negative impedance power supply...

[server]

On Mon, 17 Jul 2023 07:29:44 -0700 (PDT), Fred Bloggs
<bloggs.fredbloggs.fred@gmail.com> wrote:

Of course the same people who run into this have no clue beforehand that a 300 ft run might be problematic.

If you are going to feed e.g. 5 V any significant load at such
distances, you are going to have problems. Why not install a DC/DC
converter close to the load and fed the long line with e.g. 24 V ?

 

[Fred Bloggs]

On Tuesday, July 18, 2023 at 11:36:32 AM UTC-4, upsid...@downunder.com wrote:

On Mon, 17 Jul 2023 07:29:44 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:


Of course the same people who run into this have no clue beforehand that a 300 ft run might be problematic.
If you are going to feed e.g. 5 V any significant load at such
distances, you are going to have problems. Why not install a DC/DC
converter close to the load and fed the long line with e.g. 24 V ?

I know it\'s pathetic, but not at unusual.

 

[John Larkin]

On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
<bloggs.fredbloggs.fred@gmail.com> wrote:

On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability

It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.

They almost certainly use a low-resistance shunt to sense output
current. Voltage drops in the 10s of millivolts are typical, keeping
dissipation down.

There are some nice isolated delta-sigma ADCs around lately, which
allows the shunt to be anywhere in the circuit without common-mode
hassles. Like ADUM7703. One can play cute tricks, trading off
bandwidth against resolution in an FPGA.

 

[John Larkin]

On Tue, 18 Jul 2023 14:04:05 +0100, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

On 16/07/2023 21:24, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

I haven\'t, but it\'s one technique for powering things at the end of very
long, lossy lines, think oilwell. Only one conductor plus armour, so
remote sensing would need pretty complex telemetry.

As the current drawn increases, so does the supply voltage in an attempt
to keep the downhole voltage within some range.

There were transatlantic telephone cables with *tube* repeaters every
so often. They use the most carefully manufactured and tested tubes in
history.

The power supplies were 10s of kilovolts on each end and the amps were
in series from a power supply standpoint. Probably had the equivalent
of a shunt regulator. That was all described in an old BSTJ.

 

[Anthony William Sloman]

On Wednesday, July 19, 2023 at 1:59:11 AM UTC+10, John Larkin wrote:

On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series.. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability
It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.

Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop.

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation.

They almost certainly use a low-resistance shunt to sense output current.. Voltage drops in the 10s of millivolts are typical, keeping
dissipation down.

There are some nice isolated delta-sigma ADCs around lately, which
allows the shunt to be anywhere in the circuit without common-mode
hassles. Like ADUM7703. One can play cute tricks, trading off
bandwidth against resolution in an FPGA.

It helps if you know the basic of frequency compensation negative feedback loops. People who claim top do electronic design should understand that.

--
Bill Sloman, Sydney

 

[Lasse Langwadt Christensen]

tirsdag den 18. juli 2023 kl. 18.08.01 UTC+2 skrev Anthony William Sloman:

On Wednesday, July 19, 2023 at 1:59:11 AM UTC+10, John Larkin wrote:
On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability
It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.
Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop.

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation.

still have to be fast enough to keep the voltage in check on a load dump

 

[Fred Bloggs]

On Tuesday, July 18, 2023 at 11:59:11 AM UTC-4, John Larkin wrote:

On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series.. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability
It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.

They almost certainly use a low-resistance shunt to sense output
current. Voltage drops in the 10s of millivolts are typical, keeping
dissipation down.

There are some nice isolated delta-sigma ADCs around lately, which
allows the shunt to be anywhere in the circuit without common-mode
hassles. Like ADUM7703. One can play cute tricks, trading off
bandwidth against resolution in an FPGA.

A few 10s of milliseconds response time should be good enough. Also even a 6-bit should be enough for a power supply voltage, but 8-bits are standard so use one of them.

If you\'re willing to go the current divider route, then GMR and Hall effect should be feasible for non-contact, low loss current measurement. And they\'re low pass elements, even better.

 

[John Larkin]

On Tue, 18 Jul 2023 09:28:13 -0700 (PDT), Lasse Langwadt Christensen
<langwadt@fonz.dk> wrote:

tirsdag den 18. juli 2023 kl. 18.08.01 UTC+2 skrev Anthony William Sloman:
On Wednesday, July 19, 2023 at 1:59:11?AM UTC+10, John Larkin wrote:
On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability
It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.
Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop.

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation.

still have to be fast enough to keep the voltage in check on a load dump

If the load opens and the measured current drops, the power supply
will react by dropping its output voltage back to the no-load setpoint
value.

I wouldn\'t expect that my power supply would ever try to take out much
wire drop, maybe a volt or so.

In real life, power supplies tend to have wild and weird transient
responses.

 

[Clive Arthur]

On 18/07/2023 17:04, John Larkin wrote:

On Tue, 18 Jul 2023 14:04:05 +0100, Clive Arthur
clive@nowaytoday.co.uk> wrote:

<snip>

I haven\'t, but it\'s one technique for powering things at the end of very
long, lossy lines, think oilwell. Only one conductor plus armour, so
remote sensing would need pretty complex telemetry.

As the current drawn increases, so does the supply voltage in an attempt
to keep the downhole voltage within some range.

There were transatlantic telephone cables with *tube* repeaters every
so often. They use the most carefully manufactured and tested tubes in
history.

The power supplies were 10s of kilovolts on each end and the amps were
in series from a power supply standpoint. Probably had the equivalent
of a shunt regulator. That was all described in an old BSTJ.

This might be typical...

http://www.camesaemc.com/En/Monoconductor/1K22-Monoconductor#1K22PP

....no outer insulation, poor outer conductivity (and skin effect!), no
chance of any repeaters, it\'s on a drum and might be 10km. Plus it\'s
very expensive and you\'d rather not lose it.

Small x-section area is important as it has to go through a pressure
gland and you don\'t want to make a \'piston\' which might push the cable up.

--
Cheers
Clive

 

[Anthony William Sloman]

On Wednesday, July 19, 2023 at 2:28:18 AM UTC+10, Lasse Langwadt Christensen wrote:

tirsdag den 18. juli 2023 kl. 18.08.01 UTC+2 skrev Anthony William Sloman:
On Wednesday, July 19, 2023 at 1:59:11 AM UTC+10, John Larkin wrote:
On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability

It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.

Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop.

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation.

still have to be fast enough to keep the voltage in check on a load dump

That\'s desirable, but if there\'s too much lag around the loop, you can\'t rely on feedback to get it. Brute force - like a big enough Zener - can fill in.

--
Bill Sloman, Sydney

 

[server]

On Tue, 18 Jul 2023 18:43:46 +0100, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

What kind of power levels do you want to transfer ?

This might be typical...

http://www.camesaemc.com/En/Monoconductor/1K22-Monoconductor#1K22PP

...no outer insulation, poor outer conductivity (and skin effect!), no
chance of any repeaters, it\'s on a drum and might be 10km. Plus it\'s
very expensive and you\'d rather not lose it.

If this is a DC feed, why would be some skin effect issues ?

According to the link, the maximum cable voltage is 1200 V and
resistance 13.5 ohms/km thus the 10 km drum would have 135 Ohms. If
the ground return resistance is similar, the full loop resistance is
about 250 Ohms.With a 1000 V feed and 2 A current, only 500 V would
reach the load i.e. 1000 W. This is the maximum power since equal
power is lost is the wire resistances as in the load.

If the loop current is only 0.1 A, only 25 V (2.5%) is lost and the
load will get 975 V or 97 W. Not much (or any) regulation needed even
if a shorter drum is used.

 

[Jan Panteltje]

On a sunny day (Wed, 19 Jul 2023 09:27:04 +0300) it happened
upsidedown@downunder.com wrote in
<2quebit65m5kc8g4uc779cs9v314mb2dbo@4ax.com>:

On Tue, 18 Jul 2023 18:43:46 +0100, Clive Arthur
clive@nowaytoday.co.uk> wrote:

What kind of power levels do you want to transfer ?

This might be typical...

http://www.camesaemc.com/En/Monoconductor/1K22-Monoconductor#1K22PP

...no outer insulation, poor outer conductivity (and skin effect!), no
chance of any repeaters, it\'s on a drum and might be 10km. Plus it\'s
very expensive and you\'d rather not lose it.

If this is a DC feed, why would be some skin effect issues ?

According to the link, the maximum cable voltage is 1200 V and
resistance 13.5 ohms/km thus the 10 km drum would have 135 Ohms. If
the ground return resistance is similar, the full loop resistance is
about 250 Ohms.With a 1000 V feed and 2 A current, only 500 V would
reach the load i.e. 1000 W. This is the maximum power since equal
power is lost is the wire resistances as in the load.

Yes I have used the high voltage low current method to power my drone via a thin light coax:
https://panteltje.nl/pub/h501s_drone_remote_power_test_ground_control_1_IMG_6276.JPG
note the thin brown coax going over teh fence.

Simple transformer and rectifier at the drone end.
https://panteltje.nl/pub/h501s_drone_remote_power_drone_side_IMG_6278.JPG
the ring core transformer and diodes are cooled here by the propeller airflow ...

If the loop current is only 0.1 A, only 25 V (2.5%) is lost and the
load will get 975 V or 97 W. Not much (or any) regulation needed even
if a shorter drum is used.

In my drone case with about 100 kHz, yes, DC would have no skin effect.

 

[Clive Arthur]

On 19/07/2023 07:27, upsidedown@downunder.com wrote:

On Tue, 18 Jul 2023 18:43:46 +0100, Clive Arthur
clive@nowaytoday.co.uk> wrote:

What kind of power levels do you want to transfer ?

Not me, but kilowatts for motors isn\'t unusual.

This might be typical...

http://www.camesaemc.com/En/Monoconductor/1K22-Monoconductor#1K22PP

...no outer insulation, poor outer conductivity (and skin effect!), no
chance of any repeaters, it\'s on a drum and might be 10km. Plus it\'s
very expensive and you\'d rather not lose it.

If this is a DC feed, why would be some skin effect issues ?

I was referring to Fred\'s suggestion that remote voltage sensing could
be effected via telemetry rather than use a negative impedance supply.
It can, but skin effect makes telemetry harder, it\'s a steel outer, so
pretty bad. The signal levels are small, the attenuation is high, and
the motors may be big and noisy, so the game is often not worth the candle.

<snip>

--
Cheers
Clive

 

[Lasse Langwadt Christensen]

onsdag den 19. juli 2023 kl. 06.43.27 UTC+2 skrev Anthony William Sloman:

On Wednesday, July 19, 2023 at 2:28:18 AM UTC+10, Lasse Langwadt Christensen wrote:
tirsdag den 18. juli 2023 kl. 18.08.01 UTC+2 skrev Anthony William Sloman:
On Wednesday, July 19, 2023 at 1:59:11 AM UTC+10, John Larkin wrote:
On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting..

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability

It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.

Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop.

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation..

still have to be fast enough to keep the voltage in check on a load dump
That\'s desirable, but if there\'s too much lag around the loop, you can\'t rely on feedback to get it. Brute force - like a big enough Zener - can fill in.

it would have to be at the far end of the cable, pretty accurate, and in the case of an adjustable supply, adjustable

 

[Anthony William Sloman]

On Wednesday, July 19, 2023 at 8:19:28 PM UTC+10, Lasse Langwadt Christensen wrote:

onsdag den 19. juli 2023 kl. 06.43.27 UTC+2 skrev Anthony William Sloman:
On Wednesday, July 19, 2023 at 2:28:18 AM UTC+10, Lasse Langwadt Christensen wrote:
tirsdag den 18. juli 2023 kl. 18.08.01 UTC+2 skrev Anthony William Sloman:
On Wednesday, July 19, 2023 at 1:59:11 AM UTC+10, John Larkin wrote:
On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability

It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.

Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop.

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation.

still have to be fast enough to keep the voltage in check on a load dump
That\'s desirable, but if there\'s too much lag around the loop, you can\'t rely on feedback to get it. Brute force - like a big enough Zener - can fill in.

it would have to be at the far end of the cable, pretty accurate, and in the case of an adjustable supply, adjustable.

The accuracy required is determined by the breakdoown voltage of the parts being driven. There\'s no absolute requirement for precision.

If it is at the far end of the cable, the clamp voltage is obviously set by whatever is being driven, and since it would be installed separately it is obviously adjustable (in the sense that you buy what you need before you fitted it).

--
Bill Sloman, Sydney

 

[John Larkin]

On Wed, 19 Jul 2023 10:53:35 +0100, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

On 19/07/2023 07:27, upsidedown@downunder.com wrote:
On Tue, 18 Jul 2023 18:43:46 +0100, Clive Arthur
clive@nowaytoday.co.uk> wrote:

What kind of power levels do you want to transfer ?

Not me, but kilowatts for motors isn\'t unusual.

This might be typical...

http://www.camesaemc.com/En/Monoconductor/1K22-Monoconductor#1K22PP

...no outer insulation, poor outer conductivity (and skin effect!), no
chance of any repeaters, it\'s on a drum and might be 10km. Plus it\'s
very expensive and you\'d rather not lose it.

If this is a DC feed, why would be some skin effect issues ?

I was referring to Fred\'s suggestion that remote voltage sensing could
be effected via telemetry rather than use a negative impedance supply.
It can, but skin effect makes telemetry harder, it\'s a steel outer, so
pretty bad. The signal levels are small, the attenuation is high, and
the motors may be big and noisy, so the game is often not worth the candle.

snip

I want to design some multi-channel low-cost programmable power
supplies, and remote sense would be a nuisance and burn connector
pins. RS has hazards too, more hassle to adjust for.

Programmable +- output impedance is easy, just a little more FPGA
code. I\'ll know the output voltage and current.

I guess I should simulate a bunch of cases, especially complex remote
loads with input capacitors and switching (negative input impedance)
power supplies.

 

[Fred Bloggs]

On Tuesday, July 18, 2023 at 12:39:52 PM UTC-4, John Larkin wrote:

On Tue, 18 Jul 2023 09:28:13 -0700 (PDT), Lasse Langwadt Christensen
lang...@fonz.dk> wrote:

tirsdag den 18. juli 2023 kl. 18.08.01 UTC+2 skrev Anthony William Sloman:
On Wednesday, July 19, 2023 at 1:59:11?AM UTC+10, John Larkin wrote:
On Tue, 18 Jul 2023 05:17:57 -0700 (PDT), Fred Bloggs
bloggs.fred...@gmail.com> wrote:
On Sunday, July 16, 2023 at 4:24:43?PM UTC-4, John Larkin wrote:
Has anyone designed a power supply with a negative output impedance?
That could be a viable (ie sellable) low-cost alternative to remote
sensing.

There could be a user-settable parameter, output impedance, that could
be signed. Positive values are \"droop\" and negatives are wire
compensation. Users could take out a few ohms of wire drop without
getting extreme and unstable. The dynamics could get interesting.

I have seen some buck switchers that had native negative output
impedance but never explored why.

Write-up linked below describes product line from XP Power PLS600 series. See section on \'addressing remote sensing and calibration\' . Seems much more practical with better performance than a negative impedance approach. Set-up couldn\'t be simpler, and is a good idea. I would be interested in knowing how they sense current without taking too much of hit to efficiency. Maybe efficiency is not a consideration for them. Whatever- they most likely use the more systematic approach previously described.

https://www.digikey.com/en/articles/apply-benchtop-power-supplies-programmability
It sure looks like their shorting/compensation trick is indeed
negative impedance. They must measure the wire resistance and then
cancel most of it, but not enough to make the system oscillate.
Don\'t be silly. All they need to do is keep the compensation slow enough to prevent any lag through the wiring creating an unstable feedback loop..

They can cancel all of the resistance, but they have to roll of the gain of the compensation loop at a low enough frequency to avoid oscillation.

still have to be fast enough to keep the voltage in check on a load dump

If the load opens and the measured current drops, the power supply
will react by dropping its output voltage back to the no-load setpoint
value.

The automotive load dump is not what the term would imply. It refers to catastrophic loss of the battery connection, the system therefore loses it\'s main high power voltage clamp, the alternator continues to supply the same current as before the dump. This obviously runs the voltage up to I x R (loads), I now includes battery charging current prior to disconnect, which is pretty much indeterminate. All the automotive grade electronics manufacturers design the IC to withstand a peak of 60V tapering down to zero in 0.5 seconds, or something like that.

In the case of a general electronic load, the input decoupling required to maintain a minimum voltage margin during a full load step applies identically to an overvoltage margin when the full load is removed. Your supply, if properly designed is regulating VL. If the loading suddenly becomes a capacitor, current I takes on an exponential discharge in time, the power supply should also have a similar exponential discharge of its terminal output voltage. The exponential rise in VL acts to further reduce the output voltage by opposing the cable current. So you end up with a reaction time dependent on both your supply feedback dynamics and load energy storage.

I wouldn\'t expect that my power supply would ever try to take out much
wire drop, maybe a volt or so.

In real life, power supplies tend to have wild and weird transient
responses.

 

[server]

On Wed, 19 Jul 2023 10:53:35 +0100, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

On 19/07/2023 07:27, upsidedown@downunder.com wrote:
On Tue, 18 Jul 2023 18:43:46 +0100, Clive Arthur
clive@nowaytoday.co.uk> wrote:

What kind of power levels do you want to transfer ?

Not me, but kilowatts for motors isn\'t unusual.

Whatever you do, you are not going to transfer much over 1 kW (,ax
power transfer theorem, with the same wiring resistance as the load
resistance over the discussed cable.

To increase the power delivered, you have three options:

* use a shorter cable
* use thicker conductors with less resistance
* use a cable that allows higher voltages

Select one or more options to deliver multiple kilowatts.