Need Help Troubleshooting Agilent E3617A DC Power Supply

[wolfienuke]

I have two Agilent E3617A DC power supplies that fail to output voltage  . I have done extensive troubleshooting and awaiting spare parts, but would like a second opinion about my theory.

Problem/Symptoms:
Power supply turns on, and has all visual indications.
Power supply has no voltage output.
Operating the current control knob changes constant current limit.
No visual signs of damage to any internal components.
No blown fuses. Input fuse checked

Operating the power supply CV controls from external voltage source (per manual) allows full operation and symptoms disappear. (This info seems most vital to troubleshooting)
-Output voltage is generated on external constant voltage when CV switch is in external and a voltage applied to CV+ and CV-. (all rear control switches in the up position, except for CV)
-No Output voltage when the switches are on Internal/Master (all rear control switches in the up position)

I followed the troubleshooting steps for "Table A-5. Low Output Voltage Troubleshooting" (pg. A10-A11)...
1. Check turn on of Q1 and Q4 by disconnecting emitter of Q9.
     b. Output voltage increases. (No fault)
2. Check turn off of Q9 by shorting point 1 to +15 V.
     b. Output voltage increases. (No fault)
3. Eliminate constant current comparator as a source of trouble by disconnecting anode of CR22.
     a. Output voltage is increases.
4. Check voltage from pin 13 to pin 12 of U9.
     This was kind of ambiguous, I don't know whether pin 13-12 is neg-pos. Either way, manual says to U9/U10 may be defective.
     Check R85 is open. - This is the voltage or current control knob and both were verified to be operating normally.

I have verified the resistance range of the control knob potentiometers.
I have verified the 10V supply to the voltage/current control knobs.
I have verified Q9 to be functional by desoldering and checking with transistor checker and multimeter on diode check.
On step 2 when manually switching the transistor, the output voltage increases quickly and when disconnected, slowly decreases.

I have verified that the power supply can operate normally when using external voltage signals as per "Remote Voltage Programming, Constant Voltage" (p 1-10). In this state, the output voltage matches the specification of the operating manual. Doing the same test for Constant Current has no effect.

This tells me that the internal CV control circuit is at fault... which exact part of the circuit, I am unsure. I have ordered replacement parts for U9 (LF347BN) and U10 (LT1055CN8), but am not convinced of their failure.

Can anyone assist on this problem? I can provide as much additional information as necessary. The cost of outsourcing the repair is too high compared to the value of these power supplies, and I know that the problem can be fixed cheaply.
I have posted the applicable excerpts from the service manual. The manual can be downloaded from Agilent (http://cp.literature.agilent.com/litweb/pdf/5959-5310.pdf) but is missing the circuit diagrams and parts list. Here is the full manual (http://www.physics.fsu.edu/users/Wahl/labmanuals/instruments/ps/AgilentE361516AManual.pdf). :-

Thanks!
-wolfienuke

[free_electron]

it works with external control but not internal ? check the switches in the back of the machine fisrt !

[wolfienuke]

Switches are in the correct positions.

Thanks for the "idiot check". I actually checked that early on. I forgot to mention in the post... (Post updated)

It works on external constant voltage when CV switch is in external and a voltage applied to CV+ and CV-. (all switches in the up position, except for CV)
It doesn't work (as I described) when the switches are on internal (all in the up position)

[just_fib_it]

There's no real difference between local and remote control, the switch on the back (S3, CV L/R on the schematic) just switches either the wiper of the pot (R84) or the rear terminal to R86.

When set to internal, does the signal of the pot arrive at R86?

When looking at the PCB from the front of the unit, the left side of R86 should be equal to the voltage at the wiper of R84. The right side should be half that voltage (because R86 forms a voltage divider together with R68, both are 46.4K). This then goes to the positive input of the U9 LF347.

[wolfienuke]

There's no real difference between local and remote control, the switch on the back (S3, CV L/R on the schematic) just switches either the wiper of the pot (R84) or the rear terminal to R86.

When set to internal, does the signal of the pot arrive at R86?

When looking at the PCB from the front of the unit, the left side of R86 should be equal to the voltage at the wiper of R84. The right side should be half that voltage (because R86 forms a voltage divider together with R68, both are 46.4K). This then goes to the positive input of the U9 LF347.

That is a good point... Possibly the wiring from the wiper or bad contact at S3. I will check that out on Monday, seems simple enough of a check. I have checked the voltage at the wiper on the pot, but overlooked the check of that signal to U9.

Thanks for the tip, although it is strange to think that the failure may be in a short (and nicely secured) wire from the potentiometer on the front panel to the PCB. The switch may be more likely, but I am sure this is a switch that has had little to know use. But then again, stranger things have happened...

[just_fib_it]

I once studied the service manual a bit and got a headache of all the connections on the schematic going out to the rear terminals and back into the main circuit, so I color-coded them. The annotated PDF is here: https://github.com/ee-lco/e361xa/blob/master/doc/AgilentE361516AManual.pdf?raw=true

BTW, I'm looking into building a E3617A myself and am curious if the PCB of the E3617A differs from the E3615A (which I have here). Would you mind posting a few pictures of the PCB?

[wolfienuke]

I once studied the service manual a bit and got a headache of all the connections on the schematic going out to the rear terminals and back into the main circuit, so I color-coded them. The annotated PDF is here: https://github.com/ee-lco/e361xa/blob/master/doc/AgilentE361516AManual.pdf?raw=true

Thanks for that link. It will be of help following the traces.

BTW, I'm looking into building a E3617A myself and am curious if the PCB of the E3617A differs from the E3615A (which I have here). Would you mind posting a few pictures of the PCB?

Sure. The boards are at my lab, and I will take some pictures on Monday for you. If you know of any particular PCB area that you will need detailed pics, let me know...

[Kevin.D]

Trying to follow that schematic  without fib_it's colours  made me  dizzy  . My moneys on  r84 wiper  or s3 bad contacts .(bit of a coincidence though you having 2 with exactly the same fault symptoms)
edit :-
quote "I have verified the resistance range of the control knob potentiometers."

Sorry didn't see you had already checked that .discounts bad R84 then

[just_fib_it]

Sure. The boards are at my lab, and I will take some pictures on Monday for you. If you know of any particular PCB area that you will need detailed pics, let me know...

Thanks, that would be really helpful. The main differences are in the SCR pre-regulator circuit, which is in front of the heat sink next to the transformer (on the E3615A, I'm assuming the E3617A is not that different).

[wolfienuke]

Trying to follow that schematic  without fib_it's colours  made me  dizzy  . My moneys on  r84 wiper  or s3 bad contacts .(bit of a coincidence though you having 2 with exactly the same fault symptoms)
edit :-
quote "I have verified the resistance range of the control knob potentiometers."

Sorry didn't see you had already checked that .discounts bad R84 then

These are from a college lab with about 30 of them. The lab is used by engineering students at all knowledge levels and all different majors (mechanical, civil, etc...), so my assumption going into it was user faults such as dead shorts to ground, or over use of common adjustment knobs - which I have verified at least twice with multiple measurements and meters (voltage to common, voltage differential, and in circuit resistance).

I have about 5 or 6 of these not working, but these 2 had similar issues which I though would be a simple fix.

Honestly, it's usually a safe bet to assume the potentiometers, especially in this scenario. I did...

[wolfienuke]

https://www.dropbox.com/sh/90ypotbpt7ogtim/sUvOdRvvSr

Here are the board pics for the E3617A. If I missed anything or need better pictures of a specific component, let me know. I also have one open and can get some pictures of the underside of the pcb if needed...

Thanks for the help... the problem was the CV local/remote switch in the rear panel. The local mode was very high impedance. I bypassed the switch until I can get a replacement. Both power supplies work now.

I have one left that will not work, and the switch has the same problem but the bypass still does not fix. I will need to do some more poking around to find the issue.

Your highlighted circuit diagrams were very helpful.

-Justin

[just_fib_it]

https://www.dropbox.com/sh/90ypotbpt7ogtim/sUvOdRvvSr

Here are the board pics for the E3617A. If I missed anything or need better pictures of a specific component, let me know. I also have one open and can get some pictures of the underside of the pcb if needed...

Thanks for taking those pictures  If you haven't closed up the supply yet, some pictures of the underside would be nice, to see the layout of the traces.

I was mostly interested in seeing if there are any differences in the PCB layout that would make it difficult to have a single main PCB that supports all four variants (E3614A...17A) and secondary boards with the different SCR circuits. From the pictures it seems that that shouldn't be a problem.

[ober]

So how do you actually take apart these things so get can get at the underside?
I'm a bit wary to unscrew the heavy transformer from the sheet metal...