Author Topic: Agilent 6626A quad power supply teardown  (Read 18796 times)

0 Members and 1 Guest are viewing this topic.

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Agilent 6626A quad power supply teardown
« on: December 13, 2015, 02:41:32 am »
Product link at Keysight: http://www.keysight.com/en/pd-836789-pn-6626A/precision-system-power-supply-25w-or-50w-4-outputs?cc=US&lc=eng

Well I just got my hands on one of these and what better way to share than to open the thing up and expose it to the world:



Unlike other large HP/Agilent supplies, this one has a hinge on its right side and opens up (after removing a gazillion screws) like a book. It is a quad supply that is ridiculously accurate (we're talking 1 uA adjustment level). The output specs are:

Two outputs having: 0-7V and 0-50V at either 15 mA or 200 mA (with varying adjustment/display accuracy)
Two outputs having: 0-16V (200 mA or 2 A) and 0-50V (1A)

When you've got some high power MOSFETs:



You need some high power cooling:



They call me 'Da Fan' and you don't want to know what I did to the last set of fingers that went pokin' 'round ma' privates:



I'm a big ass transformer and I power this beast. I have more contacts than you have fingers:



Who da tranformer? No, seriously, this thing has way too many contacts (pardon the dust):



Some top notch Linear Technology and Maxim voltage references here. These guys weren't fooling around when they designed this one:



Who needs USB when you have GPIB? It looks more like an old four line modem card from this vantage point, though:



Unlike all of the bug infested crappy firmware put out these days, they got it right the first time:


« Last Edit: December 14, 2015, 02:49:32 am by SharpEars »
 

Offline Len

  • Frequent Contributor
  • **
  • Posts: 547
  • Country: ca
Re: Agilent 6626A quad power supply mini teardown ($20,755)
« Reply #1 on: December 13, 2015, 02:55:43 am »
From keysight.com:

Quote
Product will be discontinued: 01 July 2016

Get 'em while they're hot!
DIY Eurorack Synth: https://lenp.net/synth/
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8517
  • Country: us
    • SiliconValleyGarage
Re: Agilent 6626A quad power supply mini teardown ($20,755)
« Reply #2 on: December 13, 2015, 03:08:52 am »
AHH. THE HOOVER VACUUM OF POWER SUPPLIES.
i have 6624 6626 and 6627. they really are only worth like 300$ tops on ebay.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply mini teardown ($20,755)
« Reply #3 on: December 13, 2015, 03:51:05 am »
AHH. THE HOOVER VACUUM OF POWER SUPPLIES.
i have 6624 6626 and 6627. they really are only worth like 300$ tops on ebay.

Got mine ridiculously cheap on eBay. I've just had a very lucky streak with old HP/Agilent supplies, lately. However you have to be careful with these used, because if there was ever an HP/Agilent supply that you want to purchase well within factory calibration, it's this one (GPIB only calibration sucks, BTW). Just so happened that mine was perfect, calibration wise (i.e., well within spec). I might try to calibrate further, but that would be serious VoltNut level calibration.

I have to ask, why do you need a 6624 and a 6627, once you have the 6626? At a measly retail price of $12,892 each, they're junk in comparison, at least based on programming accuracy for both voltage and current.
« Last Edit: December 13, 2015, 04:47:29 am by SharpEars »
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #4 on: December 13, 2015, 05:15:54 am »
I've had a 6627A for a while but haven't used it for a long time. I recall it had a problem, it had a significant power on spike on all of the channels. Is that typical?
 

Offline 3141592

  • Regular Contributor
  • *
  • Posts: 50
  • Country: 00
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #5 on: December 13, 2015, 09:42:22 am »
That is some sexy PSU! What are these used for by the way? I'm a bit perplexed about what task needs uA level adjustments and 2A channels at the same time.
 

Offline iampoor

  • Frequent Contributor
  • **
  • Posts: 500
  • Country: us
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #6 on: December 13, 2015, 09:57:53 am »
How many logic chips does a power supply need?  >:D
 

Offline quantumvolt

  • Frequent Contributor
  • **
  • Posts: 395
  • Country: th
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #7 on: December 13, 2015, 11:33:22 am »
This device is imo a Microprocessorized Safety Belt & Air Bagged Multiple Output DAC Machine referenced by an LT1021 (see U318 in the service manual on Keysight's product web page). Good, maybe even very good - but hardly volt-nut stuff.

Hence - if the DC is what one wants, any fanboiii on the forum with an LM399 and a handful of precision resistors (and 5 minutes access to a 6.5 digit calibrated meter) can run volt donuts around this box.

If you get a kick from the high quality hardware, they are at fleebay auctions for 49.99 dolaaares now  :-DMM

http://www.ebay.com/itm/HP-Agilent-6626A-System-Quad-DC-Power-Supply-/331727510943?hash=item4d3c801d9f:g:9XgAAOSw7FRWZh7K

 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #8 on: December 13, 2015, 01:42:32 pm »
This device is imo a Microprocessorized Safety Belt & Air Bagged Multiple Output DAC Machine referenced by an LT1021 (see U318 in the service manual on Keysight's product web page). Good, maybe even very good - but hardly volt-nut stuff.

Hence - if the DC is what one wants, any fanboiii on the forum with an LM399 and a handful of precision resistors (and 5 minutes access to a 6.5 digit calibrated meter) can run volt donuts around this box.

If you get a kick from the high quality hardware, they are at fleebay auctions for 49.99 dolaaares now  :-DMM

http://www.ebay.com/itm/HP-Agilent-6626A-System-Quad-DC-Power-Supply-/331727510943?hash=item4d3c801d9f:g:9XgAAOSw7FRWZh7K

Speaking of the bay, you clearly missed the reserve not met part when you saw the $49.99 "suckers' bid."
 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #9 on: December 13, 2015, 01:45:29 pm »
I've had a 6627A for a while but haven't used it for a long time. I recall it had a problem, it had a significant power on spike on all of the channels. Is that typical?

An HP supply with a power on spike? It sounds broken to me, because HP is extremely careful in their design to not have this phenomenon commonly present in Som Tin Wong supplies from China. I wish I had a 6627A to confirm this for you with 100% certainty, but alas among the more than a dozen HP/Agilent supplies I have, the 6627A is not a member.

I strongly urge you to look into the Operating Manual and then the Service Manual at keysight and find a graph and/or text regarding power on performance. Usually, these are explicitly stated. In fact, I have an HP 4 kV supply and I can connect an LED to it, set it to 4000 volts while off and discharged, current limit it to 20 milliamps and it will turn on flawlessly and power the LED with 20 milliamps (CC) without so much as leaving a scratch in the LEDs silicon due to overshoot, so...
« Last Edit: December 13, 2015, 01:51:11 pm by SharpEars »
 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #10 on: December 13, 2015, 06:29:35 pm »
This device is imo a Microprocessorized Safety Belt & Air Bagged Multiple Output DAC Machine referenced by an LT1021 (see U318 in the service manual on Keysight's product web page). Good, maybe even very good - but hardly volt-nut stuff.

Hence - if the DC is what one wants, any fanboiii on the forum with an LM399 and a handful of precision resistors (and 5 minutes access to a 6.5 digit calibrated meter) can run volt donuts around this box.

If you get a kick from the high quality hardware, they are at fleebay auctions for 49.99 dolaaares now  :-DMM

http://www.ebay.com/itm/HP-Agilent-6626A-System-Quad-DC-Power-Supply-/331727510943?hash=item4d3c801d9f:g:9XgAAOSw7FRWZh7K

How do you get an LM399 and a handful of precision resistors to supply and regulate at voltages higher than the buried zener's zener breakdown voltage? Then there's the cost of the LM399 itself (oh wait, I mean four of them since this is a quad supply).
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #11 on: December 13, 2015, 08:32:01 pm »
I've had a 6627A for a while but haven't used it for a long time. I recall it had a problem, it had a significant power on spike on all of the channels. Is that typical?

An HP supply with a power on spike? It sounds broken to me, because HP is extremely careful in their design to not have this phenomenon commonly present in Som Tin Wong supplies from China. I wish I had a 6627A to confirm this for you with 100% certainty, but alas among the more than a dozen HP/Agilent supplies I have, the 6627A is not a member.

I strongly urge you to look into the Operating Manual and then the Service Manual at keysight and find a graph and/or text regarding power on performance. Usually, these are explicitly stated. In fact, I have an HP 4 kV supply and I can connect an LED to it, set it to 4000 volts while off and discharged, current limit it to 20 milliamps and it will turn on flawlessly and power the LED with 20 milliamps (CC) without so much as leaving a scratch in the LEDs silicon due to overshoot, so...

Yes I figured as much. Given it's a system supply it seemed unlikely to be working correctly - as disconnecting and reconnecting the devices wouldn't be practical.

I looked into the service manual and assuming I'm reading it right the overshoot should be <100mV at power on.

The thing that threw me was that all of the channels did the same thing. Since there is no shared circuitry between the channels it seemed unlikely they would all have the same fault.
 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #12 on: December 13, 2015, 10:10:41 pm »
I've had a 6627A for a while but haven't used it for a long time. I recall it had a problem, it had a significant power on spike on all of the channels. Is that typical?

An HP supply with a power on spike? It sounds broken to me, because HP is extremely careful in their design to not have this phenomenon commonly present in Som Tin Wong supplies from China. I wish I had a 6627A to confirm this for you with 100% certainty, but alas among the more than a dozen HP/Agilent supplies I have, the 6627A is not a member.

I strongly urge you to look into the Operating Manual and then the Service Manual at keysight and find a graph and/or text regarding power on performance. Usually, these are explicitly stated. In fact, I have an HP 4 kV supply and I can connect an LED to it, set it to 4000 volts while off and discharged, current limit it to 20 milliamps and it will turn on flawlessly and power the LED with 20 milliamps (CC) without so much as leaving a scratch in the LEDs silicon due to overshoot, so...

Yes I figured as much. Given it's a system supply it seemed unlikely to be working correctly - as disconnecting and reconnecting the devices wouldn't be practical.

I looked into the service manual and assuming I'm reading it right the overshoot should be <100mV at power on.

The thing that threw me was that all of the channels did the same thing. Since there is no shared circuitry between the channels it seemed unlikely they would all have the same fault.

Is this after a cold power on? What I mean is, if you leave the output turned off for say a minute and then turn the output on, do you see the overshoot?
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #13 on: December 13, 2015, 10:45:05 pm »

Is this after a cold power on? What I mean is, if you leave the output turned off for say a minute and then turn the output on, do you see the overshoot?

Yes, IIRC it happened every time it was turned on. I will set it up again tonight and double check.
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12298
  • Country: au
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #14 on: December 13, 2015, 11:20:05 pm »
This device is imo a Microprocessorized Safety Belt & Air Bagged Multiple Output DAC Machine referenced by an LT1021 (see U318 in the service manual on Keysight's product web page). Good, maybe even very good - but hardly volt-nut stuff.

Hence - if the DC is what one wants, any fanboiii on the forum with an LM399 and a handful of precision resistors (and 5 minutes access to a 6.5 digit calibrated meter) can run volt donuts around this box.

If you get a kick from the high quality hardware, they are at fleebay auctions for 49.99 dolaaares now  :-DMM

http://www.ebay.com/itm/HP-Agilent-6626A-System-Quad-DC-Power-Supply-/331727510943?hash=item4d3c801d9f:g:9XgAAOSw7FRWZh7K

Speaking of the bay, you clearly missed the reserve not met part when you saw the $49.99 "suckers' bid."

I didn't. 

I've wasted time on chasing one or two of this type of baiting auction in the past, but I don't even look twice at them now.  Yes, it may be working and the reserve may be $100 or $200 - but if it's anything half decent, the chances that they are prepared to give it away are pretty slim.
 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown - $20,755 (HW porn warning)
« Reply #15 on: December 13, 2015, 11:48:20 pm »
This device is imo a Microprocessorized Safety Belt & Air Bagged Multiple Output DAC Machine referenced by an LT1021 (see U318 in the service manual on Keysight's product web page). Good, maybe even very good - but hardly volt-nut stuff.

Hence - if the DC is what one wants, any fanboiii on the forum with an LM399 and a handful of precision resistors (and 5 minutes access to a 6.5 digit calibrated meter) can run volt donuts around this box.

If you get a kick from the high quality hardware, they are at fleebay auctions for 49.99 dolaaares now  :-DMM

http://www.ebay.com/itm/HP-Agilent-6626A-System-Quad-DC-Power-Supply-/331727510943?hash=item4d3c801d9f:g:9XgAAOSw7FRWZh7K

Speaking of the bay, you clearly missed the reserve not met part when you saw the $49.99 "suckers' bid."

I didn't. 

I've wasted time on chasing one or two of this type of baiting auction in the past, but I don't even look twice at them now.  Yes, it may be working and the reserve may be $100 or $200 - but if it's anything half decent, the chances that they are prepared to give it away are pretty slim.

I don't even bother with auctions marked as "Reserve not met." If the seller was serious and wanted to sell the items for a reasonable price, he would not be playing these sort of games. Time is money after all and every time in the past when I bid on "reserve not yet met" items, it turned out that the seller wanted an unreasonable and in fact ridiculous amount for his goods and was just trying to lure suckers in with a low and clearly unacceptable starting bid that was several times lower than the minimum acceptable amount he was after. What a waste of time! Fortunately, most buyers have caught on to the scam and stopped bidding on auctions that involve this "reserve not met" nonsense.

« Last Edit: December 13, 2015, 11:56:24 pm by SharpEars »
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #16 on: December 14, 2015, 05:04:54 pm »
Well, I got but a 6624, which I got quite cheap (150 or so for the psu, another 150 for shipping), and it works quite well for me.

Mine always switches off all outputs as it starts up. But I still connected my trusty beloved Iwatsu DS6612 to it and did a power cycle...with a definitive overshoot. In some other pics, I saw even higher spikes one even a few volts. But I didnt want to power it on and off 50 times...plus I had no load there whatsoever, and AFAIK, the minimum current the 6624 can live with is 1mA or so.

When enabling an output from disabled state, I saw no overshoot. See the two traces attached.

P.S. My other scopes all currently need maintenance done...sorry for the slight jitter in the one pic...but the 6612 is an old lady already, followed me through multiple tours of service... ;)
« Last Edit: December 14, 2015, 05:07:04 pm by TAMHAN »
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #17 on: December 15, 2015, 12:26:24 am »
Well I had a play around with my 6627A last night and I wasn't able to replicate the problem behaviour I had seen before.

I could get it to misbehave a little when powering off with an output switched on, and it looked to me like the power switch might be 'bouncing'.

So I went to the Keysight www and found the part so I tried to order it.
What a PITA that is...
I need to register and be approved by Keysight to order a $10 part.
Now they have declined my registration because my email address doesn't have the same domain name as my company.
Following _their_ instructions to request it be reconsidered got a useless response.

I've had much better response from Rigol. Keysight are supposed to be an A brand? A for Arrogant perhaps...


 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #18 on: December 15, 2015, 12:42:38 am »
Hello,
may I ask which part you are hunting? In many cases, one can find KeySight parts also from other vendors.

Plus, my experience tells, that once something is borked up, one can rarely fix it. Simply restart all over again, signing up and sheet ;)
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #19 on: December 15, 2015, 12:50:26 am »
Hello,
may I ask which part you are hunting? In many cases, one can find KeySight parts also from other vendors.

Plus, my experience tells, that once something is borked up, one can rarely fix it. Simply restart all over again, signing up and sheet ;)

Thanks. Yes you are probably right. I just expected more from Keysight after reading comments here about their customer support.
They, and other big name brands, have always appeared to me to be a bit clueless about online selling. Which is a dangerous place to be these days.

This is the part number: 3101-2862
Description: Switch-Rocker Primary-Switch DPST 15A 250V Quick Disconnect Socket
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #20 on: December 15, 2015, 09:57:16 am »
Well they were finally able to work out why they won't sell me the parts.

Keysight only sell spare parts online to US customers.
They are telling me to go through the local agent, and won't be interested as they didn't sell me the gear.

So for me there isn't much point in buying HP/Agilent/Keysight gear.
 

Offline vinicius.jlantunes

  • Regular Contributor
  • *
  • Posts: 225
  • Country: br
Re: Agilent 6626A quad power supply teardown
« Reply #21 on: December 15, 2015, 11:05:48 am »
@hendorog, maybe worth emailing the local agent... you may be surprised.

I say that as I've been through the same as you to some extent, needed a front panel replacement for a power supply I bought used on ebay.
Found the part in their website (well, didn't find it at first but asked for help via the feature in their search page, and next day received a detailed email with pictures from an engineer who had identified the part I needed) and then was informed I could not order it directly as I'm not in the US.
No problem, emailed local agent (or maybe called, can't remember) and in a matter of a week or so they had the part shipped from the US, sorted all import hassle and re-shipped it to me, all at a very reasonable cost (I would even say it was kinda cheap for the level or service I got).

So long story short, having to contact the local agent is not necessarily a bad thing - I am just a hobbyist who needed a replacement part for my used broken equipment and still received top notch support.

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown
« Reply #22 on: December 15, 2015, 02:55:33 pm »
Well they were finally able to work out why they won't sell me the parts.

Keysight only sell spare parts online to US customers.
They are telling me to go through the local agent, and won't be interested as they didn't sell me the gear.

So for me there isn't much point in buying HP/Agilent/Keysight gear.

Actually, Keysight won't even directly sell you parts if you are in the US unless you can prove you are a business (and not an individual/hobbyist). All the same e-mail domain/company name verification still applies.

Local dealers of course won't stock parts for 20-30 year old supplies and want to upcharge for a special order.
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #23 on: December 15, 2015, 08:40:51 pm »
Well they were finally able to work out why they won't sell me the parts.

Keysight only sell spare parts online to US customers.
They are telling me to go through the local agent, and won't be interested as they didn't sell me the gear.

So for me there isn't much point in buying HP/Agilent/Keysight gear.

Actually, Keysight won't even directly sell you parts if you are in the US unless you can prove you are a business (and not an individual/hobbyist). All the same e-mail domain/company name verification still applies.

Local dealers of course won't stock parts for 20-30 year old supplies and want to upcharge for a special order.


Interesting, thanks.

That just seems an odd stance for a company to take these days.
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #24 on: December 15, 2015, 08:41:40 pm »
@hendorog, maybe worth emailing the local agent... you may be surprised.

I say that as I've been through the same as you to some extent, needed a front panel replacement for a power supply I bought used on ebay.
Found the part in their website (well, didn't find it at first but asked for help via the feature in their search page, and next day received a detailed email with pictures from an engineer who had identified the part I needed) and then was informed I could not order it directly as I'm not in the US.
No problem, emailed local agent (or maybe called, can't remember) and in a matter of a week or so they had the part shipped from the US, sorted all import hassle and re-shipped it to me, all at a very reasonable cost (I would even say it was kinda cheap for the level or service I got).

So long story short, having to contact the local agent is not necessarily a bad thing - I am just a hobbyist who needed a replacement part for my used broken equipment and still received top notch support.

OK cool thanks. I'll give it a try and see how it goes.
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8517
  • Country: us
    • SiliconValleyGarage
Re: Agilent 6626A quad power supply teardown
« Reply #25 on: December 15, 2015, 09:21:15 pm »
That switch is nothing special. it is made by Preh.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #26 on: December 15, 2015, 10:01:23 pm »
free_electron to the rescue. Now it is but a question of meeting the minimum order at Farnell, RS or another distributor of choice!
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #27 on: December 16, 2015, 12:20:29 am »
That switch is nothing special. it is made by Preh.

Thanks for that info. I was hoping to get the same switch, but if it doesn't work out I'll just find something generic.
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #28 on: December 17, 2015, 10:41:43 pm »
Well the local agent has responded with two options:

1. The replacement switch for $77USD. Cost on Find a part is $9.48
2. An alternative part they have in stock for $34 USD. Cost on Find a part is $5.59

Hmm. No surprises there.
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #29 on: December 17, 2015, 10:46:59 pm »
Makes me feel like paying the findapart fee ;)
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #30 on: December 17, 2015, 10:48:15 pm »
Makes me feel like paying the findapart fee ;)

Yeah - if only they would just shut up and take my money  |O
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #31 on: December 20, 2015, 04:49:11 pm »
Hendo, I will prolly do an ASSCRATE of an order on Tuesday from Farnell. But I am not sure if all the caps I need for my scopes will bring me above 25$, and I still didnt get my trade account so  :blah: :scared: :palm: |O

If the part is available from there, I will gladly gang up with you on it. So you let me know the number, and I check if I can order it sensibly. If yes, as soon as I get it, I stick it in a slovak post first class letter and send it to wherever your HendoNess might live.
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26906
  • Country: nl
    • NCT Developments
Re: Agilent 6626A quad power supply teardown
« Reply #32 on: December 20, 2015, 05:52:28 pm »
@hendorog, maybe worth emailing the local agent... you may be surprised.

I say that as I've been through the same as you to some extent, needed a front panel replacement for a power supply I bought used on ebay.
Found the part in their website (well, didn't find it at first but asked for help via the feature in their search page, and next day received a detailed email with pictures from an engineer who had identified the part I needed) and then was informed I could not order it directly as I'm not in the US.
I have a similar experience with the support from HP/Agilent/Keysight in the NL. Unfortunately the part I needed was no longer available but still great service.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
Re: Agilent 6626A quad power supply teardown
« Reply #33 on: December 21, 2015, 08:06:08 am »
Hendo, I will prolly do an ASSCRATE of an order on Tuesday from Farnell. But I am not sure if all the caps I need for my scopes will bring me above 25$, and I still didnt get my trade account so  :blah: :scared: :palm: |O

If the part is available from there, I will gladly gang up with you on it. So you let me know the number, and I check if I can order it sensibly. If yes, as soon as I get it, I stick it in a slovak post first class letter and send it to wherever your HendoNess might live.

Hehe, great post!  :-+

I'm in the same boat and have been building up a shopping cart - and I know it will add up to enough to make the free shipping at digikey (ouch!)
We have our summer holidays now, so it will be a while...

So thanks for the offer, I appreciate it, but I'll order it a bit later when I have time to make use of it.

Cheers!
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #34 on: December 21, 2015, 09:22:55 am »
No worries, Hendo. I know that feeling just too well...swamped with work here.
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline Bud

  • Super Contributor
  • ***
  • Posts: 6911
  • Country: ca
Re: Agilent 6626A quad power supply teardown
« Reply #35 on: December 26, 2015, 04:33:36 am »
You need some high power cooling:



I have a 6622A and the fan is noisy (Papst 4600X). What is a proper method to oil fans that have Sintec bearing and what oil/grease to use ?
Facebook-free life and Rigol-free shack.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16283
  • Country: za
Re: Agilent 6626A quad power supply teardown
« Reply #36 on: December 26, 2015, 08:12:52 am »
Peel back the label, and use a needle to pop out the rubber seal if it did not come off with the label. Grab some synthetic engine oil ( Mobil one or equivalent 0W30 engine oil) or some straight SAE 15 mineral oil, and place a few drops on the bush. Leave with the bush uppermost for an hour or two, so it soaks down into the bush, then fill up level with the top of the bush, pop the rubber seal back on then stick the label back down.

If the rubber cap is still very flexible ( press down on it to see if it flexes under the label) then do not peel the label, just use a small needle on a syringe to inject 0.5ml of oil into the space under the bush.
 

Offline Bud

  • Super Contributor
  • ***
  • Posts: 6911
  • Country: ca
Re: Agilent 6626A quad power supply teardown
« Reply #37 on: December 27, 2015, 09:12:13 am »
I ended up with sewing machine oil, which is a clear low viscosity oil. The fan though did not have a rubber seal. Under the label were two screws, after removing which the fan split to two parts and the one with the motor assembly had a metal cup at the end, so I put a few oil drops there.
The power part with heat sinks of my 6622A looked different from OP's 6626A, pics attached.
Facebook-free life and Rigol-free shack.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16283
  • Country: za
Re: Agilent 6626A quad power supply teardown
« Reply #38 on: December 27, 2015, 09:19:33 am »
They don't want you changing those output transistors for sure, with that ceramic hybrid interconnect. Probably also has the ballast resistors in there as well, to save board space.
 

Offline SharpEarsTopic starter

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: Agilent 6626A quad power supply teardown
« Reply #39 on: December 31, 2015, 01:38:46 am »
All right, which one of you bastahd snipers won the auction on the HP 6626A unit on eBay? Nice bidding war on that one. :popcorn:

I sniped won the Agilent one and there's one more HP unit ending tomorrow from the same seller.
« Last Edit: December 31, 2015, 01:44:21 am by SharpEars »
 

Offline PTR_1275

  • Frequent Contributor
  • **
  • Posts: 561
  • Country: au
Re: Agilent 6626A quad power supply teardown
« Reply #40 on: November 25, 2016, 02:10:52 pm »
I know I am bringing up a dead thread, but I couldn't see much other information on the 6626a in the forum.

I have one that I purchased a few months back. I haven't had time to do much with it until recently. Channel 3 is getting abnormal over voltage errors.

The OVP is set to 55v. The power supply output voltage is pretty bang on, but if I go above about 15 volts, it shuts down with a OVP.

I have been reading through the service manual and the first thing to check is U327. On the boards I have, U327 has been replaced by a hybrid module fitted over where U327 would normally go. There are a few header pins which I am assuming are the required test points, but I cannot find any information on what pin is what.

The copy of the manual I have is from the keysight website and was printed September 2001. My unit doesn't have a serial number on it (I thought I was going crazy, but can't see a serial number or even evidence of there being one), there is a sticker on the metal case (under the cover) that says 6626A 20000342.

Does anyone have a copy of a newer service manual, or know what the amendment would be?
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Agilent 6626A quad power supply teardown
« Reply #41 on: November 27, 2016, 06:36:21 pm »
Hello,
all I know is that the unit switches its power train at 15V or so.  Maybe check the two regulator threads.

Tam
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf