Voltage Regulators in Parallel...

[John Larkin]

On Wed, 3 May 2023 15:14:37 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

 

[bitrex]

On 5/3/2023 3:28 PM, John Larkin wrote:

On Wed, 3 May 2023 15:14:37 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH


Yikes. Why not fiddle with the ground pins?

I don\'t immediately see a way to parallel only fixed regulators by
\"fiddling with the ground pins\" that (in principle, anyway) extends to
arbitrary numbers of parallel regulators...

 

[bitrex]

On 5/3/2023 3:38 PM, bitrex wrote:

On 5/3/2023 3:28 PM, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom
cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when
you
use two different kinds of power diodes and do a temp sweep like
the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have
access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max
voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep
doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the
output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH


Yikes. Why not fiddle with the ground pins?


I don\'t immediately see a way to parallel only fixed regulators by
\"fiddling with the ground pins\" that (in principle, anyway) extends to
arbitrary numbers of parallel regulators...

That can be the next challenge, lol. This one at least seems stable over
temperature and doesn\'t have a ballast resistor paradox.

 

[bitrex]

On 5/3/2023 2:28 PM, Phil Hobbs wrote:

On 2023-05-03 13:31, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like
the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.

Those spherical cows don\'t know how good they\'ve got it, for sure.

Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


Yeah, at some point you might as well build the regulator you actually
want.

One op amp, one Darlington or MOS pass transistor, a voltage reference,
and a small BJT plus a few resistors to make a foldback current limiter
to stay within the SOA and thermal limits.

Cheers

Phil Hobbs

We\'re trying to build the regulator equivalent of a lifted pick-up! Not
some elegant thing

 

[Fred Bloggs]

On Wednesday, May 3, 2023 at 1:34:28 PM UTC-4, John Larkin wrote:

On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.
The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.

And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.
Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.


Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.
That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.

Typical IQ out the GND on the 78XX is 5mA, with 8mA worst case. That particular circuit is on the verge of losing control. What\'s the cross-biasing of the diodes and C1 all about, other than to confuse the situation?

The TI datasheet lists the output tolerance as 2%, which is an old spec. The new process should be 0.2% or 25mV, they just don\'t want anyone to hold them to it. But even that applied to identical series output diodes could result in over 2:1 current imbalance, until the IC current or overtemp limits.

At least the Turkish circuit knew enough to hang the by-pass caps directly off the outputs in front of the diodes. But it\'s a crappy circuit and doesn\'t work the way they think it does. They\'re using 4007\'s??? All that stand-off voltage, missing the point their diode max current rating is being exercised by their circuit at not even max load.

 

[Fred Bloggs]

On Wednesday, May 3, 2023 at 3:28:40 PM UTC-4, John Larkin wrote:

On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail..com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

SPICE isn\'t worth a sh_t for a circuit like this. It\'s just a bunch of over-idealized garbage.

 

[John Larkin]

On Wed, 3 May 2023 15:38:26 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 3:28 PM, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH


Yikes. Why not fiddle with the ground pins?


I don\'t immediately see a way to parallel only fixed regulators by
\"fiddling with the ground pins\" that (in principle, anyway) extends to
arbitrary numbers of parallel regulators...

Use opamps (or trimpots!) to tweak the ground pin voltages to balance
regulator currents. The +12 out will track the ground pin voltage.

 

[John Larkin]

On Wed, 3 May 2023 13:44:13 -0700 (PDT), Fred Bloggs
<bloggs.fredbloggs.fred@gmail.com> wrote:

On Wednesday, May 3, 2023 at 3:28:40?PM UTC-4, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

SPICE isn\'t worth a sh_t for a circuit like this. It\'s just a bunch of over-idealized garbage.

Spice is wonderful for things like this, properly used. Well, with
decent device models.

 

[bitrex]

On 5/3/2023 5:52 PM, John Larkin wrote:

On Wed, 3 May 2023 15:38:26 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 3:28 PM, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH


Yikes. Why not fiddle with the ground pins?


I don\'t immediately see a way to parallel only fixed regulators by
\"fiddling with the ground pins\" that (in principle, anyway) extends to
arbitrary numbers of parallel regulators...

Use opamps (or trimpots!) to tweak the ground pin voltages to balance
regulator currents. The +12 out will track the ground pin voltage.

Sounds like the output voltage will have to be lifted above the
worst-case output low voltage to give some room to maneuver with that
scheme, or have a negative supply for the op amps if they\'re to balance
automatically.

If say one output is sitting at 12.05 and the other at 11.95 with both
GNDs grounded then you can\'t move the 12.05 down, the 11.95 can only go up.

 

[bitrex]

On 5/3/2023 5:53 PM, John Larkin wrote:

On Wed, 3 May 2023 13:44:13 -0700 (PDT), Fred Bloggs
bloggs.fredbloggs.fred@gmail.com> wrote:

On Wednesday, May 3, 2023 at 3:28:40?PM UTC-4, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

SPICE isn\'t worth a sh_t for a circuit like this. It\'s just a bunch of over-idealized garbage.

Spice is wonderful for things like this, properly used. Well, with
decent device models.

The 78xx models are full discrete component models, what\'s inside those
regulators is no mystery.

 

[Phil Hobbs]

On 2023-05-03 17:53, John Larkin wrote:

On Wed, 3 May 2023 13:44:13 -0700 (PDT), Fred Bloggs
bloggs.fredbloggs.fred@gmail.com> wrote:

On Wednesday, May 3, 2023 at 3:28:40?PM UTC-4, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

SPICE isn\'t worth a sh_t for a circuit like this. It\'s just a bunch of over-idealized garbage.

Spice is wonderful for things like this, properly used. Well, with
decent device models.

That is, in a purely fictitious universe.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[John Larkin]

On Wed, 3 May 2023 21:09:04 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 5:52 PM, John Larkin wrote:
On Wed, 3 May 2023 15:38:26 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 3:28 PM, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <user@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <user@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <cd@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH


Yikes. Why not fiddle with the ground pins?


I don\'t immediately see a way to parallel only fixed regulators by
\"fiddling with the ground pins\" that (in principle, anyway) extends to
arbitrary numbers of parallel regulators...

Use opamps (or trimpots!) to tweak the ground pin voltages to balance
regulator currents. The +12 out will track the ground pin voltage.


Sounds like the output voltage will have to be lifted above the
worst-case output low voltage to give some room to maneuver with that
scheme, or have a negative supply for the op amps if they\'re to balance
automatically.

If say one output is sitting at 12.05 and the other at 11.95 with both
GNDs grounded then you can\'t move the 12.05 down, the 11.95 can only go up.

See which reg is the highest voltage and move the other one up to
match.

 

[John Larkin]

On Wed, 3 May 2023 22:03:30 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2023-05-03 17:53, John Larkin wrote:
On Wed, 3 May 2023 13:44:13 -0700 (PDT), Fred Bloggs
bloggs.fredbloggs.fred@gmail.com> wrote:

On Wednesday, May 3, 2023 at 3:28:40?PM UTC-4, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

SPICE isn\'t worth a sh_t for a circuit like this. It\'s just a bunch of over-idealized garbage.

Spice is wonderful for things like this, properly used. Well, with
decent device models.

That is, in a purely fictitious universe.

Cheers

Phil Hobbs

Oh don\'t get technical.

A lot of models have truly grounded nodes inside, so you can\'t float
them off ground. And some have independent current souces, so can be
used as perpetual motion machines.

 

[Flyguy]

On Wednesday, May 3, 2023 at 7:46:19 PM UTC-7, John Larkin wrote:

On Wed, 3 May 2023 22:03:30 -0400, Phil Hobbs
pcdhSpamM...@electrooptical.net> wrote:

On 2023-05-03 17:53, John Larkin wrote:
On Wed, 3 May 2023 13:44:13 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:

On Wednesday, May 3, 2023 at 3:28:40?PM UTC-4, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:

On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:

On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com
wrote:

Gentlemen,

I have a bunch of 12VDC and 5VDC voltage regulator ICs in TO-220
packages. IIRC, I think they\'re about 1A each. Could they be
paralleled-up to get more current or would they have a tendency to
\'fight each other\' as it were? It would be nice if they further
*stablised* each other but I\'m guessing I wouldn\'t be that lucky.

Thanks,

CD

This is a little better maybe:

https://www.dropbox.com/s/0rf7274rlesc0mh/LM317s_parallel_2.jpg?raw=1



While we\'re being \"pleasingly goofy\":

https://imgur.com/a/ZK8eitu

Load regulation isn\'t great, but at leas it doesn\'t go crazy when you
use two different kinds of power diodes and do a temp sweep like the EDN
one does..

OK, that wins the goofy prize. Why use diodes to merge the two reg
outputs?

Paralleling fixed regulators is annoying, since you don\'t have access to
the feedback terminal, and can\'t connect it to the \"real\" output after
the ballast resistors. So the only thing that can bring a regulator\'s
output into line is current against the ballast resistors. But if you
use ballast resistors large enough to ensure tight current sharing you
end up with a large voltage drop at max output.

I think the diodes were a hack basically to try to limit the max voltage
drop at max output while ensuring there\'s always _some_ amount of
current sharing. vs if you just connect them directly together they\'re
guaranteed to hog.

The diodes probably make the sharing worse than resistors. Diode
behavior is complex, NTC at low currents and PTC at high currents.


And in the sim they do share very nicely, at 25 degrees C, with ideal
7812s and ideal diodes, but just start swapping diode types around and
it goes sideways over temperature. My goofy idea isn\'t really much
better. The LM7812 is +/- 5% over temperature and the temp sweep doesn\'t
take that into account, you\'d have to monte carlo that.

Sims share nicely with no sharing parts at all! Just slap the regs in
parallel.



Looks like someone was trying to maximize the current imbalance.


Any solution for fixed regulators has to take into account the output of
one regulator could be sitting at 12.5 and the other at 11.5 and be
within spec, and none of the simple ideas I\'ve played with can do much
with that. You could actively monitor the input current to each and do
something with that maybe, but that seems like a waste of time.

That\'s one reason to use LM317s or 1117s. They will share to
millivolts.

Or add opamps.


You can parallel 78xxs with a few op amps and transistors, but you have
to play the leap frog game:

https://imgur.com/a/3z7qwXH

Yikes. Why not fiddle with the ground pins?

SPICE isn\'t worth a sh_t for a circuit like this. It\'s just a bunch of over-idealized garbage.

Spice is wonderful for things like this, properly used. Well, with
decent device models.

That is, in a purely fictitious universe.

Cheers

Phil Hobbs
Oh don\'t get technical.

A lot of models have truly grounded nodes inside, so you can\'t float
them off ground. And some have independent current souces, so can be
used as perpetual motion machines.

The bottom line is to select a voltage regulator that can handle the current - paralleling several inadequate regulators is a crap solution not worthy of analysis.

 

[Anthony William Sloman]

On Thursday, May 4, 2023 at 1:06:32 PM UTC+10, Flyguy wrote:

On Wednesday, May 3, 2023 at 7:46:19 PM UTC-7, John Larkin wrote:
On Wed, 3 May 2023 22:03:30 -0400, Phil Hobbs <pcdhSpamM...@electrooptical.net> wrote:
On 2023-05-03 17:53, John Larkin wrote:
On Wed, 3 May 2023 13:44:13 -0700 (PDT), Fred Bloggs <bloggs.fred....@gmail.com> wrote:
On Wednesday, May 3, 2023 at 3:28:40?PM UTC-4, John Larkin wrote:
On Wed, 3 May 2023 15:14:37 -0400, bitrex <us...@example.net> wrote:
On 5/3/2023 1:31 PM, John Larkin wrote:
On Wed, 3 May 2023 12:07:44 -0400, bitrex <us...@example.net> wrote:
On 5/2/2023 7:14 PM, John Larkin wrote:
On Tue, 2 May 2023 11:31:07 -0400, bitrex <us...@example.net> wrote:
On 5/2/2023 10:19 AM, John Larkin wrote:
On Tue, 02 May 2023 00:20:37 +0100, Cursitor Doom <c...@notformail.com

<snip>

Spice is wonderful for things like this, properly used. Well, with decent device models.

That is, in a purely fictitious universe.

Not entirely. Some very simple integrated circuits are modelled at the transistor level. The Gummel-Poon transistor model doesn\'t capture inverted operation all that accurately, but mostly it is good enough.

> > Oh don\'t get technical.

If not here, where?

> > A lot of models have truly grounded nodes inside, so you can\'t float them off ground. And some have independent current sources, so can be used as perpetual motion machines.

You do have to use that kind of model with discretion.

> The bottom line is to select a voltage regulator that can handle the current - paralleling several inadequate regulators is a crap solution not worthy of analysis.

Flyguy has said something sensible for once. He\'s wrong of course - paralleling several very cheap voltage regulators can give you a cheap and nasty - but workable - solution and that would be worth analysing. There\'s usually an even cheaper and more elegant solution, but putting a good engineer on that kind of job is wasting their talent.

--
Bill Sloman, Sydney

 

[whit3rd]

On Wednesday, May 3, 2023 at 8:06:32 PM UTC-7, Flyguy wrote:

> The bottom line is to select a voltage regulator that can handle the current - paralleling several inadequate regulators is a crap solution not worthy of analysis.

Analysis is somewhat useful; that\'s why we can call it a crap solution... Many systems having
multiple point-of-load regulators show us an easy alternative, simply dividing the load into two
or more blocks, and using a different regulator to power each block. As long as the input power
is a single unregulated source, regulator outputs will track well-enough-for-logic.

And, in low-noise systems where correlation of multiple signals is important, having more than
one regulator can help decouple power supply noise contributions.

 

[chuck]

On 04/05/2023 2:25 p.m., whit3rd wrote:

On Wednesday, May 3, 2023 at 8:06:32 PM UTC-7, Flyguy wrote:

The bottom line is to select a voltage regulator that can handle the current - paralleling several inadequate regulators is a crap solution not worthy of analysis.

Analysis is somewhat useful; that\'s why we can call it a crap solution... Many systems having
multiple point-of-load regulators show us an easy alternative, simply dividing the load into two
or more blocks, and using a different regulator to power each block. As long as the input power
is a single unregulated source, regulator outputs will track well-enough-for-logic.

And, in low-noise systems where correlation of multiple signals is important, having more than
one regulator can help decouple power supply noise contributions.

Distributed 5V regulators was the design of choice used by HP with a raw
source of 6V (?) and a shared 5V reference.

It was used for each PCB slot in a HP Multi-Programmer 19\" Rack.

It was in my first custom design of a SCADA System with ADC/DAC/96
channel MUX based on two remote HP9825 calculators via serial RS485
datasets circa 1977. This also include Resistor controlled Voltage
Lambda power Supplies using R-DAC\'s all controlled a mile away from the
rocket launch pad.

 

[John Larkin]

On Thu, 4 May 2023 16:37:07 -0400, chuck <donnyduck@gmail.com> wrote:

On 04/05/2023 2:25 p.m., whit3rd wrote:
On Wednesday, May 3, 2023 at 8:06:32?PM UTC-7, Flyguy wrote:

The bottom line is to select a voltage regulator that can handle the current - paralleling several inadequate regulators is a crap solution not worthy of analysis.

Analysis is somewhat useful; that\'s why we can call it a crap solution... Many systems having
multiple point-of-load regulators show us an easy alternative, simply dividing the load into two
or more blocks, and using a different regulator to power each block. As long as the input power
is a single unregulated source, regulator outputs will track well-enough-for-logic.

And, in low-noise systems where correlation of multiple signals is important, having more than
one regulator can help decouple power supply noise contributions.

Distributed 5V regulators was the design of choice used by HP with a raw
source of 6V (?) and a shared 5V reference.

It was used for each PCB slot in a HP Multi-Programmer 19\" Rack.

It was in my first custom design of a SCADA System with ADC/DAC/96
channel MUX based on two remote HP9825 calculators via serial RS485
datasets circa 1977. This also include Resistor controlled Voltage
Lambda power Supplies using R-DAC\'s all controlled a mile away from the
rocket launch pad.

I used to design SCADA systems, mostly for product pipelines. More
like hundreds of miles over leased phone lines and sometimes RF out to
oil rigs. Ever hear of TANO Corp?

 

[Wandere]

Just switch them on one at a time. For example. Two regulators; one\'s on, one\'s off. Each sees a 50% duty-cycle. Expand as needed. Overlapping the on cycles might improve the switching noise.

 

[Cursitor Doom]

On Sun, 07 May 2023 11:21:28, Wanderer<dont@emailme.com> wrote:

Just switch them on one at a time. For example. Two regulators; one\'s on, one\'s off. Each sees a 50% duty-cycle. Expand as needed. Overlapping the on cycles might improve the switching noise

Novel idea.