Author Topic: Agilent 53132A Counter Repair (FIXED and Upgraded) (Read 45856 times)

HighVoltage · « **on:** January 12, 2016, 09:13:20 pm »

This 53132A counter was suppose to work well, as described by the seller but it is dead.
Symptoms: No display on, No fan, no reaction at all.
Both fuses seem to be ok.

I already got arrangements with the seller to keep it and now I have another repair job.
This thread is started at the beginning of the repair and at this stage I have no idea what the problem is.

After opening the instrument I found the fan totally plugged up with dirt and lots of dirt blown in
to the power supply area. So first I will clean the inside completely and then start analyzing the problem.
Judging by the dirty PSU, I suspect the first problem there.

After I repair the unit, I want to upgrade the the option 10/12 for a stable frequency reference and
also upgrade to to the 3 GHz Prescaler on Channel 3.

At the end there will be a calibration and comparison to my HP GPSDO with my 53230A counter.

So, anyone here having experience with the 53132A and it's problems?
One thing that sticks out right away is a missing U24 chip.
Hmmm, may be it is required for the time base upgrade?

Interestingly, this one has the Channel 1 and Channel 2 inputs on the back as well.
I think this was an option, because I have seen other units without these connectors on the back.

TheSteve · « **Reply #1 on:** January 12, 2016, 09:34:07 pm »

Wow, that is some nasty crud. I don't think U24 was ever used. The high stability option just plugs in, no changes to the main board. Same with the 030 option. I added one of the 030 ebay specials to my 53132A as well.
Being the fan should always run and the power is handled by a soft switch, the power supply really screams to be looked at first.

btw, the rear connections are option 060, they are simply in parallel with inputs 1&2, no switching or anything.

HighVoltage · « **Reply #2 on:** January 12, 2016, 10:05:03 pm »

Thanks for the feedback TheSteve,

I found the two options to be available on ebay Poland and they seem to be at an ok price level:
- 3GHz Prescaler = $89
- Ultra High stability Timebase = $ 105

Full schematics are also available here:
http://www.ko4bb.com/manuals/index.php?dir=HP_Agilent/HP_53131A_53132A_53181A_Universal_Counter

TheSteve · « **Reply #3 on:** January 12, 2016, 10:37:27 pm »

I believe there have been several different aftermarket high stability time bases made. I've seen one that is adjusted using a trimmer, and another that uses an A/D and is calibrated electronically like the factory option.
When ordering verify it is a proper clone that uses the A/D.

ManateeMafia · « **Reply #4 on:** January 13, 2016, 04:05:19 am »

I just received my option 030 from Poland. It looks good but the IC closest to the center of the board has pin 3 clipped off. Would anyone with a recent version confirm this is the case?

I contacted the eBay seller and hope to hear back soon.

ManateeMafia · « **Reply #5 on:** January 13, 2016, 04:39:58 am »

That is for the OCXO option. I did buy a blank board from Gerry Sweeney and it has worked well.

The 030 is for the 3GHz input divider. It is strange that the eBay listing shows all pins connected so I won't be surprised that this was a mistake. I will have to find the schematic for it and compare it to this board.

ManateeMafia · « **Reply #6 on:** January 13, 2016, 05:30:52 am »

Just heard from the seller that the pin was intentionally clipped.

Jf2014 · « **Reply #7 on:** January 13, 2016, 08:45:28 am »

PSU problem ?

Maybe you need a diagram PSU.

HighVoltage · « **Reply #8 on:** January 13, 2016, 10:03:49 am »

Quote from: Jf2014 on January 13, 2016, 08:45:28 am

PSU problem ?

Maybe you need a diagram PSU.

WOW!
I was looking for the PSU schematics and it was not part of the standard schematics and I could not find it.
It seems you made it?
Impressive, Thanks so much.

Orange · « **Reply #9 on:** January 14, 2016, 01:27:11 pm »

The issue with the 53132 counter series is that the are constantly powered, and the fan is blowing. This is a very disturbing noise. I have my counters disconnected from the mains if not in use.
If you get an oven controlled time-base in this thing, you will have the PSU running all the time, which will begin to irritate you (at least I had this).

Better solution is to use the external reference input and uses a GPSDO or rubidium source. It is much more accurate, and you don't need to power the counter all the time.

BTW, I noticed from the pictures that the main board in the counter has only one flash ROM for the firmware. The older type main boards have 4 separate EPROMS, which can be upgraded easy. Perhaps Agilent has a new FW upgrade possibility for this type of motherboard.

Anyhow it is a great counter

TheSteve · « **Reply #10 on:** January 14, 2016, 06:18:19 pm »

The counter does look to have newer firmware then I have seen posted anywhere - 4806. Not quite as easy to dump and post as it was with older units that had four socketed PLCC devices.

Vgkid · « **Reply #11 on:** January 15, 2016, 01:21:55 am »

Could you either substitute a different fan to run off of another rail(non standby rail).
Or, use a relay to turn on the fan(off the 5v rail, or others) and keep it wired the same way.
The 2nd one is how that is implemented in the 5335.

HighVoltage · « **Reply #12 on:** January 15, 2016, 01:11:31 pm »

Interesting comments everyone - thank you, I will look in to the fan issue, when I put it back together.
I probably have to get a new fan anyways.

Right now is cleanup time.
I took the instrument completely apart now and started with the display and front cover.
All cleaned up now, looks like new - kind of.

Next is to clean the PCB in alcohol.

EPTech · « **Reply #13 on:** January 15, 2016, 01:36:58 pm »

Hi there,

If the fan has died somewhere along the way, you should check the condition of the input cap but most certainly the low ESR output caps on the SMPS. I would just replace them by default, that may already fix the problem. I am not affiliated to any brand but I would recommend the FC, FM or FK series from Panasonic. Also check the high ohm resistors.

Does the supply work without load attached?

HighVoltage · « **Reply #14 on:** January 15, 2016, 01:53:23 pm »

Quote from: EPTech on January 15, 2016, 01:36:58 pm

Hi there,

If the fan has died somewhere along the way, you should check the condition of the input cap but most certainly the low ESR output caps on the SMPS. I would just replace them by default, that may already fix the problem. I am not affiliated to any brand but I would recommend the FC, FM or FK series from Panasonic. Also check the high ohm resistors.

Does the supply work without load attached?

Thanks EPTech
The PSU is sitting in alcohol right now since it was so full of dirt, some components were not visible.
I wanted to clean it, before I would test without load.

The corrosion on the frame is removed with a rotating steel brush but it was already deep in to the metal, even after cleaning it looked horrible.

So, I decided to cover this area of the frame with some paint.
And it seems to turned out alright.

HighVoltage · « **Reply #15 on:** January 15, 2016, 02:05:41 pm »

OK, the PCB is cleaned with alcohol and stayed in an oven for 1 hour at 50°C to get the rest of the alcohol evaporated.

EPTech · « **Reply #16 on:** January 15, 2016, 02:07:32 pm »

Really that corroded!

Makes you wonder from what environment it came out of.

I do not want to discourage you but there may be more going on than some bad caps.

Be sure to post the pics after that clean-up.

HighVoltage · « **Reply #17 on:** January 15, 2016, 02:12:18 pm »

Yes, it must have been awful in that environment for this unit.
I am actually surprised how some people treat expensive test equipment badly.
This was not the first instrument I bought used that was full of sticky and corrosive dirt.

All the sticky stuff is removed from the PCB and it looks really good now.
Here are some more PCB detail pictures.

HighVoltage · « **Reply #18 on:** January 15, 2016, 03:37:30 pm »

PSU is cleaned up with alcohol and was also one hour in the oven at 50°C
From the first look the caps seem alright (not domed out).
Time to take some measurements.

HighVoltage · « **Reply #19 on:** January 15, 2016, 04:41:03 pm »

Voltage output was all zero on all wires at the output connector.
So, I started probing at the input of the PSU.

The fix was so easy....
Probing the voltages right after the input rectifier revealed open R2.
R2 is a 15K Ohm 5 Watt power resistor.

Replacing it gave about 101V across the resistor.
That is just 0.66 Watt power and far lower than the 5 Watt rated power of the resistor.
It probably died from too much heat because of all the dirt in the PSU.

I will check the caps anyways of ESR, but I don't think they are bad.
They are all Nichicon 105 degree rated caps.
Actually looking closer at the PSU, it seems to be good quality built.

Now, without connecting the PSU to the Agilent PCB, I am getting these output voltages:

+ 5V = 5.19V
+ 12V = 12.08 V
-12 V = -11.88 V
- 9 V = -9.16 V (with separate 9V ground)

Next:
Install the PSU in to the 53132A without fan, since I don't have a fitting new one.

HighVoltage · « **Reply #20 on:** January 15, 2016, 05:36:19 pm »

Counter is fixed and working perfectly so far, tested frequency on Chanel 1 and 2.

Now I will be waiting for the parts to arrive for upgrading to 3 GHz prescaler and oven controlled oscillator.
After that I will calibrate it, based on my HP GPSDO
Parts are ordered ... stay tuned.

Thank you all for help and suggestions.

EPTech · « **Reply #21 on:** January 15, 2016, 05:40:01 pm »

Right on!

Another fine instrument added to the bench. Considering all the work you have put into it, it is probably in better shape than many 2nd handers out there.

Congrats

HighVoltage · « **Reply #22 on:** January 15, 2016, 06:07:27 pm »

Thanks.
Self Test and extended Self Test is passed without any problems.
All seems just perfect so far.

I found a fitting fan with low noise.

Sharkoon System Fan
- 7 blade fan
- 40x40x20 mm
- 6200 RPM
- 11.8 m^3/h
- 26 dB(A)
- Euro 4,38

May be this one will also solve the loud fan noise that some complained about.

TheSteve · « **Reply #23 on:** January 15, 2016, 06:56:25 pm »

Nicely done! Some one must have been measuring the frequency their fish were moving at and mounted it above a bubbling aquarium.

I plan to do the hardwired switch mod as done by Gerry Sweeney on mine, I always use an external reference anyway.

HighVoltage · « **Reply #24 on:** January 16, 2016, 02:24:01 pm »

Someone asked me, what kind of alcohol I am using for cleaning the parts and PCB's.
What I found to work really good is Flux Thinner that we use for an automatic lead free soldering robot to adjust the density of the flux.
It comes in 20 l canisters and based on pure Propanol-2
Here is the label from the canister.
Somehow this works much better that the alcohol I can buy at a drug store.

HighVoltage · « **Reply #25 on:** January 21, 2016, 11:25:30 am »

The "upgrade" parts arrived from Poland and got installed today.
After turning on the instrument it gave the message "Not Calibrated", so it detected the 10 MHz OCXO alright.
The 3 GHz Pre-Scaler also seems to work.

Here are some pictures of the installed upgrades.
Next is to calibrate and check all specifications.

WN1X · « **Reply #26 on:** January 21, 2016, 01:01:47 pm »

sorenkir · « **Reply #27 on:** January 21, 2016, 01:41:59 pm »

Hi,

I have replaced the noisy fan of my HP53181A with the following: Sunon A40201V4 (Super Silence Series

)
http://www.digikey.com/product-search/en?mpart=HA40201V4-000U-999&vendor=259

The fan is very quiet by itself (12.8dB-A), but when installed, due to airflow & high rotation speed, it is still noticeable in a quiet environment.

Michel.

HighVoltage · « **Reply #28 on:** January 22, 2016, 05:03:58 pm »

Thanks for the Digikey link to the fan.
I found a similar fan in Germany and installed it.
It is very quiet and in the lab I do not hear the noise of this instrument.

Calibration was too easy with a good 10 MHz source.
Instrument is finish now.

All done now.
Thanks to all that helped me with this one.

Vgkid · « **Reply #29 on:** January 23, 2016, 12:45:23 am »

Is that fan strong enough, 5.5 cfm is not much.

HighVoltage · « **Reply #30 on:** January 24, 2016, 02:58:38 pm »

Quote from: Vgkid on January 23, 2016, 12:45:23 am

Is that fan strong enough, 5.5 cfm is not much.

It seems very strong and blows very good and steady out of the back.
I also have a 53131A with double oven reference and still an original fan in there and it blows about the same.
So, I see no worries there.
It has been running for about 48 hours now and seems to be very stable.

Nuno_pt · « **Reply #31 on:** January 24, 2016, 03:17:20 pm »

Well, there is one mod i can do on my 53181A with Polish channel 2 option, will change the fan since it's very annoying.

I've found this two, I suppose the original it's 12Vdc fan, I don't have the schematic or manual on hand:

Ebmpapst 412/2H-048
12Vdc
100mA
7000rpm
12,5m3/h
25dB

and

Ebmpapst 412
12Vdc
0,9W
75mA
6000rpm
10m3/h
18dB

but this are more expensive about 9 to 14€

HighVoltage · « **Reply #32 on:** January 25, 2016, 09:11:55 pm »

Thanks for those fan part numbers.
The fan I installed works very well.

I had the counter officially calibrated and it passed all tests perfectly.
Here it is hooked up to my HP GPSDO and comparison to my 53230A
Seems right perfect.

Nuno_pt · « **Reply #33 on:** January 25, 2016, 09:27:16 pm »

HighVoltage,

Can you tell how the fan is attach to the board, solder wires or connector?
Can you confirm that the fan is 12Vdc?
I'm not at home till almost the end of the week, so no chance to look on the inside.

soul40k · « **Reply #34 on:** January 26, 2016, 02:02:30 am »

Hey guys, wondering if you have any possible solutions to a Front End/Measurement self test fail on this?

I have all of my information on my laptop which I left at work unfortunately, but from what I remember, I have near perfect V's from the PSU, it had many issues on the logic side though, very hot MPU chip, which when replaced, fixed every self test issue I had except for Front End/Measurement test.

I bypassed the test screen and attached the reference out to both channels, and it will measure perfect 10MHZ, but if I take off the connections, it seems to float at any random amount with nothing being input. If I am not mistaken, I had a low VCC/GND ohmage (below 20ohms), and if I were to touch the board while running, it would stabilize the read out, but still not bring it to zero with nothing being input.

Sorry for this being my first post, just got into eev, literally starting from the beginning vids/pods, hope to be here more often!

Orange · « **Reply #35 on:** January 26, 2016, 08:40:14 pm »

Quote from: Nuno_pt on January 25, 2016, 09:27:16 pm

HighVoltage,

Can you tell how the fan is attach to the board, solder wires or connector?
Can you confirm that the fan is 12Vdc?
I'm not at home till almost the end of the week, so no chance to look on the inside.

The fan is a 12 Volts type with solder leads. It's a 40x40x20mm

Nuno_pt · « **Reply #36 on:** January 26, 2016, 08:51:42 pm »

Thanks Orange.

uncle_bob · « **Reply #37 on:** February 03, 2016, 02:03:10 am »

Hi

Just a comment from a latecomer to the thread. Based on a fleet of about a hundred 53131's 132's and 181's - they mostly die from a problem in the power supply. So far we have gotten by simply swapping supplies out of the 181's into the others (the 181's are no longer used). I'll take a look at a few supplies next time we have a pile of failures to go through.

Bob

TheSteve · « **Reply #38 on:** February 03, 2016, 04:59:49 am »

I have a broken 53132A inbound, no idea what is wrong with it. If it is something simple I'll post it in this thread, if it turns into a big process I'll start a new thread.

HighVoltage · « **Reply #39 on:** February 03, 2016, 07:57:12 am »

Lately I have been blessed with repairs that turned out to be simple. You just have to take the time and go step by step through the signal path and eventually the problem will emerge.

For many days no, I have this counter hooked up to my HP GPSDO and out of the 12 digits, only the last two are changing
It is quiet impressive, how stable the counter and the GPSDO is.
I calibrated it once against the "locked" GPSDO and since then the system has been extremely stable.

And yes, it was a 12V fan that was installed in this counter.

uncle_bob,
since we now have the schematics for the power supply, you should be easily figure out, what the problem is.
BTW, how come you have hundreds of these ?
What for?
Pictures?

crispy_tofu · « **Reply #40 on:** February 03, 2016, 08:22:44 am »

Quote from: TheSteve on February 03, 2016, 04:59:49 am

I have a broken 53132A inbound, no idea what is wrong with it. If it is something simple I'll post it in this thread, if it turns into a big process I'll start a new thread.

Can't wait!

uncle_bob · « **Reply #41 on:** February 03, 2016, 12:15:55 pm »

Quote from: HighVoltage on February 03, 2016, 07:57:12 am

Lately I have been blessed with repairs that turned out to be simple. You just have to take the time and go step by step through the signal path and eventually the problem will emerge.

For many days no, I have this counter hooked up to my HP GPSDO and out of the 12 digits, only the last two are changing
It is quiet impressive, how stable the counter and the GPSDO is.
I calibrated it once against the "locked" GPSDO and since then the system has been extremely stable.

And yes, it was a 12V fan that was installed in this counter.

uncle_bob,
since we now have the schematics for the power supply, you should be easily figure out, what the problem is.
BTW, how come you have hundreds of these ?
What for?
Pictures?

Hi

Needless to say, the counters are not mine. We use them at work. To avoid the whole "disclaimer" thing I don't post pictures of work. I'd rather not become their spokesman ... That said, they are in the frequency products / timing industry.

Bob

TheSteve · « **Reply #42 on:** February 04, 2016, 05:06:37 am »

Received my "parts/repair" 53132A that was supposedly missing parts. The only part I know is missing is the back plastic bracket. I plugged it in and absolutely nothing happened. OK, lets take a closer look. The line and neutral wires were desoldered from the IEC connector. I looked over the supply and didn't see anything that looked bad so I reconnected them. The power supply seems fine and tests ok, the fan is dead of course. The counter itself fails the front end test and measure tests. I also see all of the HP branded PLCC chips near the front end have been resoldered or possibly changed. So perhaps this will be a bit of a challenge afterall. Looking at the schematic I should be able to feed a signal in and follow it through. I also have a 100% working 53132A to compare to so hopefully that will narrow things down pretty quick. Being both channels fail I am thinking the problem is a portion they have in common.

HighVoltage · « **Reply #43 on:** February 04, 2016, 10:57:50 am »

It looks like you have a much larger repair job than I had. Looking forward to your progress.

I really don't like it, when some instruments are sold as defect and then when you get them, you find out that someone has tried to fix it and usually caused more damage.

You are probably right that the fault has something to do, that is common to both input channels.
Is the self test showing any error numbers?

uncle_bob · « **Reply #44 on:** February 04, 2016, 11:59:16 am »

Hi

Be careful of the HP custom ASIC chips in these counters. About the only way I have ever seen to get them is to buy an entire board. You do *not* want to pay HP prices on those boards.

Bob

TheSteve · « **Reply #45 on:** February 04, 2016, 06:45:27 pm »

We shall see what the problem is. The front ends are totally fine. It looks to be something just after that. I may be buying a parts board in the future! The whole thing was $165.00 so I have a budget for parts.

soul40k · « **Reply #46 on:** February 05, 2016, 02:09:08 am »

Quote from: TheSteve on February 04, 2016, 06:45:27 pm

We shall see what the problem is. The front ends are totally fine. It looks to be something just after that. I may be buying a parts board in the future! The whole thing was $165.00 so I have a budget for parts.

That is where I sit right now, the front looks fine, really weird actions with just a completely spastic display with nothing connected, but I touch it when live, on the bottom, and it stops. This is going to be the death of me, probably something really simple that I am just missing.

HighVoltage · « **Reply #47 on:** February 05, 2016, 08:13:11 am »

Can you show us in the schematic where you have to touch to make the display stable?

TheSteve · « **Reply #48 on:** February 05, 2016, 09:18:22 pm »

Spent some time last night working on my 53132A. The entire time I just kept hearing Shahriar in my head saying "Follow the signal path" lol.

Anyway, some progress was made and I know I'm close. While following the signal flow between the front end and the FPGA I kept getting that feeling something obvious was wrong yet I couldn't see it. I briefly thought maybe the FPGA was bad as it was not generating any reset signals but then it occurred to me that it probably had to detect some sort of signal before it would reset the count circuitry. This proved to be true as selecting input two and then going back to input one would generate 2 or 3 reset pulses that could be traced through the logic. At this point I looked at why I was getting no LED indication showing the selected input or gate samples. So I focused on U50. I had previously gone through all of the logic chips and verified +5, -5 and ground but I thought I'd do it again on U50. The -5 was nice and steady on my 34461A but the +5 was only steady for the first 3 digits. A quick hit with the soldering iron solved that problem and I now had input and gate indications. I still received the failed front end and failed measure errors at boot but it correctly measured the frequency being input which it had never done before. I should also mention at this point I had removed U44 which is a buffer opamp to set the trigger levels as it just didn't look right comparing its in and out voltages. I gave channel 1 a fixed trigger level with a resistor. After seeing it count the frequency properly I reinstalled the original U44. At this point it will not count anything at all and still gives the same errors. I get the feeling I am now at the point of the original failure which started the previous owner resoldering everything. The bad solder connection on U50 was caused by the previous rework as nearly all of the logic chips in the front end have been removed and resoldered(maybe replaced, I'll never know). It was late so I had to pack it in for the evening. I will take a very close look at U44 tonight though. Unless I can say 100% for certain it is good I will order a replacement from Digikey, it's in stock for 5 to 6 bucks and may be all that is really wrong with the unit. I know it can count a frequency correctly so there there can't be much else wrong right?

btw, the previous person that did the rework has some skill but not really enough to do a proper job. The soldering they did is far from excellent and they removed a part or two that there was clearly no need to remove as they can easily be verified 100% good in circuit. They also left the signature amateur calling card in the top of one of the relays in the form of soldering iron melt marks.

soul40k · « **Reply #49 on:** February 06, 2016, 01:29:39 am »

Quote from: HighVoltage on February 05, 2016, 08:13:11 am

Can you show us in the schematic where you have to touch to make the display stable?

It is literally anywhere near the front end, no specific point, I got busy with another job, so I haven't tried other places yet, have to get back to it soon though.

HighVoltage · « **Reply #50 on:** February 06, 2016, 09:36:52 am »

Quote from: TheSteve on February 05, 2016, 09:18:22 pm

The entire time I just kept hearing Shahriar in my head saying "Follow the signal path" lol.

Yes, I know what you mean, I learned so much from Shahriar' repair videos, he really is a master at this.

Quote

btw, the previous person that did the rework has some skill but not really enough to do a proper job. The soldering they did is far from excellent and they removed a part or two that there was clearly no need to remove as they can easily be verified 100% good in circuit. They also left the signature amateur calling card in the top of one of the relays in the form of soldering iron melt marks.

If that previous person did not have the skills and / or tools to do the job right, he may even have damaged some of the new components he installed. May be it is a good idea to check them all or replace them properly.

TheSteve · « **Reply #51 on:** February 06, 2016, 11:20:27 pm »

Cleaned up all of the soldering, its starting to look pretty good inside. I do need to order a replacement U44 next week. I've bypassed the chip and while it fails the self tests still it can count on both channels. If I give it the proper level input though on ch1 while booting it passes all self tests - so I'm hopeful a new U44 fixes it. I forgot to mention the other "surprise" I got the with unit. I am sure they just threw a power supply into the main board and then tossed a case on it. On the bottom of the case where it is riveted together you can see a hint of something black on the outside, on the inside it is a different story. It did not come off the 53132A PCB though as its very clean. I assume something leaked from a different board, sure is disgusting looking.

TheSteve · « **Reply #52 on:** February 07, 2016, 06:49:49 am »

Well I fired up the bread board and tested U44 - it looks and works exactly as an op-amp should so its fine. Back to the drawing board on this one...

edpalmer42 · « **Reply #53 on:** February 07, 2016, 07:29:21 am »

Quote from: TheSteve on February 06, 2016, 11:20:27 pm

Cleaned up all of the soldering, its starting to look pretty good inside. I do need to order a replacement U44 next week. I've bypassed the chip and while it fails the self tests still it can count on both channels. If I give it the proper level input though on ch1 while booting it passes all self tests - so I'm hopeful a new U44 fixes it. I forgot to mention the other "surprise" I got the with unit. I am sure they just threw a power supply into the main board and then tossed a case on it. On the bottom of the case where it is riveted together you can see a hint of something black on the outside, on the inside it is a different story. It did not come off the 53132A PCB though as its very clean. I assume something leaked from a different board, sure is disgusting looking.

Is that black stuff dry like blown capacitor guts or sticky/gooey? In some equipment HP used stick-on rubber feet as spacers to make sure that the board couldn't touch the case. Years later, they found that the rubber feet melted and turned into goo. If that's what you've got, it cleans up easily with IPA.

Zucca · « **Reply #54 on:** February 07, 2016, 07:36:27 am »

well, after reading this post here my choice:

Quote from: Nuno_pt on January 24, 2016, 03:17:20 pm

Ebmpapst 412
12Vdc
0,9W
75mA
6000rpm
10m3/h
18dB

expensive but I saved the shipping in the Digikey order...

I am pretty pleased with the result. When I fired it up the first time with the new fan installed, I thought I messed up something since I couldn't hear it. Only in a second step I looked at it and saw the fan spinning...

Just a head's up, in my PSU the ZD601:

had one leg flapping in the breeze. Also the corresponding solder pad below it did not show any sign of proper soldering. My guess is a production issue.

Nuno_pt · « **Reply #55 on:** February 07, 2016, 01:48:39 pm »

Hi zucca,

Here the price is 11.20€ + 23% VAT + Shipping so should be around 15€ from a local dealer.

I'll look for the ZD601 zener in my 53181A.

uncle_bob · « **Reply #56 on:** February 07, 2016, 01:58:19 pm »

Hi

One of the wonderful things about the 53181 is that most people hate them. As a result they saw no use in their "first life" and sell at a discount to the 53131 (or at least they should). I have one sitting here maybe two feet from me. The factory stock fan is completely silent. Yes, it's running. Yes I do check every so often. The display is nice and bright. There's no big pile of crud inside it from the fan.

They either are a great source of parts for the 53132 or a nice little counter all on their own.

Bob

TheSteve · « **Reply #57 on:** February 08, 2016, 02:24:25 am »

I am 99% the 53132A is repaired - or will be when a new AD7243AR 12 bit DAC arrives. The counter has one for each front end. Data goes through the CH1 DAC (U46) into the CH2 DAC(U47). I was able to see that CH2 was mostly alive so I removed both DAC's and installed the CH2 one into CH1. I still get failures at boot time of course(CH2 DAC is missing) but CH1 now counts perfectly and reports the correct input levels. Its not the cheapest part but is easy to get which is nice. I will order the DAC and a replacement fan next week and post an update once the new part is installed.

btw, based on date code the following IC's were previously replaced - U50, U51, U52, U53, U54, U55, U65 and possibly U44. The date codes on all original parts are 96/97, the replacement parts are 2011.

uncle_bob · « **Reply #58 on:** February 08, 2016, 05:58:54 pm »

Quote from: TheSteve on February 08, 2016, 02:24:25 am

btw, based on date code the following IC's were previously replaced - U50, U51, U52, U53, U54, U55, U65 and possibly U44. The date codes on all original parts are 96/97, the replacement parts are 2011.

Hi

Either somebody was very confused, or that counter took a *major* hit to it's front end back in 2011.

Bob

TheSteve · « **Reply #59 on:** February 08, 2016, 06:37:20 pm »

Quote from: uncle_bob on February 08, 2016, 05:58:54 pm

Quote from: TheSteve on February 08, 2016, 02:24:25 am

btw, based on date code the following IC's were previously replaced - U50, U51, U52, U53, U54, U55, U65 and possibly U44. The date codes on all original parts are 96/97, the replacement parts are 2011.

Hi

Either somebody was very confused, or that counter took a *major* hit to it's front end back in 2011.

Bob

I am quite sure it didn't take any kind of hit and I have finally found the real problem. I am thinking they just started swapping parts trying to fix it but never got to the proper part. If it had taken some kind of hit I think something closer to the input would be damaged. Will order the DAC today, should have it tomorrow.

edit - DAC on the way via overnight fedex.

TheSteve · « **Reply #60 on:** February 10, 2016, 06:51:06 am »

I am pleased to report another 53132A is once again 100% operational.

Here is the summary of the unit:

Bought from ebay as not working/missing parts. The pics were decent in the auction, the only thing actually missing was the back plastic frame and 3 screws. Made a low best offer, went back and forth a few times and settled at $165.00.

Received it a few days later - it was initially dead.

Discovered the hot and neutral wires were desoldered from the IEC power connector.
Looked over the supply, saw nothing wrong and decided to try soldering them back on, supply fired up just fine other then the fan which was dead.

Unit powered up with two failures - front end and measurement. Absolutely nothing could be measured initially.

Noticed lots of previous rework. U50, U51, U52, U53, U54, U55 and U56 had been replaced. U44 was possibly replaced as well.

Found a bad solder connection on the U50 +5 volt line. With this fixed the trigger lights now turned on but everything was still dead.

Removed U44 as its readings looked a little off to me. With the outputs of U44 jumpered to ground both channels would count properly with a 2 Vpp signal.
However the signal level readings were fixed at +- 5 volts on both channels.

Tested U44 out of circuit, it worked perfectly so I reinstalled it. I had the same power up errors but channel 2 seemed to actually work, it even gave the correct signal level. I assume U44 had a bad solder connection as well initially.

This lead to me to look at U39 and U40 which are the 12 bit DACs that feed voltage to U44. Upon power up a good DAC will toggle its output a few times during the self test. I was able to see U40 toggle properly but U39 was always a fixed -5 volts on its output. I removed both of them and soldered the U40 DAC in place of the U39 DAC. I didn't reinstall U40. At this point I received three errors at power on - front end, measurement and QSPI. However channel 1 now counted properly and indicated the proper signal levels. So a DAC was certainly bad.
Ordered a replacement U40 DAC and installed it. All self tests now pass and all measurements are bang on as they should be. I also installed a replacement fan.
Lastly I ordered a bumper and handle set from ebay and will be ordering a replacement rear plastic frame and the missing 3 screws from Keysight in a week or two when I finalize what parts are needed for two other ebay purchase projects.

Good times indeed, another very satisfying repair project.

Has anyone else ever seen a factory HP sticker on the front of a unit like this one? It has the standard serial # plate on the back so I am not sure what the front one indicates unless this was an HP factory asset at some point. For now I am leaving it on the unit.

Zucca · « **Reply #61 on:** February 10, 2016, 08:34:04 am »

Well done TheSteve thanks for sharing.
Interesting that sticker on the front... never saw that before.

HighVoltage · « **Reply #62 on:** February 10, 2016, 09:55:53 am »

Great Job and congratulations on the success.

You could order also a new front display cover, to get rid of the HP sticker and have it more modern looking with the Agilent branding.
I have never seen such sticker before.

uncle_bob · « **Reply #63 on:** February 10, 2016, 12:37:49 pm »

Quote from: TheSteve on February 10, 2016, 06:51:06 am

Has anyone else ever seen a factory HP sticker on the front of a unit like this one? It has the standard serial # plate on the back so I am not sure what the front one indicates unless this was an HP factory asset at some point. For now I am leaving it on the unit.

Hi

I suspect it lived in an HP cal center for some portion of it's life. They cycled the 53131's out of their system after they went EOL. This one may have gone to somebody else after that (and thus the odd repair). They have a lot of calibration sites around the world, so a lot of gear sitting on benches.

All of that *sounds* pretty cool. The gotcha is the name change. Most of these outfits come up with new asset tags as soon as there is a name change. Within a year or so after the change, you don't see the old tags going on gear anymore. It may have just been a "cool sticker" somebody came up with.

Bob

TheSteve · « **Reply #64 on:** February 10, 2016, 05:44:52 pm »

It does appear as it if was rack mounted most of its life - there is some thread locker in the mount holes. The sticker is the standard embossed aluminum one used by HP so I am sure it is a true HP sticker, what it means will likely always be a mystery. I could order a new front lense but I rather like original HP branded equipment.

bingo600 · « **Reply #65 on:** May 21, 2016, 09:10:06 am »

Quote from: sorenkir on January 21, 2016, 01:41:59 pm

Hi,

I have replaced the noisy fan of my HP53181A with the following: Sunon A40201V4 (Super Silence Series )
http://www.digikey.com/product-search/en?mpart=HA40201V4-000U-999&vendor=259

The fan is very quiet by itself (12.8dB-A), but when installed, due to airflow & high rotation speed, it is still noticeable in a quiet environment.

Michel.

Found this Fan with comparable specs on ebay.de
http://www.ebay.com/itm/HA40201V4-999-Axial-Lufter-40x40x20mm-12V-9m-h-von-Sunon-/111317257709

/Bingo

HighVoltage · « **Reply #66 on:** May 21, 2016, 09:48:04 am »

That fan should work well. If it is still too loud, just add a series resistor in the cable. The instrument does not need much cooling, just a small constant airflow.

Dwaine · « **Reply #67 on:** August 26, 2016, 02:03:18 pm »

I just bought a HP 53132a with the second channel BNC connector smashed. Does anyone know where I can find a replacement BNC connector? I looked on ebay and could not find anything.

Other than that, the unit is perfectly working. I also added the oven crystal to the unit and that is also working great.

TheSteve · « **Reply #68 on:** August 26, 2016, 10:46:28 pm »

Quote from: Dwaine on August 26, 2016, 02:03:18 pm

I just bought a HP 53132a with the second channel BNC connector smashed. Does anyone know where I can find a replacement BNC connector? I looked on ebay and could not find anything.

Other than that, the unit is perfectly working. I also added the oven crystal to the unit and that is also working great.

EDIT - I have found a better connector that is confirmed to work - TE Connectivity AMP Connector 5227677-1

To me it looks like an Amp 5413879-1 - or pretty close to that. You might want to consider just stealing one of the connectors from the back panel(whichever one you'll use the least).

Dwaine · « **Reply #69 on:** August 27, 2016, 12:36:13 am »

That's a good idea to use one from the back. I'll keep looking for a good match BNC connector. If I find one, then I'll post the details.

Dwaine

edpalmer42 · « **Reply #70 on:** August 27, 2016, 01:08:49 am »

How mangled is the connector? If you're careful, it's surprising how easy it is to restore a bent BNC.

Ed

TheSteve · « **Reply #71 on:** August 27, 2016, 03:24:09 am »

This one is an even closer match - http://www.digikey.ca/product-detail/en/te-connectivity-amp-connectors/5227677-1/A32268-ND/811166

HighVoltage · « **Reply #72 on:** August 27, 2016, 08:48:04 am »

Quote from: TheSteve on August 27, 2016, 03:24:09 am

This one is an even closer match - http://www.digikey.ca/product-detail/en/te-connectivity-amp-connectors/5227677-1/A32268-ND/811166

That one looks like it should fit well.
I ordered an original from Keysight a while back but that one was much more expensive

Dwaine · « **Reply #73 on:** August 27, 2016, 01:04:52 pm »

Thanks everyone. I ordered the connector from digikey.

Dwaine · « **Reply #74 on:** September 09, 2016, 06:12:03 pm »

The Digikey part number listed here is the exact fit. Replaced the BNC connected and channel 2 calibrated/passed. Working great.

Thanks everyone for the assistance. I appreciate it.

HighVoltage · « **Reply #75 on:** September 09, 2016, 06:20:19 pm »

Good to have the confirmation that it fitted and congratulations of the repair.

HighVoltage · « **Reply #76 on:** December 16, 2017, 11:20:15 am »

I got two more for repair
1 x 53132A
1 x 53131A

Both were free but they are in horrible condition.
One turns on with errors (But weak VFD), the other does not do anything
Missing windows, missing switch buttons
Broken and corroded BNC connectors.

Seems like a nice winter job.

On one of them the little rubber (?) distance foot is melted away to gunk.
May be the instrument was in a bad environment.

TheSteve · « **Reply #77 on:** December 16, 2017, 07:39:33 pm »

Quite the winter project indeed. One of my 53132A's had the rubber melt, it was a puddle on the lower chassis, no sign of it was left on the PCB itself. Looks like your unit was stored on its side and it ran down the PCB which is uglier and likely no fun to clean up at all. I have a spare display window but it is likely easier for you to get a few from Keysight directly along with whatever other parts you need.

texaspyro · « **Reply #78 on:** December 17, 2017, 02:00:33 am »

Quote from: HighVoltage on December 16, 2017, 11:20:15 am

On one of them the little rubber (?) distance foot is melted away to gunk.
May be the instrument was in a bad environment.

Nope, just polyurethane doing what it wants to do... revert back to its natural state of primordial goo. HP's (electrostatic printer for 9100 series calculators) and Tektronix's formulation for printer platen rollers (1503 TDR) are particularly good at it... (Hint... look at 3M cold shrink tubing for rebuilding those)

I had to clean and replace the snot glob in my 53132A and it lived a cushy indoor life.

HighVoltage · « **Reply #79 on:** December 23, 2017, 03:58:22 pm »

Ok, the first one is fully cleaned up and ready to be assembled again.
It was so much sticky, nasty dirt everywhere, it took several deep cycle bath tubs in pure alcohol, to take it all off.

Repair so far:
1. One power resistor exchanged in the power supply
2. All 5 BNC connectors replaced to new
3. New VFD installed

Seems to work again, the new VFD is nice a bright.

jadew · « **Reply #80 on:** December 23, 2017, 05:19:48 pm »

That looks great. I had the same experience with the little rubber foot, got on everything.

HighVoltage · « **Reply #81 on:** December 27, 2017, 01:44:54 pm »

Ok, the second broken counter is fixed.

Faillure:
- All LED light up but nothing showed up on the VFD.

Problem:
- No high voltage power to the VFD.
- 5V supply was OK
- 38V supply was bad

Broken parts:
- VR1 and VR2 both bad
- (DIODE-ZNR 6.2V 5pct PD=1.5W IR=5UA)

Solution:
- Installed 2 new zener diodes
- Installed new VFD

Repair done!

glarsson · « **Reply #82 on:** December 27, 2017, 04:16:38 pm »

That PCB is missing one XLINX so it's a 53131A, no a 53132A.
Never seen that extra PCB, with the three blue coax cables, in any 5313xA. What option is that?

texaspyro · « **Reply #83 on:** December 27, 2017, 04:41:23 pm »

Where are you getting the VFDs?

HighVoltage · « **Reply #84 on:** December 27, 2017, 05:04:34 pm »

Quote from: glarsson on December 27, 2017, 04:16:38 pm

Never seen that extra PCB, with the three blue coax cables, in any 5313xA. What option is that?

There are a few interesting things in this 53131A
I am not sure if this board with the blue cables is an option or was standard in the early counters.
The small blue cable is to take the 10MHz output from the Option 010 (High Stability) board to the main board output.

The little circuit board withe the 3 blue cables has a relay and can switch the 10 MHz output to two different locations on the main board:
Location 1: EXTERNAL 10 MHz OUT
Location 2: Ref. IN
May be it is for self calibration purpose? I am not sure and did not read up on it.

Interestingly this counter has the high stability crystal option 010 but there are no power supply cables going from the power supply to the Option 010 board, like usual (J2 on PCB not populated).
Instead it is powered by the ribbon cable.

HighVoltage · « **Reply #85 on:** December 27, 2017, 05:20:43 pm »

Quote from: texaspyro on December 27, 2017, 04:41:23 pm

Where are you getting the VFDs?

Here is a picture after calibration with the new VFD installed.
What a difference!
It is like new from a brightness and sharpness point of view.

After searching forever, I got very lucky to get two brand new VFDs but unfortunately the seller does not have any more.

glarsson · « **Reply #86 on:** December 27, 2017, 05:25:21 pm »

My Agilent 53132A with factory HS oven (option 010) does not have the extra power cable. The oven is powered through the flat cable only. The service manual mentions that the extra power cable isn't always used. Perhaps only when the oven requires it (depending on manufacturer/model number?).

HighVoltage · « **Reply #87 on:** December 27, 2017, 06:23:20 pm »

Quote from: glarsson on December 27, 2017, 05:25:21 pm

My Agilent 53132A with factory HS oven (option 010) does not have the extra power cable. The oven is powered through the flat cable only. The service manual mentions that the extra power cable isn't always used. Perhaps only when the oven requires it (depending on manufacturer/model number?).

Very interesting.
My two other 53131A both have the extra power cable from the power supply to the 010 option board.
Slightly different part numbers on the crystal oscillator:
10811-60160
10811-60161

glarsson · « **Reply #88 on:** December 27, 2017, 07:12:45 pm »

The HS oven in my 53132A is 10811-60160. The PCB it is mounted on does not have P2 mounted (the coax connector for the blue wire) but it does have J2 mounted (for the power supply) but plugged into it is a connector that only connects the two pins closest to P2 together.

glarsson · « **Reply #89 on:** December 27, 2017, 07:24:38 pm »

Image of P2 and J2.

glarsson · « **Reply #90 on:** December 27, 2017, 07:34:12 pm »

The main PCB of my 53132A does not have connectors mounted for the blue coax wires. Also, the tension converter for the display is not on a transformer based design on a daughter board, it is based on a capacitor/diode ladder and is placed on the main PCB. The melting rubber for is not a melted gray goo, it is black in pristine condition.

HighVoltage · « **Reply #91 on:** December 27, 2017, 08:14:00 pm »

Quote from: glarsson on December 27, 2017, 07:34:12 pm

The main PCB of my 53132A does not have connectors mounted for the blue coax wires. Also, the tension converter for the display is not on a transformer based design on a daughter board, it is based on a capacitor/diode ladder and is placed on the main PCB. The melting rubber for is not a melted gray goo, it is black in pristine condition.

Can you show a picture of this?
How old is this 53132A of yours?

All my 53131A and 53132A have the daughter board for making the VFD voltage.

glarsson · « **Reply #92 on:** December 27, 2017, 08:58:57 pm »

I don't know when it was manufactured as it was bought used. I got it cheap with handle and bumpers missing. Unexpectedly it arrived with option 010, squeaky clean inside and outside, intact rubber foot on PCB backside, bright VFD, etc. Could have been worse.

Bought NOS handle and bumpers on eBay...

The main PCB has a date code of week 30 2008 and the latest component date code I have found is week 24 2008.

One picture showing 38V converter. Need more pictures?

HighVoltage · « **Reply #93 on:** December 29, 2017, 04:34:30 pm »

Quote from: glarsson on December 27, 2017, 08:58:57 pm

I don't know when it was manufactured as it was bought used. I got it cheap with handle and bumpers missing. Unexpectedly it arrived with option 010, squeaky clean inside and outside, intact rubber foot on PCB backside, bright VFD, etc. Could have been worse. Bought NOS handle and bumpers on eBay...

The main PCB has a date code of week 30 2008 and the latest component date code I have found is week 24 2008.

One picture showing 38V converter. Need more pictures?

Very nice!
My counters are all older and have the extra board.
I will have a look out for a newer model.

HighVoltage · « **Reply #94 on:** December 29, 2017, 04:37:53 pm »

Yesterday I calibrated all my "old" counters against my HP GPS based 10 MHz reference and had them run over night to see the drift.

Here is the result in the picture.
Without analyzing the numbers, it appears that they are very stable after 24h.
They all have the option 010 installed.

HighVoltage · « **Reply #95 on:** December 29, 2017, 04:41:18 pm »

It seems that there are at least 2 different VFDs used in these counters:

One made by itron / Japan and one made by Samsung in South Korea
Both are interchangeable, but the metal frame around them has to be bent a little, to make the itron version fit, when the samsung was installed before.

Jf2014 · « **Reply #96 on:** December 31, 2017, 08:02:55 pm »

Quote from: texaspyro on December 27, 2017, 04:41:23 pm

Where are you getting the VFDs?

Probably in eBay

https://www.ebay.de/itm/Agilent-HP-Frequenz-Zahler-Display-Itron-CC1187G/222736101378

chinapp · « **Reply #97 on:** February 27, 2018, 09:23:51 am »

I encounted the same error: fail FRN /fail mesr ,but the falt part is not the U40 IC

HighVoltage · « **Reply #98 on:** February 27, 2018, 10:00:01 am »

Quote from: chinapp on February 27, 2018, 09:23:51 am

I encounted the same error: fail FRN /fail mesr ,but the falt part is not the U40 IC

FRN errors are not good!
Usually it is the front end that is dead.
You need to look at the the first amplifier stage, which is most likely dead.
But the service manual and schematics are available and you can find your way through and get it repaired.

analogRF · « **Reply #99 on:** May 07, 2018, 01:26:04 pm »

Quote from: HighVoltage on December 23, 2017, 03:58:22 pm

Ok, the first one is fully cleaned up and ready to be assembled again.
It was so much sticky, nasty dirt everywhere, it took several deep cycle bath tubs in pure alcohol, to take it all off.

Repair so far:
1. One power resistor exchanged in the power supply
2. All 5 BNC connectors replaced to new
3. New VFD installed

Seems to work again, the new VFD is nice a bright.

Hi, where did you get the new VFD? I need to replace mine as it is very dim with non-uniform intensity on various segments

HighVoltage · « **Reply #100 on:** May 07, 2018, 01:33:23 pm »

Quote from: analogRF on May 07, 2018, 01:26:04 pm

Hi, where did you get the new VFD? I need to replace mine as it is very dim with non-uniform intensity on various segments

I bought one on ebay Germany and that worked well.
When I wanted to buy more, the guy had sold them all.

Since these VFDs were the Japanese ITRON made version, I looked on the Japanese Yahoo auction site and found a few more and upgraded all my counters to new VFDs.

You need to look for Itron model number: CC1187G

analogRF · « **Reply #101 on:** May 07, 2018, 01:41:07 pm »

Hi
I have a 53132A which seems to be working OK. Passes the power on self test, also passes the ALL TEST self tests (in Utility menu: power + recall key) and passes calibration procedures.
However, when I choose specifically the FR END test in the utility menu, for which you have to connect the 10MHz out to ch1 and ch2 via a BNC-T, it always fails.
Can anyone with a working unit try this test and see what happens? My unit came with original option 010.

History: the power supply and the 38V VFD supply were dead which I fixed and now all voltages are well within specs and solid but I am not sure about their noise. Could that be the problem? Perhaps replacing the PSU caps would help?

HighVoltage · « **Reply #102 on:** May 07, 2018, 01:44:49 pm »

Actually there are some VFDs on ebay.com right now from China for $95 each
This is not the ITRON model but it should work well.

https://www.ebay.com/itm/112732008985

analogRF · « **Reply #103 on:** May 07, 2018, 01:49:56 pm »

Quote from: HighVoltage on May 07, 2018, 01:33:23 pm

Quote from: analogRF on May 07, 2018, 01:26:04 pm

Hi, where did you get the new VFD? I need to replace mine as it is very dim with non-uniform intensity on various segments

I bought one on ebay Germany and that worked well.
When I wanted to buy more, the guy had sold them all.

Since these VFDs were the Japanese ITRON made version, I looked on the Japanese Yahoo auction site and found a few more and upgraded all my counters to new VFDs.

You need to look for Itron model number: CC1187G

thanks! I will keep looking. Right now I could not find anything on japanese yahoo auctions.

I have seen those on ebay but I was hoping to find it cheaper. The seller offered $85 but the shipping to Canada is also $30
So $115 for display is a bit too much...

ArthurDent · « **Reply #104 on:** May 23, 2018, 06:54:50 pm »

I just bought an HP 53132A on Ebay that was listed as parts only with a blank display. I gambled and got it and it looked pretty good. It had the 3 Ghz card and the HS oscillator (010 option) but as shown on Ebay, the display was completely dark. I took the case off and visually inspected the inside and the only problem at first glance was the white bumper on the main board was completely melted with just some goo remaining.

Looking more closely I noticed one cap in the power supply had a bulging top so I replaced that 1000 uF/16V cap and the counter works perfectly. It took a while to find which REV I had so I could access the cal procedure but once I got that info and went through the cal menu, everything seems to be working as it should, even the display looks fairly bright. I had to relube the fan because it wouldn't always start on its own but after the oil worked its way onto the entire shaft it worked o.k. as well. If the fan was stopping that might have shortened the life of the faulty cap. I'll replace all of the power supply caps, just in case.

HighVoltage · « **Reply #105 on:** May 23, 2018, 09:24:41 pm »

Congratulations on an easy repair.
Yes, it is a good idea to exchange all electrolytic capacitors and measure the voltages and ripple.

Carefully remove the goo and install a new silicone bumper.
I would also replace the fan with a new silent type.

analogRF · « **Reply #106 on:** May 25, 2018, 03:29:51 pm »

Quote from: ArthurDent on May 23, 2018, 06:54:50 pm

I just bought an HP 53132A on Ebay that was listed as parts only with a blank display. I gambled and got it and it looked pretty good. It had the 3 Ghz card and the HS oscillator (010 option) but as shown on Ebay, the display was completely dark. I took the case off and visually inspected the inside and the only problem at first glance was the white bumper on the main board was completely melted with just some goo remaining.

Looking more closely I noticed one cap in the power supply had a bulging top so I replaced that 1000 uF/16V cap and the counter works perfectly. It took a while to find which REV I had so I could access the cal procedure but once I got that info and went through the cal menu, everything seems to be working as it should, even the display looks fairly bright. I had to relube the fan because it wouldn't always start on its own but after the oil worked its way onto the entire shaft it worked o.k. as well. If the fan was stopping that might have shortened the life of the faulty cap. I'll replace all of the power supply caps, just in case.

how do you get rid of that white glue sticking the caps together in the power supply and how do you re-apply that glue?
I know what it is used for but i don't know what is the material and where I can get it. It's used in many power supplies...

ArthurDent · « **Reply #107 on:** May 25, 2018, 04:58:51 pm »

When I removed the bad cap I found on the power supply in my HP 53132A, I scraped the ‘glue’ off the board then washed the area thoroughly with isopropyl alcohol to make sure that the electrolyte residue that leaked out of the cap was removed and wouldn’t cause future problems with the board. After repairing the board I make sure the board was good and dry before reapplying power.

The ‘glue’ on this power supply board was hard (or had hardened with heat/age) but some glue used on other pieces of equipment I’ve had was somewhat flexible like silicone rubber sealant that you could buy in any hardware store. I’ve seen Dave in some videos and others refer to this flexible glue as ‘snot’. I wouldn’t get too carried away with applying it because you still want some air flow around the heat sinks on the board.

HighVoltage · « **Reply #108 on:** December 21, 2018, 12:10:41 pm »

Keysight just released a new firmware 3.00 for all 532xx Series counters.
The Option 150 - Pulse Microwave Measurements is now a free upgrade
And the "about" screen is more clear about installed options.

I updated the FW of my 53230A counter from v2.05 to v3.00

TheSteve · « **Reply #109 on:** December 21, 2018, 04:00:07 pm »

So far I have found the html5 interface quite buggy.

SoundTech-LG · « **Reply #110 on:** March 28, 2019, 08:54:15 pm »

Got my hands on another 53230A... this one however, has CH1 dead, but the price was right, so need to figure out if the Front End got hit hard enough to fry something obvious. All is under that shield can I do believe, (lots of 5V relays as well). If anyone has ventured in there and repaired anything, it would be nice to hear of your adventure.
No schematics on these of course, (at least that we can have) so even more fun.

TheSteve · « **Reply #111 on:** March 28, 2019, 09:22:43 pm »

I made some offers on that one

If you're lucky it is a broken connection right where the BNC is mounted. If not, feed in a signal to both channels and see where it stops.

SoundTech-LG · « **Reply #112 on:** March 28, 2019, 09:27:53 pm »

Well MAYBE I got lucky... we'll see.

I just found the old post...

https://www.eevblog.com/forum/testgear/53230a-counter-ch1-input-trace-fails!/

HighVoltage · « **Reply #113 on:** March 29, 2019, 08:21:29 am »

Over the years I had two counters with a broken BNC solder connection / PCB crack at the BNC connector. Sometimes the BNC connector is not well mounted. If this is the case on yours, make sure to make the mechanical connection between the BNC connector and housing is well done and not wiggly.

SoundTech-LG · « **Reply #114 on:** March 29, 2019, 01:21:31 pm »

Hi,

Possible a bit of board flex caused it, but since it is purposely a very, very thin trace, it appears the intent was for the trace to pop before anything else, on high current, maybe. If you look closely at the OP photo, you can see the BNC shell is directly soldered onto the board, as is the signal connection. To replace that thin trace with a bus bar is probably not a great idea. I plan to use a very small piece of wire, if it turns out this is the failure. I would much rather re-visit the wire, than troubleshoot, and replace components in the Front End.

SoundTech-LG · « **Reply #115 on:** March 31, 2019, 07:55:41 pm »

I got VERY lucky. Broken fusible trace at CH1 BNC input...
Working again.

HighVoltage · « **Reply #116 on:** March 31, 2019, 09:52:18 pm »

Congratulations !

Now make sure the BNC connectors are good and stiff mounted in the housing.

SoundTech-LG · « **Reply #117 on:** April 01, 2019, 01:53:59 pm »

No "adjustments" to the BNCs are possible. They are soldered securely to the main board, pass through the tight fitting plastic front panel openings. That's it. It is what it is...

HighVoltage · « **Reply #118 on:** April 01, 2019, 02:35:39 pm »

On one of my counters, the BNC wiggled inside the housing. So, before inserting it, I added a very thin layer of shrink-tubing on the BNC to make it tight.

May be not all counters have this issue with space between BNC and housing.

SoundTech-LG · « **Reply #119 on:** April 01, 2019, 03:32:24 pm »

yes... the plastic is tight around the BNCs... no room for the thinnest heatshrink ever manufactured.

Which makes me conclude this was never a mechanical issue at all. Purely an overload condition that popped the fusible portion of the trace. If you look closely under magnification you can see the fusible spots along the input trace. Thank you HPAK.

richard.cs · « **Reply #120 on:** May 12, 2020, 09:31:03 pm »

Bumping an old thread... I figure as this morphed into a general repair thread over 3 years it can do no harm and I could use some advice from people who've had them apart.

I have a HP53132A that was a skip rescue, it was working perfectly, though I had not used it very much. This evening I go to turn it on and the front panel flashes and internal relays chatter continuously. I am thinking PSU problems causing it to power cycle when the load is increased (the rear fan runs fine in standby, doesn't slow noticeably when I try to turn it on but it's rather hard to tell with all the chattering). Sound plausible?

I see there's not a great deal of detail in the service manual, it being geared more towards replacing modules, though there is at least a set of testpoints and target voltages. I don't suppose anyone found further details than that? I've seen schematic snippets in this thread, some reverse engineered and some looking like they come from HP/Agilent documentation. Of course I may yet find an obvious failure when I crack her open, but it's 22:30 here so that's not going to happen today.

I suspect it may well have had some rough treatment before I owned it, but it's been in (very) occasional use for a couple of years without problems.

HighVoltage · « **Reply #121 on:** May 13, 2020, 10:01:53 am »

Did you test all power supply output and especially for any ripple on the dc rails?

richard.cs · « **Reply #122 on:** May 13, 2020, 10:13:55 am »

Quote from: HighVoltage on May 13, 2020, 10:01:53 am

Did you test all power supply output and especially for any ripple on the dc rails?

First thing on my list to do, hopefully I'll get a chance today. Is there a target ripple specification, the service manual only gives DC tolerances.

richard.cs · « **Reply #123 on:** May 13, 2020, 06:48:49 pm »

It's very clearly a PSU fault, the supply voltages are low in standby, and fall further and become noisier when I turn the instrument on.

Supply	Standby	On
5V	3.5	O2.4
+12V	10.6	10.1
-12V	-11.9	-7.3
-9V	-8.9	-8.9

Overall it's quite dusty inside and in need of a good clean, but there is no visible damage (no domed caps or visible heat damage apart from perhaps a slight PCB yellowing near Q1).

The -12 and 5V seem to be the worst offenders, with +12V being out-of-spec but not dropping much at switch on. Looking at the schematic Jf2014 posted, likely candidates would seem to be the primary current sense (R5) failing high, problems with the aux supply (D2, R4, etc.) causing cycling when more gate drive power is demanded, feedback problems, etc. I am not sure I quite understand the whole circuit* but there are some interesting features in there like making the current sense dependent on primary voltage. I'm not quite sure what IC3 is about either - cleaner shutdown when power is removed?

*What's that complex arrangement on the bottom left? I know it's one of the controllers where V(comp) sets Ipeak, but the role of Q22, Q23, Q24 does not seem clear to me.

HighVoltage · « **Reply #124 on:** May 14, 2020, 09:57:33 am »

That is a very good start for the repair.
Most likely one or more components are pulling too much current.

Just follow the current path and remove a possible suspect and see if the voltage jumps back to its normal values.

Unless the problem is within the power supply rails, check that first.

You should find the culprit easily.

richard.cs · « **Reply #125 on:** May 14, 2020, 10:13:43 am »

Ah, so you think it may be an overload problem on the counter rather than a fault with the PSU itself. Not something I had considered but quite plausible. Hmmm, lots of tantalum capacitors on the counter that I am now suspicious of.

I think my next step will be to separate the PSU and the remainder of the board so I can test them independently. Probably need to do that anyway for the cleaning.

richard.cs · « **Reply #126 on:** May 17, 2020, 10:20:31 am »

PSU board removed and cleaned. Tested in isolation with resistive loads on 5V and 12V - still faulty so not overcurrent from the rest of the counter. Scoping the outputs shows full-amplitude ripple on the 5V output and about 6V of ripple on 12V output. The PSU is appears to be starting up and running for around 15 ms, every 60 ms (the 60 ms being mains-synchronous). This could be controller Vcc or perhaps the tripping of a protection circuit.

Probing around shows the controller Vcc swinging between 16 V and 10 V with that same 60 ms period. These voltages are completely consistent with the controller datasheet values. I infer from this that the fault lies either with the aux winding and its rectifier, or with the IC3/Q25 circuit (JF2014 component IDs).

As I side note I have observed that something on the PSU board is capable of holding (a small amount of) charge for several days. I am not sure what I touched - probably a line-earth EMC capacitor as everything else has a bleed path.

richard.cs · « **Reply #127 on:** May 17, 2020, 10:53:30 am »

And the answer is... R26B is open circuit. R26A is a little high but within tolerance. It looks to be a 0.5 or 1 W part, though worst case dissipation should be under 100 mW, probably for voltage rating. As I don't stock those I have temporarily replaced it with two 1/4 W components in series and the supply now operates with the outputs in specification.

@ Jf2014 Thank you very much for the reverse-engineered schematic.

Perhaps before I put it back together for good I'll do as others on here have and source a quieter fan.

HighVoltage · « **Reply #128 on:** May 17, 2020, 01:12:11 pm »

Nice, that you fond the culprit, congratulations.

Agilent power supplies have a problem from time to time on many of their instruments and most of the time its is a single component failure. May be a good idea to install a little higher wattage resistor.

An a better fan is always a good idea.

pizzigri · « **Reply #129 on:** August 01, 2021, 10:45:24 am »

So I am working on the 53131a ps as well, and got curious while I am waiting for some components. I decided to post here since it is more related to this general post than my specific repair problem.
The point is, I have noticed the PS has a -9 / -9VG output, and I cannot understand what is that for. Actually, I even found a video on YT in which I think Gerry Sweeney experiments with leaving this line disconnected and the counter boots and works fine; and this, to eventually build a linear PS for the counter. He never followed up in this project, but I was still very curious to learn what does the 9v Power is needed for..... anyone knows?

Plus, since the Delta PS is so unreliable, maybe it could be a good idea to make a linear ps for it. Replacement Delta PS go for 250-350 Euro which together with the VFD going prices, is totally crazy...

HighVoltage · « **Reply #130 on:** August 01, 2021, 06:23:48 pm »

Quote from: pizzigri on August 01, 2021, 10:45:24 am

Plus, since the Delta PS is so unreliable, maybe it could be a good idea to make a linear ps for it. Replacement Delta PS go for 250-350 Euro which together with the VFD going prices, is totally crazy...

This was discussed before in a different thread also in a Keysight scope thread since the PSUs in some Keysight gear are not very reliable.
The positive aspect of this is that we can find broken test gear for low cost and repair it for fun.

f14 · « **Reply #131 on:** April 02, 2022, 10:14:31 am »

sorry for bump old topic.
I have two error QSPI fail and INTERP fail. it suddently happens. My 53131a is "newer" version ( diffirent A1 board with agilent service doc)
Have any idea with this fail ?. power supply is good. voltage output within specs

HighVoltage · « **Reply #132 on:** April 02, 2022, 11:01:10 am »

The service manual gives these indications:

QSPI: FAIL
A1 Motherboard Assembly, or A2 Display Board Assembly

INTERP: FAIL
A1 Motherboard Assembly, or A4 AC Power Supply Assembly

So, most likely something is wrong in the A2 motherboard assembly

What are the symptoms of the instrument?
Do these fails show up doing starting of the instrument?

f14 · « **Reply #133 on:** April 02, 2022, 11:07:45 am »

Quote from: HighVoltage on April 02, 2022, 11:01:10 am

The service manual gives these indications:

QSPI: FAIL
A1 Motherboard Assembly, or A2 Display Board Assembly

INTERP: FAIL
A1 Motherboard Assembly, or A4 AC Power Supply Assembly

So, most likely something is wrong in the A2 motherboard assembly

What are the symptoms of the instrument?
Do these fails show up doing starting of the instrument?

it suddently appear when power on self test.
I have read service doc from agilent but Im not clear about they explain this errors
now my 53131a like a brick it appear INTERP:FAIL when hit RUN or enter
I have test A1 mainboard. and I got something strange for me . C60 C61 cap. votage not stable (changing ~2v <->5v) I remove both caps and check the voltage it stable at 5v but the caps is fine (I check with LCR)
thanks for your anwers

HighVoltage · « **Reply #134 on:** April 02, 2022, 12:13:46 pm »

Check all DC voltages and rails for AC ripple and stability.

f14 · « **Reply #135 on:** April 02, 2022, 12:49:44 pm »

Im checked . all rail stable and AC ripple ~<300mV for 4 rails 5V/12V/-12V and -9V. I measured on power input connector (A1 mainboard)
I just remove some E-cap from A4 power suppy to check value with LCR meter and it still within specs

HighVoltage · « **Reply #136 on:** April 02, 2022, 02:52:08 pm »

If I look at an older manual, it says:

INTERP FAIL
Interpolator calibration failed on the last measurement; therefore, no valid measurement was taken.

So, I would start at the signal input and work my way through signal path until you find something that makes no sense. Sometimes this can be tedious.

f14 · « **Reply #137 on:** April 03, 2022, 10:08:40 am »

Quote from: HighVoltage on April 02, 2022, 02:52:08 pm

If I look at an older manual, it says:

INTERP FAIL
Interpolator calibration failed on the last measurement; therefore, no valid measurement was taken.

So, I would start at the signal input and work my way through signal path until you find something that makes no sense. Sometimes this can be tedious.

I use a Printer ON via RS232 to see more details about this error and I have 4 line:
-Interpolator : A , MRC mode Fail
-Interpolator : B , MRC mode Fail
-Interpolator : B , CC mode Fail
-QSPI : HEX DAC Fail
Because I have only one 53131a so I can't check a good 53131a signal to compare

HighVoltage · « **Reply #138 on:** April 03, 2022, 11:27:38 am »

It is not easy to give advise at this stage.
All I know from experience is that you start looking at signals and suddenly you come to find one culprit.

Maybe you can borrow the same counter and compare signals.


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