Author Topic: Tektronix AWG2020 Restoration  (Read 6507 times)

0 Members and 1 Guest are viewing this topic.

Offline danpoarchTopic starter

  • Newbie
  • Posts: 8
Tektronix AWG2020 Restoration
« on: March 19, 2015, 05:12:40 am »
I plan on posting a much longer piece on the cleaning and restoration of this waveform generator, but I have a quick question about cleaning the analog output boards.

There's a lot of components on these boards that I either haven't worked with before, or haven't had to deal with as part of a cleaning. I just want to know if someone sees anything here that I should watch out for. The generator works, but has some output scaling issues and gives off failure codes for these particular boards. My aim is to improve its performance, not degrade it. ; )

My typical process:
- Soak with foaming Simple Green Crystal
- Rinse with distilled water (would use DI if I was rich or had a place I could steal it from)
- Spot clean with 99.9% isopropyl while removing the old capacitors
- Spot clean again
- Use Caig DeOxit on the corroded legs of the regulators and xstrs.
- Rinse-repeat, if necessary

This method works beautifully on the beige Macs I restore, and served me well on the other boards on this unit.

The data sheet for the Omron G6H-2 relays says, "When washing the product after soldering the Relay to a PCB, use a water-based solvent or alcohol-based solvent, and keep the solvent temperature to less than 40°C. Do not put the Relay in a cold cleaning bath immediately after soldering." I won't be reworking those, but they look safe to get wet as part of the cleaning process.

I really like this generator and for $300 and a lot, but not unreasonable amount, of elbow grease it has excellent value. The frequencies seem to be dead on (at least at 4.5 digits, which is all I need) and I like having two channels, not to mention the arbitrary waveform features. Like I said, my hope is help, not harm it.

« Last Edit: March 19, 2015, 05:39:48 am by danpoarch »
 
The following users thanked this post: lordhex

Offline danpoarchTopic starter

  • Newbie
  • Posts: 8
Re: Tektronix AWG2020 Restoration
« Reply #1 on: March 19, 2015, 05:26:11 am »
Sorry if I wasn't clear, I'm tearing into these boards and completely replacing the aluminum SMT caps with Nichicon solid polymer CVs. And trying to clean and prevent further damage from the leaking old ones as I go.

Caps:
http://www.mouser.com/Search/ProductDetail.aspx?R=PCV1V220MCL2GSvirtualkey64700000virtualkey647-PCV1V220MCL2GS
http://nichicon-us.com/english/products/pdfs/e-cv.pdf
 

Offline meriororen

  • Contributor
  • Posts: 11
  • Country: id
Re: Tektronix AWG2020 Restoration
« Reply #2 on: April 27, 2017, 09:00:01 am »
Hi,

I know this is an old topic.
I also recently have just got an AWG2020, all working except it fails at Setup CH1 (642) , apparently analog board fault.
Also, sometimes the Clock too has 212 fault, synthesizer board fault. However when I run calib, it passes, run again and it failed.

I opened the case only to find all those leaky caps has been replaced with new throughhole electrolytic caps with either 22uF 50v value
or 33uF 50v value.

 

Offline danpoarchTopic starter

  • Newbie
  • Posts: 8
Re: Tektronix AWG2020 Restoration
« Reply #3 on: April 27, 2017, 02:03:31 pm »
Yikes. That probably means that the boards that the boards were never cleaned. I would think about washing them in Simple Green, or similar, ASAP.

I also discovered two more caps buried in soldered-down RF cages on the board that seems to handle timing/triggering (been a while since I've worked on this project so I think that's what the board does). Haven't quite figured out how I'll get to those to replace them.

If our Signal generators can last thirty years, this thread can last just as long... ; )
 

Offline meriororen

  • Contributor
  • Posts: 11
  • Country: id
Re: Tektronix AWG2020 Restoration
« Reply #4 on: April 28, 2017, 02:47:51 am »
Despite of these faults, actually it can still output waveform. I wonder what effect does these faults cause.

Also, Is this thing can only generate 2.5 MHz? I can't seem to go higher than that frequency in F.G. menu.
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Tektronix AWG2020 Restoration
« Reply #5 on: April 29, 2017, 09:21:03 am »
Hello,
just checked on my AWG2021, same "problem" re the 2.5MhZ!

You are dealing with two units tacked together here: a "function generator" in the FG menu, and a seperate arb. They are different in quite a few aspects, notably that the Arb expects 50 Ohm termination while the FG can also do high impendance loads (to some extent).

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

Offline meriororen

  • Contributor
  • Posts: 11
  • Country: id
Re: Tektronix AWG2020 Restoration
« Reply #6 on: April 29, 2017, 02:09:36 pm »
I actually "figured" it out without reading manual.  :-+

You need to make a waveform, and then "play" it. To make a waveform, use EDIT menu, save it into a file in either diskette or nvram,
then "play" it on SETUP menu.

Here's one in action outputting 30MHz, it's upside down because I am probing the analog board which is located at the bottom. The pic's rather blurry though.

 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Tektronix AWG2020 Restoration
« Reply #7 on: May 02, 2017, 11:48:42 pm »
Please test square waveforms or ramps, so we can easily see memory related problems.

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

Offline meriororen

  • Contributor
  • Posts: 11
  • Country: id
Re: Tektronix AWG2020 Restoration
« Reply #8 on: May 04, 2017, 03:31:02 am »


This is as fast as I can get (right now as I have yet to figure out more of its setup) with 250MHz clock using square wave. 250 cycles of square waves resulting 62.5 MHz. If I go further than 251 cycles I start seeing discontinued waveform.

Will keep posted.


Sent from my iPhone using Tapatalk
 

Offline TAMHAN

  • Frequent Contributor
  • **
  • Posts: 407
  • Country: sk
Re: Tektronix AWG2020 Restoration
« Reply #9 on: May 04, 2017, 02:02:14 pm »
Merio, we want the LONGEST POSSIBLE waveform as to exercise ALL memory cells.

Speed, for now, is not so important.
Feel like some additional tamile wisdom? Visit my YouTube channel -> https://www.youtube.com/user/MrTamhan for 10min tid-bits!
 

Offline TheNewLab

  • Frequent Contributor
  • **
  • Posts: 290
  • Country: us
Re: Tektronix AWG2020 Restoration
« Reply #10 on: July 06, 2018, 06:07:50 am »
Excellent! I am staying on this thread to keep it alive. I would love to hear anymore of your progress. i recently got two AWG2021. It says up to 250MHz. I am not sure how to do it. So far I think it is about turning up the internal clock and I have to create my own custom AWG..and somehow push enough into the file I created to get the 250MHz.

 

Offline lordhex

  • Newbie
  • Posts: 5
  • Country: us
Re: Tektronix AWG2020 Restoration
« Reply #11 on: June 29, 2020, 07:06:19 pm »
Hi - I realize this is an old topic and old posts. How did your restore came out? How is the generator performing? Any issues at all?
YO3FCU
 

Offline tec

  • Newbie
  • Posts: 1
  • Country: de
Re: Tektronix AWG2020 Restoration
« Reply #12 on: March 07, 2021, 02:00:32 am »
Hi,

got an AWG2020 too and replaced all the leaky 22uF caps and cleaned the pcb's - typical Tektronix problem (same with TDS XXX scopes).
Before and after doing this I get the "Setup CH1 - FAIL (642)" error.
Measured all the power rails and they seem to be stable and present.
Not sure if you need to load a "cal" file or anything.
Output looks good on CH1 (arbitrary waveform and FG).

Maybe anyone got further than me. - Also just found the AWG2021 service manual, but the error-code is out of range in the manual.

greetings from Berlin :)
 

Offline Repairguy42

  • Newbie
  • Posts: 3
  • Country: us
Re: Tektronix AWG2020 Restoration
« Reply #13 on: January 05, 2023, 01:47:30 am »
I've got an AWG2021 I bought on Ebay. It is giving trigger fail "810" on second channel at power on diagnostics. I have a second unit which works fine so could fortunately board-swap-troubleshoot. Service manual did not help locate problem. Finally was able to isolate problem to the "Synthesizer Board A10", after trying all other boards.
PN/ 671-3459-00
      389-A371-00
There is no visible issue and I did replace many of the SMD mount electrotytic caps as they are from late 90's.
I have not removed the RF-cages to get at the "SERIAL INPUT PLL FREQUENCY SYNTHESIZER WITH 520MHz PRESCALER {Fujitsu MB1504H}
The RF-Cage looks like a bear to get off.
Would like to repair board or find a replacement but parts are pretty slim for these.
The working 2 channel AWG2021 is a really nice tool and at today's prices such products cost $10K ++
 

Offline gianrisa

  • Newbie
  • Posts: 8
  • Country: de
Re: Tektronix AWG2020 Restoration
« Reply #14 on: November 19, 2023, 09:34:28 am »
Hi, did you progress anyhow with the FAIL 642 I have read around that few of the Fail that are attributed to the Analog board are calibration issues that are coming from a faulty battery. Just to know did you check the battery ? -

VG Gian aus Fürth Bayern
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9246
  • Country: us
  • $
Re: Tektronix AWG2020 Restoration
« Reply #15 on: November 19, 2023, 09:37:35 am »
get the caps near crt, those are really bad
 
The following users thanked this post: gianrisa

Offline gianrisa

  • Newbie
  • Posts: 8
  • Country: de
Re: Tektronix AWG2020 Restoration
« Reply #16 on: November 22, 2023, 11:22:46 pm »
Fail 641 it was due to the battery low the cal data are not peristing during the cal process, new battery no errors

but...

one trouble which I discovered on this device is the following
waveform has a common mode failure, all the 3 waveform have a jump now debugging starts to be nicer and nicer

In the scope you can see 2 traces, the 1 ( top ) is the output of the f.g. and you can see the "jump"
the trace 2 ( bottom ) is basically the pin 8 of the amplifier and there we have already the "jump"
I've disconnected the cable that comes from the DAC and already there I see the malformed waveform ( this defect it is common to all the 3 waveforms ) I did also test with a generated waveform from a custom function
and I got the same result. So My thinking is the following, it cold be that the data have been stored in the ram in the incorrect way or it can be that the DAC converts it in the wrong way.
Any thought ?

Cheers Gian
« Last Edit: November 22, 2023, 11:24:45 pm by gianrisa »
 

Offline danpoarchTopic starter

  • Newbie
  • Posts: 8
Re: Tektronix AWG2020 Restoration
« Reply #17 on: January 03, 2024, 06:30:06 pm »
Following up on this post after 8 years. I've had two kids and received a BSEE since I started this thread. This has happened before, but it's always funny when I google a problem and the first result was my own post from years ago.

I did end up getting this generator mostly off the ground in 2015 and I've attached some of my results.

Here's what I recall doing:
- Backed up all of the settings and presets to a floppy (still worked after all these years)
- Fixed traces on the memory board as necessary (this resolves wave sections at the wrong Y position)
- Replaced the supercapacitor for good measure
- Replacing all SMT aluminum caps on the socketed PC boards (with the exception of those in the RF cage
- Replaced the CMOS battery on the host board (did NOT replace the 10-12 electrolytic caps on that board)
- Reload all the settings

The generator performed pretty well after that. The clock generation seemed on point, and the vertical amplifiers were within reasonable accuracy.

Two calibration issues persisted:

[NOTE: Just noticed that CH1 and CH2 are reversed between the scope and Generator, I have edited to reflect this as AWG-CH1 to say CH1 on the generator)

After power-on:
  CH1 Fail [690]
  CH2 Fail [620]

After 90 minutes of usage:
  CH1 Fail [642]
  CH2 Fail [620]

Also, compare the trace close-ups in one of the JPEGs. AWG-CH1 looks much worse than AWG-CH2. Admittedly, I see now that the scope vertical setting is twice what AWG-CH2 shows, but I recall thinking at the time after much comparison that AWG-CH1 was def in worse shape.

I need to calibrate some reel-to-reel decks, and this generator should be fine for that. My next move will be to disassemble and clean all of the analog board on the underside of the scope. I'm hoping they survive dunking in acetic acid, simple green, etc.

Sorry for the 8 year lag.
« Last Edit: January 03, 2024, 06:33:22 pm by danpoarch »
 

Offline danpoarchTopic starter

  • Newbie
  • Posts: 8
Re: Tektronix AWG2020 Restoration
« Reply #18 on: January 04, 2024, 07:30:43 pm »
Most of the repair pron was shot on my old Canon s95. Those images are stored on my NAS and I'll dig them up to share more of that at some point (in the next 8 years! :palm: )

This shot shows the RF cage on the left, for those that haven't seen them, note the solder seal along the bottom edge. I have an old roll of Radio Shack braid I intend to use up on that.

On the bottom right you can see some delightful cap damage.
« Last Edit: January 04, 2024, 07:32:15 pm by danpoarch »
 

Offline gtr-electro

  • Newbie
  • Posts: 1
  • Country: fr
Re: Tektronix AWG2020 Restoration
« Reply #19 on: January 07, 2024, 09:49:15 am »
Hello, I try to repair a AWG2041 with some fail on the Analog board.
I succeed to repair Setup CH1 FAIL 610, but now have Fail 620 like yours....
do you have some hints about the error codes?
someone on the forums said that is related to relays...
I checked all the relay positions when OFF and whenON ? they are all good!

all advices are appreciated! and finger crossed to repair your!

Best regards
« Last Edit: January 07, 2024, 08:24:02 pm by gtr-electro »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf