Author Topic: HP 54502A - nearly vintage digitizing scope  (Read 37348 times)

0 Members and 1 Guest are viewing this topic.

Offline MetraCollectorTopic starter

  • Regular Contributor
  • *
  • Posts: 97
  • Country: cz
HP 54502A - nearly vintage digitizing scope
« on: August 06, 2012, 12:53:08 pm »
Greetings gentlemen !

I would like to introduce you this Hewlett Packard 54502A digitizing oscilloscope in this short teardown.
I bought it for very reasonable price, because seller told me that it show some errors and I hoped it can be repaired.
So let's have a look.

Device after turning on and ran self test:


My friend told me this is very bad. I agree with him :(


This machine was a virgin  ;D because of his sticker "warranty void if removed", so I was likely the first one who opened it after it's last calibration in January 1998 (cal sticker with date).
Dirty and nasty, the same smell when you open an old TV, black dust on HV accessories.




Lot of dust:






Power supply. This looks primitive for me compared to evilness of the mainboard. ;D


I cleaned HV supplies:


Everything clean now. Very easy assembling and dissasembling. However I recommend to seal all screws and write on it where they belong to :)


These orange Nippons are perfect, they always hold their capacity, ESR and tgD even in rough conditions.


I noticed that Packards alway use golden plating on their keyboard accessories :)


Computer. M68000. Where can I get this in gold ceramic ?


Should I replace this Dallas IC ?


Preamps and attenuators


A/D hybrids


More hybrids and ... mmm, thermal oscillator :)


I was talking about evilness of the mainboard. Do you agree with me ? I can count 8 layers in close up of the laminate.
Anyway, it is beautiful, I like these clear laminate that I can read traces in inner layers :)


More hybrids:


I didn't take a photo, but this was very dirty, balls of dust:





I hope you like it, feel free to ask and require more photos.
Now I am going to check power supplies and find out what's wrong with this machine.
Do you have any experience with this HP digiscopes series ?

All of photos here:
https://picasaweb.google.com/100683312221737332313/HP54502A

Regards, Ondrej
« Last Edit: August 06, 2012, 12:58:18 pm by MetraCollector »
 

Offline firewalker

  • Super Contributor
  • ***
  • Posts: 2450
  • Country: gr
Re: HP 54502A - nearly vintage digitizing scope
« Reply #1 on: August 06, 2012, 01:10:25 pm »
Clean everything (contacts etc), reassemble it and test it again. Some times taking apart something will do the trick!  :) :) :)

Nice piece of equipment btw!

Alexander.
Become a realist, stay a dreamer.

 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: HP 54502A - nearly vintage digitizing scope
« Reply #2 on: August 06, 2012, 01:15:56 pm »
lovely bit of gear, i actually have one of a very similar vintage using a 68000K aswell,
i must say i am annoyed to discover those HV wires unclip, i almost trashed my ones HV supply desoldering its HV transformer,
 

Offline anotherlin

  • Regular Contributor
  • *
  • Posts: 244
  • Country: fr
Re: HP 54502A - nearly vintage digitizing scope
« Reply #3 on: August 06, 2012, 02:15:33 pm »
Any idea of the date of introduction of this beauty ?
"Lots of people have made $100K or more mistakes and didn't get the boot. It's called training, why fire them after such an expensive lesson?" -- EEVblog Electronics Community Forum
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 6680
  • Country: gb
  • Electronics Hobbyist & FPGA/Embedded Systems EE
Re: HP 54502A - nearly vintage digitizing scope
« Reply #4 on: August 06, 2012, 02:28:29 pm »
Had a HP 54501A, made in 1993.

Introduced 1991.

Fault is with the DALLAS IC holding the calibration factors. It must be replaced and a re-cal should be done, though it may be okay without a cal (just not super accurate.)
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: HP 54502A - nearly vintage digitizing scope
« Reply #5 on: August 06, 2012, 03:51:01 pm »
Change the dallas chip and do a self cal. Most likely the machine will work again.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline aghp

  • Regular Contributor
  • *
  • !
  • Posts: 196
  • Country: fi
Re: HP 54502A - nearly vintage digitizing scope
« Reply #6 on: August 06, 2012, 04:13:24 pm »
"Now I am going to check power supplies and find out what's wrong with this machine."

As can see in self test there stay NVRAM FAIL.

Just change it or modify with new compatible NVRAM. (exactly same you can not find maybe)
Or modify it using Gold contact socket + NVRAM socket + low standby current CMOS Ram. (this whole sandwitch is still just enough low that it can go inside through bakcside hole so these can assemble all ready outside. (if need many times take main board out then no need take NVRAM/CMOS out from socket. Advantage in modification is that you can then measure lithium voltage.

I have done it for several HP5450x oscilloscopes (years ago as these was popular here on second hand markets).

Also restored these input attenuators in some cases.

Also  common problem is sometimes power supply. But typically only after extremely long power up hours.
EU: Owon oscilloscopes and in Finland also  some Siglent selected equipments. All with our  lab deep Q.C. in Finland.  Surplus test equipments. Repair service. 40 year experience about electric and electronic repair, service and design. Local Owon SDS-Series  repair service for our customers.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16272
  • Country: za
Re: HP 54502A - nearly vintage digitizing scope
« Reply #7 on: August 06, 2012, 05:53:28 pm »
Those can be "modified" to place a new battery on top. There is a thread on that somewhere in the forum.
 

Offline MetraCollectorTopic starter

  • Regular Contributor
  • *
  • Posts: 97
  • Country: cz
Re: HP 54502A - nearly vintage digitizing scope
« Reply #8 on: August 06, 2012, 08:36:53 pm »
Hello gentlemen again.

I am very pleased to seeing your responses, thank you :)

I read the service manual and ran self-test again to look what's wrong and I was a little bit confused about it, because I tried to load default factory calibration, and it gives me this errors:





This is the error that indicates corrupted Dallas IC you are talking about ?
I am confused because I just thought that this IC can operate without backup battery, that it just loss it's data after power off.
For example, my M1T380 can be calibrated without backup battery, but it loss it's calibration after it is powered off.

Of course, I would follow your advices, here are some information about these Dallas ICs:
https://www.eevblog.com/forum/product-reviews-photos-and-discussion/defective-hp-54504a-nvram-substitute/msg92282/#msg92282
http://www.mcamafia.de/mcapage0/dsrework.htm
http://classic-computers.org.nz/blog/2009-10-10-renovating-a-dallas-battery-chip.htm
http://www.radiocollection.be/uk/HP54503_uk.html

I consider a new DS1230Y on eBay:
http://www.ebay.com/itm/1PCS-DALLAS-DS1230Y-100-DS1230Y100-1230Y100-IC-NEW-/250912608886?pt=LH_DefaultDomain_0&hash=item3a6b8e9276
Weren't these IC's counterfeited by asian copiers or it is OK to buy it even from China ?

Unluckily, my DS1235Y is soldered into the mainboard. I wonder if it is a good idea just drill to the package to disconnect old battery element and solder a new Li-battery and attach it to the package ?
I do not want to destroy the PCB, even I have an access to profesional equipment in my work. (BTW, I work for ON Semiconductor as a student :) )

What do you suggest to me ?
Thank you and regards Ondrej
 

Offline aghp

  • Regular Contributor
  • *
  • !
  • Posts: 196
  • Country: fi
Re: HP 54502A - nearly vintage digitizing scope
« Reply #9 on: August 06, 2012, 11:15:34 pm »
Turn cal protect switch non protect and do default. What reads then.
Then if all read D just do full selfcal.
(but also in your first image, there is test table where reads:
Failed: Protected Non Volatile RAM.

But in your las image there read that *C in every place = Corrupted.
But then top of screen there read that switch is protected. Turn it unprot and then load factory defaults. There need then read D in every place where now is C.
EU: Owon oscilloscopes and in Finland also  some Siglent selected equipments. All with our  lab deep Q.C. in Finland.  Surplus test equipments. Repair service. 40 year experience about electric and electronic repair, service and design. Local Owon SDS-Series  repair service for our customers.
 

Offline MetraCollectorTopic starter

  • Regular Contributor
  • *
  • Posts: 97
  • Country: cz
Re: HP 54502A - nearly vintage digitizing scope
« Reply #10 on: August 07, 2012, 03:17:55 pm »
Hello

You all were right, I switched calibration to unprotected and then oscilloscope automatically loaded default calibration, as you can see here:


Selftest also passes without any problems, relays click quite loud :)


Are these reports all-right ?
I am confined about drilling holes into Dallas IC to connect new battery and attach it to the package. Can you someone tell where are connections located ? I have some training DS12887 from old motherboard and I found it by drilling, following instructions from the WWW I wrote above. But this DS1235Y has different pinout, I expect different layout.

Anyway, oscillscope seems to work, offsets are little bit off, but I am very pleased to have almost functional HP :)
I tried some simple Wien's oscillator to test it:


Thank you and regards, Ondrej
 

Offline Wim_L

  • Regular Contributor
  • *
  • Posts: 212
  • Country: be
Re: HP 54502A - nearly vintage digitizing scope
« Reply #11 on: August 07, 2012, 04:01:41 pm »
Had the same problem with a 54504A:
https://www.eevblog.com/forum/product-reviews-photos-and-discussion/defective-hp-54504a-nvram-substitute/

I tested it, and it would selfcal perfectly if started from a default setup, and everything worked after that. Of course it lost calibration as soon as it was powered down, but it'd work fine as long as you left it running. If those match your symptoms, it definitely is the Dallas chip. You don't need the exact same model, http://www.maxim-ic.com/app-notes/index.mvp/id/202#q13 (and see the previous thread for more discussion).

A note on removing it: the legs of that chip are covered completely by the package, so cutting the chip out and then desoldering pin by pin won't work. I ended up doing it with hot air, but the heating slightly discoloured the board and made a nearby cap drop out of the board too. Not too bad, no serious damage, but if you have access to a good vacuum desoldering tool you can probably do better.

Mine also had a bit of error on the low frequency adjustment. While you're replacing it anyway, grab yourself a copy of the service manual (search for the model number on the Agilent site, they have a scanned PDF version online) and follow the whole procedure for tests and calibration. Some parts are a bit tricky because you'll have to adjust pots under the power supply, some of those cheap plastic adjustment screwdrivers might come in handy (after cutting them in half, if they're the long type).
 

Offline MetraCollectorTopic starter

  • Regular Contributor
  • *
  • Posts: 97
  • Country: cz
Re: HP 54502A - nearly vintage digitizing scope
« Reply #12 on: August 18, 2012, 09:11:03 pm »
VOILA  :D

Hardly believe I managed to do it with only chinese sucker and pro's kit solder, but it went much more better then expected :)
It went absolutely with no resistance, almost every pin on the first try, pins were free and when sucked the last pin, I only deattached the IC with no force :)



DS1230Y substituent is on it's way :)
Unluckily have ho socket at home right now, but it is a question of one hour to local store :)
« Last Edit: August 18, 2012, 09:12:39 pm by MetraCollector »
 

Offline ice595

  • Regular Contributor
  • *
  • Posts: 98
  • Country: ca
Re: HP 54502A - nearly vintage digitizing scope
« Reply #13 on: November 19, 2013, 05:08:35 am »
If I buy a working HP 54502A oscilloscope from ebay that passes all self-tests, should I be worried about the DALLAS IC failing in the next few years?
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: HP 54502A - nearly vintage digitizing scope
« Reply #14 on: November 19, 2013, 09:35:48 am »
yes, as the battery in the chip will be almost 2 decades old and well past its used by date
 

Offline ice595

  • Regular Contributor
  • *
  • Posts: 98
  • Country: ca
Re: HP 54502A - nearly vintage digitizing scope
« Reply #15 on: November 19, 2013, 02:38:48 pm »
so when that day comes, do I can just replace it with a DS1230Y and do a self-cal and its all good?
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1380
Re: HP 54502A - nearly vintage digitizing scope
« Reply #16 on: November 20, 2013, 01:26:40 pm »
yes, as the battery in the chip will be almost 2 decades old and well past its used by date

Are they just SRAMs?  since the battery is internal, could it be pulled while its still alive and read out to a PC to preserve the data while a replacement it en-route?
 


Offline ice595

  • Regular Contributor
  • *
  • Posts: 98
  • Country: ca
Re: HP 54502A - nearly vintage digitizing scope
« Reply #18 on: January 31, 2014, 01:28:55 am »
I bought 2 HP 54502A scopes off ebay from the same seller last month
1 passed all self tests and the other 1 with failed protected nvram, DAC, ADC and analog trigger
I paid $200 for the working 1 and $50 for the broken 1

Then I also ordered some DS1235YW, after swapping the dead NVRAM
I did a firmware self-calibration on it and it worked like a champ!
 

Offline scientist

  • Frequent Contributor
  • **
  • !
  • Posts: 317
  • Country: 00
  • User banned.
Re: HP 54502A - nearly vintage digitizing scope
« Reply #19 on: January 31, 2014, 01:31:04 am »
Vintage what? 1990?

Also, why are my posts being "watched"? What am I going to do, illegally help a beginner?
« Last Edit: January 31, 2014, 03:30:09 am by scientist »
 

Offline ice595

  • Regular Contributor
  • *
  • Posts: 98
  • Country: ca
Re: HP 54502A - nearly vintage digitizing scope
« Reply #20 on: January 31, 2014, 03:24:05 am »
Vintage what? 1990?

The service manual PDF that I found was last revised in January 1991, so I guess its about late 1980s perhaps 1990?

Silly HP decided to use IC sockets for systems ROMs and other stuff but not the NVRAM  :palm:
This unit cost well over $7000 USD at their time so its not exactly "disposable" electronics.
Its pretty obvious that the NVRAM will fail sooner or later regardless of power up hours (just like a ticking death clock)
What were they thinking?  :-//

ordering information on PDF in the link down below
http://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=10&cad=rja&sqi=2&ved=0CGgQFjAJ&url=http%3A%2F%2Fwww.mrtestequipment.com%2Fgetfile.php%3Fs%3DAgilent%2B54502A%2B400%2BMHz%2B2%2BChannel%2BOscilloscope%2BData%2BSheet.pdf&ei=HBXrUuqDEqmuyQGDtoHQCQ&usg=AFQjCNH6aFpnNs7KEvkunYHS-xA58Xxsyw&sig2=RhA6GyLyweBBHADD442bLA


« Last Edit: January 31, 2014, 03:29:10 am by ice595 »
 

Offline thetundrawolf

  • Newbie
  • Posts: 2
Re: HP 54502A - nearly vintage digitizing scope
« Reply #21 on: July 16, 2014, 08:54:09 am »
Those can be "modified" to place a new battery on top. There is a thread on that somewhere in the forum.

Any idea where it is? I have this same oscilloscope. I want to repair it with a piggyback.
 

Offline thetundrawolf

  • Newbie
  • Posts: 2
Re: HP 54502A - nearly vintage digitizing scope
« Reply #22 on: July 17, 2014, 08:19:16 pm »
I found these resources.

http://www.hardwaresecrets.com/article/Replacing-the-Motherboard-Battery/81/3

This recommends replacing the battery inside the NVRAM module. It even has a pinout. At the worst case, you could solder a battery to the pins, observing polarity. This leaves the old battery inside, to possibly leak and corrode.

He is talking about removing the top cover and replacing the battery.

Here is another option, and an 'able.

http://www.instructables.com/id/HP-54501A-NVRAM-Replacement/

This guy not only replaced the module itself, he soldered in a socket so he could replace it later if he wanted. He has an HP 54501A.

I personally prefer to remove the cover and replace the battery. The best option, of course, is to solder in the socket so you can replace it any time you wish without desoldering.

Here is a search I did for replacing the battery in the Dallas NVRAM module. Unfortunately, it looks a bit more complicated than just popping the cover off.

https://www.google.com/search?q=replacing+dallas+nvram+battery&client=firefox-a&hs=qcE&rls=org.mozilla:en-US:official&channel=fflb&source=lnms&tbm=isch&sa=X&ei=YTDIU9ebNYzfoATC24DICQ&ved=0CAkQ_AUoAg&biw=1366&bih=631#imgdii=_

I am not sure what to think. The replacement module is a little over $30 which is worth it to keep my digitizing scope running good.

Apparently the Dallas modules have the batteries encased inside. This is what you must do to get at it.

http://www.classic-computers.org.nz/blog/2009-10-10-renovating-a-dallas-battery-chip.htm
« Last Edit: July 19, 2014, 06:15:17 am by thetundrawolf »
 

Offline jjbnks

  • Newbie
  • Posts: 1
Re: HP 54502A - nearly vintage digitizing scope
« Reply #23 on: August 09, 2014, 12:18:41 pm »
Greetings gentlemen

I have this scope for nearly a decade, now, and just ran into problems with it yesterday.
First, it seems to have a rather big offset of 46 V...

So, using "Util" menu, I ran a selftest, and got Analog trigger and D/A converter failed.

Well, this would have been manageable, but...

From this morning, I can't keep the screen on any more than a few minutes... and then it fades out quickly and I have a screen without any spot on it.

Do you have any idea about this ?
This rather looks like HV cathodic screen fails ; is it repairable ? How can I test it ?

Any idea is welcome.

BR,

José
 

Offline wa4rts

  • Newbie
  • Posts: 1
Re: HP 54502A - nearly vintage digitizing scope
« Reply #24 on: December 18, 2014, 08:08:35 pm »
Took me a long time to find this post, but it was of great help in fixing my 504503A.  The Dallas part was indeed bad, and once replaced, it calibrated just fine and is running right now.  The Dallas chips can be found in China and Hong Kong, reasonably priced, but takes about two weeks for them to arrive.  I recommend socketing the part once you have desoldered it, as it will die again (hopefully not soon...) and have to be replaced. Surprisingly, I had sockets.

Thanks for all the information and photos!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf