Protecting a power supply

[# 42]

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a diode
inline with the positive terminal. The purpose is to prevent reverse current
flow the instant the supply is turned on. The power supply is a Mastech
HY3020e switching supply (30volts, 20amps) and if it sees even a small
voltage when turned on, it does bad things to itself. Radio Shack has 50v,
6amp diodes. Can I put 4 of these in parallel with each other to get what I
need? I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

 

[John Fields]

On Mon, 2 Mar 2009 12:58:44 -0800, "# 42" <paulinlapine@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a diode
inline with the positive terminal. The purpose is to prevent reverse current
flow the instant the supply is turned on. The power supply is a Mastech
HY3020e switching supply (30volts, 20amps) and if it sees even a small
voltage when turned on, it does bad things to itself. Radio Shack has 50v,
6amp diodes. Can I put 4 of these in parallel with each other to get what I
need?

---
No.
---

I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

It's a bad idea because diodes aren't generally matched so their forward
voltages track when you buy them randomly, so they'll all have slightly
different Vf's and they'll all have a Vf with a negative temperature
coefficient.

That means that the diode with the lowest Vf will initially carry more
current than its neighbors, and the hotter it gets the lower its forward
drop will become, which will allow more charge to flow through it than
through its neighbors, which will make it get even hotter, which will
allow even more charge to flow through it than through it its
neighbors...

Eventually it could/will get hot enough to let the magic smoke out, and
then, if it fails shorted, its neighbors will no longer be diodeing or,
if it fails open, the diode with the next lowest Vf will start taking
its disproportionate share of the load and will also eventually let the
magic smoke out. Then #3 and then #4.

IMO, your best bet would be to do it with a single rectifier which can
handle 20A continuous and block what you'd expect the reverse voltage to
be with the supply off and parts in the tank. A couple of volts, maybe?

JF

 

[ian field]

"# 42" <paulinlapine@msn.com> wrote in message
news:cMXql.5186$rp7.4203@en-nntp-02.dc1.easynews.com...

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a
diode inline with the positive terminal. The purpose is to prevent reverse
current flow the instant the supply is turned on. The power supply is a
Mastech HY3020e switching supply (30volts, 20amps) and if it sees even a
small voltage when turned on, it does bad things to itself. Radio Shack
has 50v, 6amp diodes. Can I put 4 of these in parallel with each other to
get what I need? I'm just not sure if you can parallel diodes to increase
the current rating. Thanks in advance.

Generally its not best practice to put diodes in parallel as there will be
small differences in forward volt drop between diodes. The one with the
lowest drop will hog more than its share of the current and heat up more -
as temperature rises the drop falls even further causing the hottest diode
to hog an even bigger share of the current.

If you give a big enough safety margin in current rating, you can get away
with stud mounted diodes mounted close together on the same heatsink.

It seems very likely that a single stud mounted diode rated to take the
current would work out cheaper and less hassle.

 

[# 42]

Thank you, guys. Your answers probably kept the smoke in, because that's not
how I would have thought they would behave. (Electronically illiterate here)
I'll look for a single 20+ amp diode to do the job.


"# 42" <paulinlapine@msn.com> wrote in message
news:cMXql.5186$rp7.4203@en-nntp-02.dc1.easynews.com...

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a
diode inline with the positive terminal. The purpose is to prevent reverse
current flow the instant the supply is turned on. The power supply is a
Mastech HY3020e switching supply (30volts, 20amps) and if it sees even a
small voltage when turned on, it does bad things to itself. Radio Shack
has 50v, 6amp diodes. Can I put 4 of these in parallel with each other to
get what I need? I'm just not sure if you can parallel diodes to increase
the current rating. Thanks in advance.

 

[Tim Wescott]

On Mon, 02 Mar 2009 15:58:40 -0600, John Fields wrote:

On Mon, 2 Mar 2009 12:58:44 -0800, "# 42" <paulinlapine@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum.
The supplier says to protect it when using it in that fashion by putting
a diode inline with the positive terminal. The purpose is to prevent
reverse current flow the instant the supply is turned on. The power
supply is a Mastech HY3020e switching supply (30volts, 20amps) and if it
sees even a small voltage when turned on, it does bad things to itself.
Radio Shack has 50v, 6amp diodes. Can I put 4 of these in parallel with
each other to get what I need?

---
No.
---

I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

It's a bad idea because diodes aren't generally matched so their forward
voltages track when you buy them randomly, so they'll all have slightly
different Vf's and they'll all have a Vf with a negative temperature
coefficient.

That means that the diode with the lowest Vf will initially carry more
current than its neighbors, and the hotter it gets the lower its forward
drop will become, which will allow more charge to flow through it than
through its neighbors, which will make it get even hotter, which will
allow even more charge to flow through it than through it its
neighbors...

Eventually it could/will get hot enough to let the magic smoke out, and
then, if it fails shorted, its neighbors will no longer be diodeing or,
if it fails open, the diode with the next lowest Vf will start taking
its disproportionate share of the load and will also eventually let the
magic smoke out. Then #3 and then #4.

IMO, your best bet would be to do it with a single rectifier which can
handle 20A continuous and block what you'd expect the reverse voltage to
be with the supply off and parts in the tank. A couple of volts, maybe?

JF

_However_, if you're in a bind you can ballast the diodes with low-value
resistors -- something that gives a few tenths of a volt at the diode's
rated current should do. Be sure they've got a high enough power rating.

That'll keep the diode to diode current balance constant, at the cost of
a few more watts puffing up into the air.

--
http://www.wescottdesign.com

 

[John Fields]

On Mon, 02 Mar 2009 19:54:27 -0600, Tim Wescott <tim@seemywebsite.com>
wrote:

On Mon, 02 Mar 2009 15:58:40 -0600, John Fields wrote:

On Mon, 2 Mar 2009 12:58:44 -0800, "# 42" <paulinlapine@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum.
The supplier says to protect it when using it in that fashion by putting
a diode inline with the positive terminal. The purpose is to prevent
reverse current flow the instant the supply is turned on. The power
supply is a Mastech HY3020e switching supply (30volts, 20amps) and if it
sees even a small voltage when turned on, it does bad things to itself.
Radio Shack has 50v, 6amp diodes. Can I put 4 of these in parallel with
each other to get what I need?

---
No.
---

I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

It's a bad idea because diodes aren't generally matched so their forward
voltages track when you buy them randomly, so they'll all have slightly
different Vf's and they'll all have a Vf with a negative temperature
coefficient.

That means that the diode with the lowest Vf will initially carry more
current than its neighbors, and the hotter it gets the lower its forward
drop will become, which will allow more charge to flow through it than
through its neighbors, which will make it get even hotter, which will
allow even more charge to flow through it than through it its
neighbors...

Eventually it could/will get hot enough to let the magic smoke out, and
then, if it fails shorted, its neighbors will no longer be diodeing or,
if it fails open, the diode with the next lowest Vf will start taking
its disproportionate share of the load and will also eventually let the
magic smoke out. Then #3 and then #4.

IMO, your best bet would be to do it with a single rectifier which can
handle 20A continuous and block what you'd expect the reverse voltage to
be with the supply off and parts in the tank. A couple of volts, maybe?

JF

_However_, if you're in a bind you can ballast the diodes with low-value
resistors -- something that gives a few tenths of a volt at the diode's
rated current should do. Be sure they've got a high enough power rating.

That'll keep the diode to diode current balance constant, at the cost of
a few more watts puffing up into the air.

---

At six amps per diode:

--+--[DIODE>]--+--
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+

|<---0.3V--->|


That comes out to 0.0125 ohms across the diodes which takes away their
diodeness.

JF

 

[Peter Bennett]

On Tue, 03 Mar 2009 08:59:27 -0600, John Fields
<jfields@austininstruments.com> wrote:

On Mon, 02 Mar 2009 19:54:27 -0600, Tim Wescott <tim@seemywebsite.com
wrote:


_However_, if you're in a bind you can ballast the diodes with low-value
resistors -- something that gives a few tenths of a volt at the diode's
rated current should do. Be sure they've got a high enough power rating.

That'll keep the diode to diode current balance constant, at the cost of
a few more watts puffing up into the air.

---

At six amps per diode:

--+--[DIODE>]--+--
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+

|<---0.3V--->|


That comes out to 0.0125 ohms across the diodes which takes away their
diodeness.

JF

I suspect that Tim meant to place a resistor in series with each
diode, not in parallel.


--
Peter Bennett, VE7CEI
peterbb4 (at) interchange.ubc.ca
GPS and NMEA info: http://vancouver-webpages.com/peter
Vancouver Power Squadron: http://vancouver.powersquadron.ca

 

[John Fields]

On Tue, 03 Mar 2009 09:11:23 -0800, Peter Bennett
<peterbb@somewhere.invalid> wrote:

On Tue, 03 Mar 2009 08:59:27 -0600, John Fields
jfields@austininstruments.com> wrote:

On Mon, 02 Mar 2009 19:54:27 -0600, Tim Wescott <tim@seemywebsite.com
wrote:


_However_, if you're in a bind you can ballast the diodes with low-value
resistors -- something that gives a few tenths of a volt at the diode's
rated current should do. Be sure they've got a high enough power rating.

That'll keep the diode to diode current balance constant, at the cost of
a few more watts puffing up into the air.

---

At six amps per diode:

--+--[DIODE>]--+--
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+

|<---0.3V--->|


That comes out to 0.0125 ohms across the diodes which takes away their
diodeness.

JF

I suspect that Tim meant to place a resistor in series with each
diode, not in parallel.

---
Aaarghhh!!!

I'm sure you're right.

What _was_ I thinking?

JF

 

[ian field]

"John Fields" <jfields@austininstruments.com> wrote in message
news:4frqq414akh8lfjmb6tdp16u36mt71il9c@4ax.com...

On Tue, 03 Mar 2009 09:11:23 -0800, Peter Bennett
peterbb@somewhere.invalid> wrote:

On Tue, 03 Mar 2009 08:59:27 -0600, John Fields
jfields@austininstruments.com> wrote:

On Mon, 02 Mar 2009 19:54:27 -0600, Tim Wescott <tim@seemywebsite.com
wrote:


_However_, if you're in a bind you can ballast the diodes with low-value
resistors -- something that gives a few tenths of a volt at the diode's
rated current should do. Be sure they've got a high enough power
rating.

That'll keep the diode to diode current balance constant, at the cost of
a few more watts puffing up into the air.

---

At six amps per diode:

--+--[DIODE>]--+--
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+

|<---0.3V--->|


That comes out to 0.0125 ohms across the diodes which takes away their
diodeness.

JF

I suspect that Tim meant to place a resistor in series with each
diode, not in parallel.

---
Aaarghhh!!!

I'm sure you're right.

What _was_ I thinking?

JF

Or could that be - what were you smoking?!

 

[Tim Wescott]

On Tue, 03 Mar 2009 09:11:23 -0800, Peter Bennett wrote:

On Tue, 03 Mar 2009 08:59:27 -0600, John Fields
jfields@austininstruments.com> wrote:

On Mon, 02 Mar 2009 19:54:27 -0600, Tim Wescott <tim@seemywebsite.com
wrote:


_However_, if you're in a bind you can ballast the diodes with
low-value resistors -- something that gives a few tenths of a volt at
the diode's rated current should do. Be sure they've got a high enough
power rating.

That'll keep the diode to diode current balance constant, at the cost
of a few more watts puffing up into the air.

---

At six amps per diode:

--+--[DIODE>]--+--
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+

|<---0.3V--->|


That comes out to 0.0125 ohms across the diodes which takes away their
diodeness.

JF

I suspect that Tim meant to place a resistor in series with each diode,
not in parallel.

I'm pretty sure that's what he was thinking, too:

o---o---/\/\/---->|---o----o
| |
o---/\/\/---->|---o
| |
o --- etc --- o

But what do I know?

--
http://www.wescottdesign.com

 

On Mon, 2 Mar 2009 12:58:44 -0800, "# 42" <paulinlapine@msn.com>
wrote:

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a diode
inline with the positive terminal. The purpose is to prevent reverse current
flow the instant the supply is turned on. The power supply is a Mastech
HY3020e switching supply (30volts, 20amps) and if it sees even a small
voltage when turned on, it does bad things to itself. Radio Shack has 50v,
6amp diodes. Can I put 4 of these in parallel with each other to get what I
need? I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

All Electronics has some 40 amp stud mounts for $3.50 each:
http://www.allelectronics.com/make-a-store/item/1N1190A/40-AMP-600-VOLT-RECTIFIER/-/1.html
http://www.allelectronics.com/make-a-store/item/1N1190AR/40-AMP-600-VOLT-RECTIFIER/-/1.html

Shipping is a flat $7.

John

 

[Pors]

"John Fields" <jfields@austininstruments.com> wrote in message
news:uqgqq4ddh1loonampmpocndra50dtmbidl@4ax.com...

On Mon, 02 Mar 2009 19:54:27 -0600, Tim Wescott <tim@seemywebsite.com
wrote:

On Mon, 02 Mar 2009 15:58:40 -0600, John Fields wrote:

On Mon, 2 Mar 2009 12:58:44 -0800, "# 42" <paulinlapine@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum.
The supplier says to protect it when using it in that fashion by putting
a diode inline with the positive terminal. The purpose is to prevent
reverse current flow the instant the supply is turned on. The power
supply is a Mastech HY3020e switching supply (30volts, 20amps) and if it
sees even a small voltage when turned on, it does bad things to itself.
Radio Shack has 50v, 6amp diodes. Can I put 4 of these in parallel with
each other to get what I need?

---
No.
---

I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

It's a bad idea because diodes aren't generally matched so their forward
voltages track when you buy them randomly, so they'll all have slightly
different Vf's and they'll all have a Vf with a negative temperature
coefficient.

That means that the diode with the lowest Vf will initially carry more
current than its neighbors, and the hotter it gets the lower its forward
drop will become, which will allow more charge to flow through it than
through its neighbors, which will make it get even hotter, which will
allow even more charge to flow through it than through it its
neighbors...

Eventually it could/will get hot enough to let the magic smoke out, and
then, if it fails shorted, its neighbors will no longer be diodeing or,
if it fails open, the diode with the next lowest Vf will start taking
its disproportionate share of the load and will also eventually let the
magic smoke out. Then #3 and then #4.

IMO, your best bet would be to do it with a single rectifier which can
handle 20A continuous and block what you'd expect the reverse voltage to
be with the supply off and parts in the tank. A couple of volts, maybe?

JF

_However_, if you're in a bind you can ballast the diodes with low-value
resistors -- something that gives a few tenths of a volt at the diode's
rated current should do. Be sure they've got a high enough power rating.

That'll keep the diode to diode current balance constant, at the cost of
a few more watts puffing up into the air.

---

At six amps per diode:

--+--[DIODE>]--+--
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
| |
+--[DIODE>]--+
| |
+--[0.05R]---+
2 x & amp Resistors & Horry Don't knows
|<---0.3V--->|


That comes out to 0.0125 ohms across the diodes which takes away their
diodeness.

JF

 

[NASH]

On Mar 3, 1:58 am, "# 42" <paulinlap...@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a diode
inline with the positive terminal. The purpose is to prevent reverse current
flow the instant the supply is turned on. The power supply is a Mastech
HY3020e switching supply (30volts, 20amps) and if it sees even a small
voltage when turned on, it does bad things to itself. Radio Shack has 50v,
6amp diodes. Can I put 4 of these in parallel with each other to get what I
need? I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

NO Because their forward resistance will not be the same and the diode
of lower resistance become more hot & burns therefore use single diode.

 

[Jasen Betts]

On 2009-03-02, # 42 <paulinlapine@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a diode
inline with the positive terminal. The purpose is to prevent reverse current
flow the instant the supply is turned on. The power supply is a Mastech
HY3020e switching supply (30volts, 20amps) and if it sees even a small
voltage when turned on, it does bad things to itself. Radio Shack has 50v,
6amp diodes. Can I put 4 of these in parallel with each other to get what I
need? I'm just not sure if you can parallel diodes to increase the current
rating. Thanks in advance.

If RS is the only electronics supplier you can find nearby
Go to an auto-electrician and get a rectifier diode from them,
they'll have diodes rated at 50V 50A and higher for use in
automotive rectifiers.

 

[ian field]

"Jasen Betts" <jasen@xnet.co.nz> wrote in message
news:golneh$nq1$1@reversiblemaps.ath.cx...

On 2009-03-02, # 42 <paulinlapine@msn.com> wrote:

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a
diode
inline with the positive terminal. The purpose is to prevent reverse
current
flow the instant the supply is turned on. The power supply is a Mastech
HY3020e switching supply (30volts, 20amps) and if it sees even a small
voltage when turned on, it does bad things to itself. Radio Shack has
50v,
6amp diodes. Can I put 4 of these in parallel with each other to get what
I
need? I'm just not sure if you can parallel diodes to increase the
current
rating. Thanks in advance.

If RS is the only electronics supplier you can find nearby
Go to an auto-electrician and get a rectifier diode from them,
they'll have diodes rated at 50V 50A and higher for use in
automotive rectifiers.

Come to that, suitable rectifiers can be had for free out of a scrap car
alternator.

 

[ian field]

"JeffM" <jeffm_@email.com> wrote in message
news:c93fd0c3-6523-4563-8015-62945ac20ccf@y33g2000prg.googlegroups.com...

Jasen Betts wrote:
If RS is the only electronics supplier you can find nearby
Go to an auto-electrician and get a rectifier diode from them,
they'll have diodes rated at 50V 50A and higher for use in
automotive rectifiers.

ian field wrote:
Come to that, suitable rectifiers can be had for free
out of a scrap car alternator.

Aren't those suckers all press-fit?
Getting them out of the unit's housing is a bit of a chore. Right?

Not if you leave them in the heatsink and use the hole for a terminal lug
that's already there.

Why turn an easy job into an ordeal?!!!

 

[JeffM]

Jasen Betts wrote:
If RS is the only electronics supplier you can find nearby
Go to an auto-electrician and get a rectifier diode from them,
they'll have diodes rated at 50V 50A and higher for use in
automotive rectifiers.

ian field wrote:
Come to that, suitable rectifiers can be had for free
out of a scrap car alternator.

Aren't those suckers all press-fit?
Getting them out of the unit's housing is a bit of a chore. Right?
(...and half are polarized one way; half the other.)

Connecting to them is another issue:
high thermal mass and no actual second terminal.

 

[Varactor]

On Mar 3, 11:29 am, "ian field" <gangprobing.al...@ntlworld.com>
wrote:

"# 42" <paulinlap...@msn.com> wrote in message

news:cMXql.5186$rp7.4203@en-nntp-02.dc1.easynews.com...

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a
diode inline with the positive terminal. The purpose is to prevent reverse
current flow the instant the supply is turned on. The power supply is a
Mastech HY3020e switching supply (30volts, 20amps) and if it sees even a
small voltage when turned on, it does bad things to itself. Radio Shack
has 50v, 6amp diodes. Can I put 4 of these in parallel with each other to
get what I need? I'm just not sure if you can parallel diodes to increase
the current rating. Thanks in advance.

Generally its not best practice to put diodes in parallel as there will be
small differences in forward volt drop between diodes. The one with the
lowest drop will hog more than its share of the current and heat up more -
as temperature rises the drop falls even further causing the hottest diode
to hog an even bigger share of the current.

If you give a big enough safety margin in current rating, you can get away
with stud mounted diodes mounted close together on the same heatsink.

It seems very likely that a single stud mounted diode rated to take the
current would work out cheaper and less hassle.

Off the top of my head, how about a grunty MOSFET on the PS output to
delay the PS connecting to the load? The mosfet turn on would be
delayed with a RC charging circuit...

Cheers

 

[Jasen Betts]

On 2009-03-04, JeffM <jeffm_@email.com> wrote:

Jasen Betts wrote:
If RS is the only electronics supplier you can find nearby
Go to an auto-electrician and get a rectifier diode from them,
they'll have diodes rated at 50V 50A and higher for use in
automotive rectifiers.

ian field wrote:
Come to that, suitable rectifiers can be had for free
out of a scrap car alternator.

Aren't those suckers all press-fit?

depends, some are soldered to the heatsink
eg: http://www.diotec-usa.com/srp7500.pdf

Getting them out of the unit's housing is a bit of a chore. Right?
(...and half are polarized one way; half the other.)

Connecting to them is another issue:
high thermal mass and no actual second terminal.

if from a scrap altenator leave them in situ or cut the heatsink to a
suitable size leaving a bolt hole or stud for connecting the other
terminal.

(remeber to mount the heatsink in a non-conductive way,)

 

[# 42]

"# 42" <paulinlapine@msn.com> wrote in message
news:cMXql.5186$rp7.4203@en-nntp-02.dc1.easynews.com...

I've got a new power supply that I will be using to anodize aluminum. The
supplier says to protect it when using it in that fashion by putting a
diode inline with the positive terminal. The purpose is to prevent reverse
current flow the instant the supply is turned on. The power supply is a
Mastech HY3020e switching supply (30volts, 20amps) and if it sees even a
small voltage when turned on, it does bad things to itself. Radio Shack
has 50v, 6amp diodes. Can I put 4 of these in parallel with each other to
get what I need? I'm just not sure if you can parallel diodes to increase
the current rating. Thanks in advance.

Thanks for all the suggestions, I think I'll go this route.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=360131630146

Paul