I loved working with those 6812B sources. We have 6 (3 HP and 3 Agilent) of those in the HW lab and are very reliable. I have used them about 15 years before I retired this year. Thank you for the video. It is very interesting to see the inside of the unit. We use it with AC Source GUI software for running IEC/IEEE power supply compliance tests. I can create test profiles to perform power supply tests such as voltage dips and run it with ACGui software. I trigger the oscilloscope via trigger output of 6812 during test runs. When we purchased newer Keysight 6812C it didn't work with the ACgui software. I called up service center and they said that we need to use newer software since ACgui software is obsolete and no longer supported. However, the nice tech on the phone told me that I just need to change the scheme in the unit setting from 6812C to 6812B to make compatible with the ACgui software. I believe the ACgui software is free download at Keysight.
I've been thinking about making something along the same lines as this (obviously just the poor man's version), so I spent some time studying the Agilent service manual, especially the schematics. I was amazed at just how much stuff is in these things. Not just the major components, but the huge number of smaller supporting components throughout. That got me really starting to wonder what these looked like inside, and so I searched for a teardown video on UA-cam. This video was the perfect result, thanks! I've just been tearing down a couple of 3000VA transformerless UPS's, which are not dissimilar in topology. (my attempt will use parts from them) Again I was amazed at how much more stuff is in this unit, versus the UPS which outputs four times as much power. Especially in the number of electrolytic capacitors and MOSFETS, and the size of the heatsinks and cooling. Thanks again for a very informative video!
Glad the video was helpful and thanks for watching! Yes these AC power sources are complex as it takes a lot to accurately control high power levels across a wide voltage and frequency range. This 6812B can also do DC offset, or DC and no AC, plus all the measurement capabilities so that complicates things too. The Behlman AC source I have does the design very differently, massive transformers and inductor, voltage pre regulation but not at a high frequency like the 6812B, and push-pull high power amplifier to drive the output transformer.
Thanks for watching! The power source was manufactured in 2011 so the RIFA are the newer version. I inspected them but did not find the telltale hairline cracks older units develop that lead to failure.
Perfect! (I suspect that somebody like you cannot miss something like that). You are absolutely right. Excuse me. And please, keep on doing your excellent and extremely clear and clean videos. TIA!
Very interesting PSU I never heard of before. I have a 6632B with the exact same VFD display and wondered what all those unused segments are for that light up in the self test. In particular the phase displays on the left. I notice yours does a display of Φ1 (phi or theta). Now it makes sense that the VFD could also be purposed for 3 phase AC as there are 3 of them. I can also see why there are AC+DC display annunciators on the right of the display.
Yes the display has the extra phase indicators for other applications that involved a 3 phase setup. The AC and DC for the voltmeter are for reading or blocking AC, DC, or AC + DC. The Output AC and DC indicators tell you if the output allows DC offset with AC or only AC (DC nulled to zero even if offset is entered).
Nice one! One has to be really careful with "DC coupling"/DC offset too. Would be a great way to destroy a transformer. I found the CC on my 6811B to be really slow and wasn't sure if it was meant to be *that* slow, but it surely looks like it. In my light bulb tests the control loop also seemed a bit unstable depending on the use case - the slow control loop plus the negative tempco of the filament might have something to do with it... Mine also shows a certain offset. Funny thing: They recommend a 3458A for the calibration/adjustment. I guess the poor man's 3458A - the trusty 34401A - will have to suffice when I get around to the calibration. If you do the procedure on your unit, I'd be interested in that video ;)
Yes the DC offset with AC would be destructive to transformers and other devices. This instrument has to be used with caution for sure. You can shut off the AC and have only DC up to 425 volts at a few hundred watts which makes a nice DC power supply too. The user manual mentions that the CC mode may be slow (I read that after filming the video) at lower voltages, and other situations, which certainly seems to be the case. Interesting on the 8.5 digit 3458A.. that seems overkill for the accuracy of this power source. Maybe an excuse to get the 3458A :) Thanks for watching!
@@InssiAjaton yes, you are right. What I wanted to say was: Source limits current (due to newly set lower current limit), voltage drops, now the resistance decreases (because of the positive tempco), hence the controller has to reduce current further, voltage drops further, … until the source basically shuts down.
The RIFA caps in older equipment definitely have problems. I have done videos about that. They get hairline cracks and eventually fail. I inspected the caps in this unit and they are all ok. This unit was made in 2011 and RIFA caps that fail are older 80's/90's vintage.
You might want to consider replacing the RIFA capacitors on the AC section. They have a history of going up in smoke, and are usually placed in the AC circuit before the power switch so always energized when plugged in.
The RIFA caps in older equipment definitely have problems. I have done videos about that. They get hairline cracks and eventually fail. I inspected the caps in this unit and they are all ok. This unit was made in 2011 and RIFA caps that fail are older 80's/90's vintage. Yes, some RIFA caps are connected before the power switch.
Hi from Norway! From your first name, are you of Scandinavian origin? That is a very versatile, and complicated power source, must have been very expensive new. Nice!
Calibration procedure is basically taking various readings on the output such as voltage and current with a known accurate DMM and entering those values into the instrument. That creates calibration constants (or offsets) that the instrument uses to make the output and displayed values accurate.
Hi! Have you back-up the firmware in EPROMs? I have this one but unfortunately it could not power up and only can display shift if I pressed the shift button. I'd like to repair it. I suspect the program is corrupted.
Sorry I do not have a firmware backup to offer. The service manual actually lists your fault symptom of only the shift indicator turning on. It may not be the firmware. There are some test and troubleshooting procedures listed. The service manual is available free at www.keysight.com/th/en/product/6812B/performance-ac-power-source-750-va-300-v-65-a.html#resources
Thank you. I have seen the service manual. I checked the signals followed the instruction and found that there was no UART signal from the DSP. I tried to switch on and off many times and there was only one time the display segments were all lit up. More situations were either the shift could light up or the keyboard had no response.
It can generate any AC voltage up to 300 volts so think of it as an electronic version of a variac. Any frequency DC (0 Hz) to 1 KHz so you can test 50 Hz and 400 Hz equipment or check how frequency variation affects a device's performance. Measuring abilities built in so you can monitor performance of a device being tested. Isolated output from main AC line which helps add safety during testing. Clean and stable output waveform for testing sensitive devices that you want to isolate from line fluctuations.
@@kevinedwards5719 Used checked working $1500 and up. Parts (complete boards) can run into the many hundreds or another thousand if you are trying to repair a unit. Non working or "unknown condition" units can be around $500 which is the description my unit had.
I loved working with those 6812B sources. We have 6 (3 HP and 3 Agilent) of those in the HW lab and are very reliable. I have used them about 15 years before I retired this year. Thank you for the video. It is very interesting to see the inside of the unit. We use it with AC Source GUI software for running IEC/IEEE power supply compliance tests. I can create test profiles to perform power supply tests such as voltage dips and run it with ACGui software. I trigger the oscilloscope via trigger output of 6812 during test runs. When we purchased newer Keysight 6812C it didn't work with the ACgui software. I called up service center and they said that we need to use newer software since ACgui software is obsolete and no longer supported. However, the nice tech on the phone told me that I just need to change the scheme in the unit setting from 6812C to 6812B to make compatible with the ACgui software. I believe the ACgui software is free download at Keysight.
Glad you enjoyed the video and thanks for sharing your experience.
Nice AC power synthesizer. There are actually some which can generate an arbitrary waveform.
Yes very versatile for testing. Although not completely arbitrary, this one can be set to do various transient output waveforms.
I've been thinking about making something along the same lines as this (obviously just the poor man's version), so I spent some time studying the Agilent service manual, especially the schematics. I was amazed at just how much stuff is in these things. Not just the major components, but the huge number of smaller supporting components throughout.
That got me really starting to wonder what these looked like inside, and so I searched for a teardown video on UA-cam. This video was the perfect result, thanks!
I've just been tearing down a couple of 3000VA transformerless UPS's, which are not dissimilar in topology. (my attempt will use parts from them) Again I was amazed at how much more stuff is in this unit, versus the UPS which outputs four times as much power. Especially in the number of electrolytic capacitors and MOSFETS, and the size of the heatsinks and cooling.
Thanks again for a very informative video!
Glad the video was helpful and thanks for watching! Yes these AC power sources are complex as it takes a lot to accurately control high power levels across a wide voltage and frequency range. This 6812B can also do DC offset, or DC and no AC, plus all the measurement capabilities so that complicates things too. The Behlman AC source I have does the design very differently, massive transformers and inductor, voltage pre regulation but not at a high frequency like the 6812B, and push-pull high power amplifier to drive the output transformer.
I saw several Rifa caps in the very first PCB you showed. Better get rid of these ASAP...Nice video, as ever. Thank you.
Thanks for watching! The power source was manufactured in 2011 so the RIFA are the newer version. I inspected them but did not find the telltale hairline cracks older units develop that lead to failure.
Perfect! (I suspect that somebody like you cannot miss something like that). You are absolutely right. Excuse me. And please, keep on doing your excellent and extremely clear and clean videos. TIA!
@@fichambawelby2632 there will be more videos :) glad you are enjoying them!
Very thorough as usual. Thanks
Thanks for watching!
Very interesting PSU I never heard of before. I have a 6632B with the exact same VFD display and wondered what all those unused segments are for that light up in the self test. In particular the phase displays on the left. I notice yours does a display of Φ1 (phi or theta). Now it makes sense that the VFD could also be purposed for 3 phase AC as there are 3 of them. I can also see why there are AC+DC display annunciators on the right of the display.
Yes the display has the extra phase indicators for other applications that involved a 3 phase setup. The AC and DC for the voltmeter are for reading or blocking AC, DC, or AC + DC. The Output AC and DC indicators tell you if the output allows DC offset with AC or only AC (DC nulled to zero even if offset is entered).
Nice one! One has to be really careful with "DC coupling"/DC offset too. Would be a great way to destroy a transformer. I found the CC on my 6811B to be really slow and wasn't sure if it was meant to be *that* slow, but it surely looks like it. In my light bulb tests the control loop also seemed a bit unstable depending on the use case - the slow control loop plus the negative tempco of the filament might have something to do with it... Mine also shows a certain offset. Funny thing: They recommend a 3458A for the calibration/adjustment. I guess the poor man's 3458A - the trusty 34401A - will have to suffice when I get around to the calibration. If you do the procedure on your unit, I'd be interested in that video ;)
Yes the DC offset with AC would be destructive to transformers and other devices. This instrument has to be used with caution for sure. You can shut off the AC and have only DC up to 425 volts at a few hundred watts which makes a nice DC power supply too. The user manual mentions that the CC mode may be slow (I read that after filming the video) at lower voltages, and other situations, which certainly seems to be the case. Interesting on the 8.5 digit 3458A.. that seems overkill for the accuracy of this power source. Maybe an excuse to get the 3458A :)
Thanks for watching!
Actually the light bulb temperature coefficient is POSITIVE. Low ohms in cold state & high inrush current. Settles to low current when heated up.
@@InssiAjaton yes, you are right. What I wanted to say was: Source limits current (due to newly set lower current limit), voltage drops, now the resistance decreases (because of the positive tempco), hence the controller has to reduce current further, voltage drops further, … until the source basically shuts down.
Looks like there are some RIFA caps on the AC input board right behind the terminal strip and possibly next to the relay that should be changed...
The RIFA caps in older equipment definitely have problems. I have done videos about that. They get hairline cracks and eventually fail. I inspected the caps in this unit and they are all ok. This unit was made in 2011 and RIFA caps that fail are older 80's/90's vintage.
yep, timebomb. if one of these goes off it's silver confetti time.
You might want to consider replacing the RIFA capacitors on the AC section. They have a history of going up in smoke, and are usually placed in the AC circuit before the power switch so always energized when plugged in.
The RIFA caps in older equipment definitely have problems. I have done videos about that. They get hairline cracks and eventually fail. I inspected the caps in this unit and they are all ok. This unit was made in 2011 and RIFA caps that fail are older 80's/90's vintage. Yes, some RIFA caps are connected before the power switch.
Hi from Norway! From your first name, are you of Scandinavian origin? That is a very versatile, and complicated power source, must have been very expensive new. Nice!
Hi and thanks for watching the video. I have German ancestors. Yes the AC power source was quite expensive new!
👍
Very mr carlson's lab, but digital. needs some more AVE phrases though.
I was thinking the same thing. Mini Mr Carlson. Even his voice sound like a younger Mr Carlson. 🙂
will you video the cal process
Calibration procedure is basically taking various readings on the output such as voltage and current with a known accurate DMM and entering those values into the instrument. That creates calibration constants (or offsets) that the instrument uses to make the output and displayed values accurate.
Hi! Have you back-up the firmware in EPROMs? I have this one but unfortunately it could not power up and only can display shift if I pressed the shift button. I'd like to repair it. I suspect the program is corrupted.
Sorry I do not have a firmware backup to offer. The service manual actually lists your fault symptom of only the shift indicator turning on. It may not be the firmware. There are some test and troubleshooting procedures listed.
The service manual is available free at www.keysight.com/th/en/product/6812B/performance-ac-power-source-750-va-300-v-65-a.html#resources
Thank you. I have seen the service manual. I checked the signals followed the instruction and found that there was no UART signal from the DSP. I tried to switch on and off many times and there was only one time the display segments were all lit up. More situations were either the shift could light up or the keyboard had no response.
whats the max wattage output?.
750 VA which is 750 watts on purely resistive loads.
What are these things used for?
It can generate any AC voltage up to 300 volts so think of it as an electronic version of a variac. Any frequency DC (0 Hz) to 1 KHz so you can test 50 Hz and 400 Hz equipment or check how frequency variation affects a device's performance. Measuring abilities built in so you can monitor performance of a device being tested. Isolated output from main AC line which helps add safety during testing. Clean and stable output waveform for testing sensitive devices that you want to isolate from line fluctuations.
@@EriksElectronicsWorkbench how much is used one cost or to refurbish one ?
@@kevinedwards5719 Used checked working $1500 and up. Parts (complete boards) can run into the many hundreds or another thousand if you are trying to repair a unit. Non working or "unknown condition" units can be around $500 which is the description my unit had.