Author Topic: Keithley 2601 (Read 6812 times)

gaminn · « **on:** March 21, 2016, 06:29:34 pm »

Hi,
I have Keithley 2601 sourcemeter (manufactured 2005). Suddenly, it outputs -25 V constantly. No matter what I set on the source meter. Even, when the output is turned off. The "meter" part of the sourcemeter is OK, it measures real voltage and current.

Any advise?

TopLoser · « **Reply #1 on:** March 21, 2016, 06:34:15 pm »

There's a good chance it's still under warranty. Email Gerorge at Tektronix with the serial number and he will let you know the warranty status. If it's in warranty he'll send you an RMA and arranged for it to be collected from you.

gaminn · « **Reply #2 on:** March 21, 2016, 06:43:22 pm »

Oh, I'm sorry for the typo - it was manufactured in 2005, not 2015.

TiN · « **Reply #3 on:** March 21, 2016, 06:52:26 pm »

Take it apart and lets see what's inside. I'd expect few mosfets on output stage are blown.

gaminn · « **Reply #4 on:** March 22, 2016, 11:03:25 am »

Hi, I took it apart. I checked 4 big power mosfets - they all switch, they are ok.

I noticed, that the area near power mosfets gets really hot. I can't even touch some components for longer time, they are really hot.

If I run sourcemeter without power mosfets it outputs still - 16 V, no matter if the output is switched on or off. The output slightly changes when I set some output voltage on the front panel.

plesa · « **Reply #5 on:** March 23, 2016, 11:02:03 pm »

Thanks for pictures and tear-down. I have no experience with 2601, but I will try to replace the compactors LM339 and check zeners and current sources.

TiN · « **Reply #6 on:** March 24, 2016, 01:21:00 am »

gaminn
Can you post hi-res pics or upload them (no size limits).
Since we don't have schematics for 26xx, it's bit more difficult to track everything down.

If I were you, I'd start looking power path from output connector and trace all power components on the path.

gaminn · « **Reply #7 on:** March 24, 2016, 11:03:14 am »

Thanks for your help. I uploaded picture of the board: ftp.xdevs.com/pics/Keithley2601.jpg

I checked 4 big power mosfets (two IRFP240 and two IRFP9140) and they seem to be OK. They are connected in a such way, that all their sources are nearly on the same potential (2.5 ohms between them max) and they are somehow connected to L4 and to LO output. Drains of N mosfets are connected together and drains of P mosfets also.

I don't see another power components on the board. 4 big caps are probably OK, two have 17 V and two have 53 V.

I measured either 0 or 18 V on all OUTs of all 6 LM339s (measured against their GND). I will eventually replace them when I buy them.

HI output is +15 V higher than LM339s GND and LO output is +40V higher than LM339s GND (hence -25 V between HI and LO).

The area around CZTA42s and CZTA92s gets hot.

TiN · « **Reply #8 on:** March 24, 2016, 04:24:31 pm »

How about pair of DPAK transistors in center and LT1166 (U56)?
Also check resistors R27-R84, I believe they are 4.7ohms each.

What are power supply rail voltages?

gaminn · « **Reply #9 on:** March 24, 2016, 06:56:31 pm »

That DPAK NPN and PNP transistors are OK. I don't know how to check LT1166. Without it, output voltage (LO to HI) drops to almost zero. Also two TO220 dual schottky diodes are OK.

The board's power rails are +-5 V, +-15 V, +-17 V, +-55 V, +-130 V

plesa · « **Reply #10 on:** March 24, 2016, 08:12:28 pm »

I looks like quite interesting repair project, but I do not have experience with SMU repairs. Based on description I suspect zeners and comparators as most sensitive to the overvoltage on input-

gaminn · « **Reply #11 on:** March 24, 2016, 08:43:18 pm »

LM339 comparators are OK. Measured their input voltage and the corresponding output has always correct value (0 V or 15 - 18 V).

What components on the board are zeners? That MELF orange components?

Kleinstein · « **Reply #12 on:** March 24, 2016, 09:01:15 pm »

Likely the next step would be getting a rough idea of the circuit plan, especially the output stage I would guess it to be similar to other keithley SMUs (e.g. 2400 there is some information available about this one). So something like using a floating regulator and two raw supplies (e.g. +-15 V and +-55 V) for the power.

The comparators and JFET switches are more like a problem it the meter part is not working.

For the power output, it's more like the power MOSFETs or the supply rails.

It might also help to know how the meter got damaged.

gaminn · « **Reply #13 on:** March 24, 2016, 09:30:51 pm »

I removed power mosfets and connected external power supply to LO and HI. Both voltage and current readings are ok for my source meter. SO really only the source part is KO.

I didn't notice how the damage happened. I was using the sourcemeter as liion battery simulator (it supplied/consumed max 4.2 V at max 500 mA).

plesa · « **Reply #14 on:** March 26, 2016, 06:20:28 pm »

If only the source section is faulty check the 2400 repair manual [pages 60-63].
http://www.tek.com/keithley-source-measure-units/smu-2400-series-sourcemeter-manual/model-2400-sourcemeter-service-manu
The source is H class apmlifier. I used this hint during 6517 electrometer repair. In my case the power MOSFETs, LM234 current sources, BJT and few zeners were damaged.
I have 2604B, if you needs any mesurement.

TiN · « **Reply #15 on:** March 27, 2016, 12:48:05 am »

I was seeing some block diagrams of 26xx series SMUs output stage on one of Keithley's patents before. I think it's bit different to 2400, as it's using optoisolated drive train (but that was long time ago I checked, don't take my word directly).

gaminn · « **Reply #16 on:** March 28, 2016, 02:08:57 pm »

Thanks guys

I discovered that two AQV252 optoisolated switches are probably bad. The LED side of the switch seems to be OK (Vf = 1.1 to 1.2 V @ 10 mA) however the transistor sides of the switches conduct all the time (they are not fully open) even with no LED current. Nothing changes when current flows via the LED side. So I hope replacing them will help. We will see...

gaminn · « **Reply #17 on:** March 30, 2016, 11:23:48 am »

I replaced AQV252. Nothing changed.

Do you know what this component is:

TiN · « **Reply #18 on:** March 30, 2016, 12:43:31 pm »

High-value resistor usually.

gaminn · « **Reply #19 on:** March 30, 2016, 01:13:27 pm »

I'm afraid I'm not able to repair it. I will sell parts from the sourcemeteron eBay. Should anyone need something...

plesa · « **Reply #20 on:** April 30, 2016, 04:29:09 pm »

What can you help with repair is Keithley 236/237 service manual.
https://doc.xdevs.com/doc/Keithley/236_237_238/236_902_01D.pdf
I'm just using it for Keithley 2400 repair. On page 89 / chapter 4 Theory of operation are simplified schematic and lot of magic explained.
Or alternative
http://exodus.poly.edu/~kurt/manuals/manuals/Keithley/KEI%20238%20Service.pdf
on page 66 with minor differences.
Or you give up and send unit to Keithley for board swap?


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Keithley 2601 (Read 6812 times)

Share me