Re: HP 6012B volt zero and current regulation
Hi Dave. I got to thinking a bit about the point Alan made below, and
it's a very good one. If you consider the turn-on behavior of an
unregulated linear supply (i.e. the classic transformer, bridge
rectifier, filter capacitor construct), the shape of the voltage ramp-up
curve should be dominated by the capacitor's charging characteristics.
But if you place a regulator after that, the output voltage (of the
regulator) will rise, and go potentially quite high above the setpoint
until the regulation loop takes over. The width of the waveform above
the setpoint voltage will correspond in inverse proportion to the
bandwidth of the control loop of the regulator.
Linear and switching regulators will both exhibit this behavior,
though modern switching regulators with high switching frequencies and
high-bandwidth control loops will keep it well under control.
In practice this is usually so brief that it doesn't bother anything,
a few dozen microseconds at most. I myself have watched it on an
oscilloscope (I design and characterize a lot of voltage regulators) but
have never seen it do any harm.
-Dave
toggle quoted message
Show quoted text
On 08/20/2018 05:06 PM, David Speck wrote: OK, I'll bite--
What kind of power supply would you recommend?? I always thought that a
regulated supply would be the safest.?
Were these linear or switcher supplies that produced the spikes?
Thanks,
Dave
On 8/20/2018 2:31 PM, Alan Melia via Groups.Io wrote:
During 10 years or so doing life tests on transistors in the 1960s and
70s I would not use regulated power supplies for powering the devices
during their life test. When asked, I demonstrated the switch on pulse
with a Tek storage scope. ok it was only microseconds wide but peaked
at the unregulated level.~35v .....enough to damage 24v
transistors.This can also happen with a transcient load. It is to do
with the bandwidth of the feedback loop.
?
Alan
G3NYK
--
Dave McGuire, AK4HZ
New Kensington, PA
|
Re: HP 6012B volt zero and current regulation
OK, I'll bite--
What kind of power supply would you recommend?? I always thought
that a regulated supply would be the safest.?
Were these linear or switcher supplies that produced the spikes?
Thanks,
Dave
On 8/20/2018 2:31 PM, Alan Melia via
Groups.Io wrote:
toggle quoted message
Show quoted text
During 10 years or so doing life
tests on transistors in the 1960s and 70s I would not use
regulated power supplies for powering the devices during their
life test. When asked, I demonstrated the switch on pulse with
a Tek storage scope. ok it was only microseconds wide but
peaked at the unregulated level.~35v .....enough to damage 24v
transistors.This can also happen with a transcient load. It is
to do with the bandwidth of the feedback loop.
?
Alan
G3NYK
|
Joel Kist, was Re: [HP-Agilent-Keysight-equipment] HP 6012B volt zero and current regulation
On 08/20/2018 02:48 PM, Mark Wendt wrote: Off subject ¨C I still get duplicates ¨C Joel?
By chance are you subscribed to the group with two different email
addresses? Check the headers of the emails that are dupes the next time
you receive them.? Yes, he is. There have been no duplicate messages here. That's a Yahoo problem specifically. Joel, you are subscribed at both your sbcglobal.net address and your att.net address. If you let me know which one you want to keep, I can chop the other one for you. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
toggle quoted message
Show quoted text
----- Original Message -----
Sent: Monday, August 20, 2018 8:49
PM
Subject: Re:
[HP-Agilent-Keysight-equipment] A huge thank you!
Awesome move... thanks to all involved in it! Bravo!
Phil, N6OMM?
Ditto from me Dave!!!
Jim
|
|
|
Re: HP 6012B volt zero and current regulation
I did a lot of power supply testing for HP, Agilent, and NGC, including the 6012A/B¡¯s, Here is the spec sheet:?. ?What you want to look at is the transient response - right hand column. ?Calibration transient response is tested for both voltage and current modes and is tested full to no resistive load, and also with 100/90% resistive load change. ?2ms/100mv. ?I believe that a min. 5 min. warmup time is required for specs to apply.
|
Also thank you for the move. I did not realize it until I got a redirect and log in request.
I also didn¡¯t realize I had not been getting the summaries except rarely.
Nicely done. I now have the website referenced and logged.
Thank you,
Don Bitters (ret. HP/Agilent/NGC engineer)
|
Re: 8566B SA all PLLs unlocked
I too have seen a failed power resistor. ?Other failed item maybe be a leaky diode that then shuts down the power supply. ?If the -40v supply is other than 0, -40v suspect a leaky diode, or shorted or leaky cap on the affected supply. ?
You will also get a similar symptom if you have an ext 10 MHz ref and it drops out, or you lose the internal 10MHz, or the 100MHz loop is down.
Don Bitters
|
Re: HP 6012B volt zero and current regulation
Off subject ¨C I still get duplicates ¨C Joel?
By chance are you subscribed to the group with two different email addresses? Check the headers of the emails that are dupes the next time you receive them.?
Mark
|
Re: HP 6012B volt zero and current regulation
?
During 10 years or so doing life tests on
transistors in the 1960s and 70s I would not use regulated power supplies for
powering the devices during their life test. When asked, I demonstrated the
switch on pulse with a Tek storage scope. ok it was only microseconds wide but
peaked at the unregulated level.~35v .....enough to damage 24v transistors.This
can also happen with a transcient load. It is to do with the bandwidth of the
feedback loop.
?
Alan
G3NYK
?
----- Original Message -----
Sent: Monday, August 20, 2018 7:18
PM
Subject: Re:
[HP-Agilent-Keysight-equipment] HP 6012B volt zero and current
regulation
Hi
In my experience most bench power
supplies show a similar behaviour. ?At one time I suffered repeated
damage to equipment that was powered from a bench supply, when checked with a
storage scope it went to a high voltage on turn on before settling back to the
set voltage.
I still think that for valid
measurements you cannot use a light bulb, they are NOT a linear devices if you
do an applied voltage plot against observed resistance, from a sixtey year old
memory of an experiment they are square law devices
73 George G6HIG
??
On Monday, August 20, 2018 5:50 PM,
"wilson2115@..." <wilson2115@...>
wrote:
I refined my test...
Instead of ramping the voltage with
the knob. I basically set the voltage to 20V and the amperage to 2 amps. I
connected the voltmeter in series with the bulb and performed a dynamic load
test by opening and closing the circuit by touching the probe lead of the
voltmeter to the output of the power supply... I also have the Fluke voltmeter
set to peak amperage measurement and these are the results that I found.
With the 6012B set to 20V and 2A, upon closing the
circuit the peak amps are 4 amps but quickly settles to 2 amps
With the 6024A set to 20V and 2 A, upon closing the circuit the
peaks amps are about 2.5 but quickly settles to 2 amps.
The important observation here is the 6012B settles back to 2 amps,
it seems the settle time is longer and the deviation is larger than the 6024A.
Although these are two completely different supplies. If anything I could be
nit picking here on the 6012B for its regulation settle time and overshoot,
which is why I wanted to see if someone else had a similar supply and could
see if they could duplicate the same results to see if this is normal...
Thanks for the help
|
Re: HP 6012B volt zero and current regulation
Also many bench supplies have very large output capacitors. I have a smaller HP on my bench and recorded over 100 amps instantaneous on a short with a couple feet of #10 leads. That was all from the output cap, it¡¯s a 10 amp supply.?
toggle quoted message
Show quoted text
On Aug 20, 2018, at 2:18 PM, george edmonds via Groups.Io < G6HIG@...> wrote: Hi
In my experience most bench power supplies show a similar behaviour. ?At one time I suffered repeated damage to equipment that was powered from a bench supply, when checked with a storage scope it went to a high voltage on turn on before settling back to the set voltage.
I still think that for valid measurements you cannot use a light bulb, they are NOT a linear devices if you do an applied voltage plot against observed resistance, from a sixtey year old memory of an experiment they are square law devices
73 George G6HIG ??
I refined my test...
Instead of ramping the voltage with the knob. I basically set the voltage to 20V and the amperage to 2 amps. I connected the voltmeter in series with the bulb and performed a dynamic load test by opening and closing the circuit by touching the probe lead of the voltmeter to the output of the power supply... I also have the Fluke voltmeter set to peak amperage measurement and these are the results that I found.
With the 6012B set to 20V and 2A, upon closing the circuit the peak amps are 4 amps but quickly settles to 2 amps
With the 6024A set to 20V and 2 A, upon closing the circuit the peaks amps are about 2.5 but quickly settles to 2 amps.
The important observation here is the 6012B settles back to 2 amps, it seems the settle time is longer and the deviation is larger than the 6024A. Although these are two completely different supplies. If anything I could be nit picking here on the 6012B for its regulation settle time and overshoot, which is why I wanted to see if someone else had a similar supply and could see if they could duplicate the same results to see if this is normal...
Thanks for the help
|
Re: HP 6012B volt zero and current regulation
Off subject ¨C I still get duplicates ¨C Joel STS SS ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of wilson2115@...
Sent: Monday, August 20, 2018 12:50 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 6012B volt zero and current regulation? I refined my test...
Instead of ramping the voltage with the knob. I basically set the voltage to 20V and the amperage to 2 amps. I connected the voltmeter in series with the bulb and performed a dynamic load test by opening and closing the circuit by touching the probe lead of the voltmeter to the output of the power supply... I also have the Fluke voltmeter set to peak amperage measurement and these are the results that I found.
With the 6012B set to 20V and 2A, upon closing the circuit the peak amps are 4 amps but quickly settles to 2 amps
With the 6024A set to 20V and 2 A, upon closing the circuit the peaks amps are about 2.5 but quickly settles to 2 amps.
The important observation here is the 6012B settles back to 2 amps, it seems the settle time is longer and the deviation is larger than the 6024A. Although these are two completely different supplies If anything I could be nit picking here on the 6012B for its regulation settle time and overshoot, which is why I wanted to see if someone else had a similar supply and could see if they could duplicate the same results to see if this is normal...
Thanks for the help
|
Re: HP 6012B volt zero and current regulation
Hi
In my experience most bench power supplies show a similar behaviour. ?At one time I suffered repeated damage to equipment that was powered from a bench supply, when checked with a storage scope it went to a high voltage on turn on before settling back to the set voltage.
I still think that for valid measurements you cannot use a light bulb, they are NOT a linear devices if you do an applied voltage plot against observed resistance, from a sixtey year old memory of an experiment they are square law devices
73 George G6HIG ??
On Monday, August 20, 2018 5:50 PM, "wilson2115@..." <wilson2115@...> wrote:
I refined my test...
Instead of ramping the voltage with the knob. I basically set the voltage to 20V and the amperage to 2 amps. I connected the voltmeter in series with the bulb and performed a dynamic load test by opening and closing the circuit by touching the probe lead of the voltmeter to the output of the power supply... I also have the Fluke voltmeter set to peak amperage measurement and these are the results that I found.
With the 6012B set to 20V and 2A, upon closing the circuit the peak amps are 4 amps but quickly settles to 2 amps
With the 6024A set to 20V and 2 A, upon closing the circuit the peaks amps are about 2.5 but quickly settles to 2 amps.
The important observation here is the 6012B settles back to 2 amps, it seems the settle time is longer and the deviation is larger than the 6024A. Although these are two completely different supplies. If anything I could be nit picking here on the 6012B for its regulation settle time and overshoot, which is why I wanted to see if someone else had a similar supply and could see if they could duplicate the same results to see if this is normal...
Thanks for the help
|
Re: HP 6012B volt zero and current regulation
I refined my test...
Instead of ramping the voltage with the knob. I basically set the voltage to 20V and the amperage to 2 amps. I connected the voltmeter in series with the bulb and performed a dynamic load test by opening and closing the circuit by touching the probe lead of the voltmeter to the output of the power supply... I also have the Fluke voltmeter set to peak amperage measurement and these are the results that I found.
With the 6012B set to 20V and 2A, upon closing the circuit the peak amps are 4 amps but quickly settles to 2 amps
With the 6024A set to 20V and 2 A, upon closing the circuit the peaks amps are about 2.5 but quickly settles to 2 amps.
The important observation here is the 6012B settles back to 2 amps, it seems the settle time is longer and the deviation is larger than the 6024A. Although these are two completely different supplies. If anything I could be nit picking here on the 6012B for its regulation settle time and overshoot, which is why I wanted to see if someone else had a similar supply and could see if they could duplicate the same results to see if this is normal...
Thanks for the help
|
Sincerely,
James L. Hudson
WA5JAT
|
Re: 8566B SA all PLLs unlocked
Hello Calvin,
did you check the 6 power resistors? It did happen to me that a broken one on A1A6 once killed an opamp in the IF/display unit.
best regards,
Peter
|
many thanks to those that moved the group, I hope our new home proves to be a good one.
all the best,
walter (walter2 -at- sphere.bc.ca
sphere research corp.
|
Thanks for the smooth move!
On Mon, Aug 20, 2018 at 6:32 AM Keith Monahan via Groups.Io <keith= [email protected]> wrote: Thanks indeed!
I guess there was no problem at all transferring messages?
I scraped the group awhile ago when I thought yahoo might disappear. No
jails, no throttling, no VPN required. It just worked. I think it was a
little slow, maybe overnight for 88K messages.
Glad to see you had no troubles!
Thanks again!
Keith
On 8/20/2018 9:08 AM, James Allen wrote:
> Ditto from me Dave!!!
>
> Jim
>
|
Re: migration test message
On Mon, 20 Aug 2018, f1chf wrote: Loud and clear :) hell just to check !
thanks --- * * KSI@home KOI8 Net < > The impossible we do immediately. * * Las Vegas NV, USA < > Miracles require 24-hour notice. * *
|
very nice and smooth :-)
congratulations!
|
Dave McGuire
Phillip Potter
Sergey Kubushyn