Author Topic: Need help with repair of Agilent E3647A power supply  (Read 973 times)

0 Members and 1 Guest are viewing this topic.

Offline Arek_R

  • Regular Contributor
  • *
  • Posts: 63
  • Country: gb
  • Polish guy living in the UK
Need help with repair of Agilent E3647A power supply
« on: April 14, 2018, 04:40:01 pm »
Hi, got this PSU from ebay, this is my first attempt of repairing industrial grade equipent.
In short it displays same low value on the display all the time and changing output voltage doesn't make any difference, but you can get max 13V on output 1 with 60V range and full 62V on CH2.
Cleary ADC portion on both channels doesnt work and also DAC portion on CH1.
First I want to repair ADC portion of the circuit, so I focused on part of the schematic responsible for it on smaller "daughter board" with main processors.
Started by measuring around multiplexer U27 MC74HC4051D(handles V/A measurements voltages for both channels), connected negative lead of my multimeter to positive terminal of CH1(connected to "A1" ground on schematic), set both channels to supply actual 5V, and started measuring, but then I have problem with understanding what's happening with voltages between U28 and U24 opamps, here's part of the schematic and what I measure:
https://drive.google.com/file/d/1IYsnK9iRd133jwK9XqMq3pcEMRtA-Vkm/view
And my question is: How is it possible that I measure some voltage directly on pin #6 of U28 but nothing directly on pin #2 of U24???
« Last Edit: April 14, 2018, 05:19:42 pm by Arek_R »
 

Offline orbanp

  • Regular Contributor
  • *
  • Posts: 160
Re: Need help with repair of Agilent E3647A power supply
« Reply #1 on: April 14, 2018, 06:09:48 pm »
Hi Arek,

Your measurements and the voltages are fine!

The gain of the second op-amp is roughly (pretty close to this) R342 / R343 = 8.25k / 10k  = 0.825
If you multiply 0.127V * 0.825 = 0.1048V which is close what you measure there!
That is about 5.5% error, which sounds reasonable at such low voltages.

Peter
« Last Edit: April 14, 2018, 06:16:31 pm by orbanp »
 
The following users thanked this post: Arek_R

Offline Arek_R

  • Regular Contributor
  • *
  • Posts: 63
  • Country: gb
  • Polish guy living in the UK
Re: Need help with repair of Agilent E3647A power supply
« Reply #2 on: April 14, 2018, 06:18:17 pm »
Thanks for reply!
Yeah these opamps+resistors seems to be working fine, but why I measure 0V at negative input of U24?
Also at "COMP" I have constant value of 5-6V(don't remember exactly) and it's not changing when I change output voltage by turning the encoder, it shouldn't be like that right?
 

Offline orbanp

  • Regular Contributor
  • *
  • Posts: 160
Re: Need help with repair of Agilent E3647A power supply
« Reply #3 on: April 14, 2018, 06:29:56 pm »
Quote from: Arek_R on Today at 04:18:17 AM
Thanks for reply!
Yeah these opamps+resistors seems to be working fine, but why I measure 0V at negative input of U24?
...
The - input is a virtual ground! The + input is tied ground. An op-amp amplifies the difference between + and - inputs with its gain.
In other words, with the output voltage between the supply voltages, both inputs are (nearly) at the same potential.
Divide that 0.1V output by the gain of 90dB - 100dB!

I hope that makes sense to you. If not, look up op-amps in a text book, it will probably explain it better than me.

I have not looked at your other question.

Peter
 

Offline Arek_R

  • Regular Contributor
  • *
  • Posts: 63
  • Country: gb
  • Polish guy living in the UK
Re: Need help with repair of Agilent E3647A power supply
« Reply #4 on: April 14, 2018, 07:05:49 pm »
Oh so it was just my misunderstanding of opamps, sorry....

Ok so now I measure 5.848V on "COMP" output of this circuit, and it doesn't change when I change output of the PSU, it shouldn't be like that, right?
 

Offline orbanp

  • Regular Contributor
  • *
  • Posts: 160
Re: Need help with repair of Agilent E3647A power supply
« Reply #5 on: April 14, 2018, 09:34:10 pm »
Hi Arek,

I take you have the manual for this PS!
There is one on the Agilent site, that does not have the schematics, on an other site the schematics was included as well. I guess you have that.

This is a complex (and lovely) PS!
The maintenance section describes self tests as well.
Did your PS pass all those tests?

The manual is 220 pages long!
So it would take some time to go through that and understand the operation, the basic architecture, and the circuitry of the PS.
Also, as far as I can see, there is not much there describing the principles and circuitry of the PS.

Let me know if I am missing something (e.g. another manual).

Peter
 

Offline Arek_R

  • Regular Contributor
  • *
  • Posts: 63
  • Country: gb
  • Polish guy living in the UK
Re: Need help with repair of Agilent E3647A power supply
« Reply #6 on: April 14, 2018, 10:05:19 pm »
Yeah I have seen this manual, that's where I got schematics.
There's almost nothing about actual repairing, all I found is "troubleshooting hints" where they just tell you to measure few voltage rails, which are ofc fine in my case.
Here's this manual(schematics at the end): http://ridl.cfd.rit.edu/products/manuals/Agilent/power%20supplies/E3646AE2.pdf
« Last Edit: April 14, 2018, 10:07:45 pm by Arek_R »
 

Offline Arek_R

  • Regular Contributor
  • *
  • Posts: 63
  • Country: gb
  • Polish guy living in the UK
Re: Need help with repair of Agilent E3647A power supply
« Reply #7 on: April 15, 2018, 12:25:12 pm »
Quick update, decided to measure some resistors around DAC/ADC and I found two open resistors.
R361 should be 10k, after desoldering shows 60MR+(limit of my DMM). <- connected to signal switch/multiplexer(U30) responsible for ADC portion of the circuit.
R331 should be 30k, after desoldering shows 60MR+ <- connected to quad opamp(U25) from DAC portion of the circuit.
It looks like we may have plague with these green precision resistors, because I have seen guy on youtube(Gerry Sweeney) having same issues with this type of precision resistors being open in these agilent PSUs.

So now I have to replace them, unfortunatelly don't have any resistors with me(moved to another country some time ago).
Will order kit of 50 value of 0805 resistors, replace, see if it works, then order some proper 0.1% low ppm resistors.

EDIT: Found another open resistor - R339 100k
And another one, not open, should be 10k but I read 11.74k which seems to me quite a bit off for 0.1% resistor...
Most of these green 0.1% 10k resistors show 9.99k, so I probably should replace all of them that show +/- 50R off.


EDIT2: Another suspicious resistors that I haven't removed but show slightly off:
MARKING | VALUE | MEASURED VALUE
R357 10k 10.13k
R334 10k 10.20k
R332 10k 10.10k
R329 10k 10.08k

EDIT3: plague continues on main power board as well - R390 100K is open...
« Last Edit: April 15, 2018, 01:22:04 pm by Arek_R »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf