Author Topic: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)  (Read 59717 times)

0 Members and 1 Guest are viewing this topic.

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
I have successfully swapped modules between 8502A, 8505A, and 8506A.   Of course, just because the modules "work" doesn't mean the parts can't be made to different tolerances etc. -  I haven't tested them that thoroughly (yet!... now that I had the idea...)  -  but some cards, like the GPIB card etc., should be a no-brainer.

The transformers hum loudly in every single unit I've ever laid hands on - it is safe to say this is a "feature", not a bug...   I guess it might be fixable by rubber mounting it - I have a theory that it is perhaps  too stiffly coupled with the chassis, mechanically? - but the hum is hardly the worst noise offender in the old test gear collection...  :D

Yes, 8505/06 I'd expect, and I hadn't really considered GPIB or isolator, but with the 8500A I can tell you the ohms, current and AC modules don't swap and I think it seem obvious that the controller and A2D won't either.  Not sure on the DC filter or active filter parts.

I tried tightening the bolts on the transformer and had it out on a bench and it doesn't seem to be a mounting issue, just something moving inside.  I had this idea of vacuum-potting it with thin epoxy, but I have better things to do--as if that has ever mattered before!  The 8500A has a different type of transformer and it is completely quiet, but it just won't work.

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
For DCV (where the bias spec I copied came from),  it routes to the DC signal conditioner, then to the filter and finally to the ADC.   Not sure what peaks you are expecting to see.   

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
For DCV (where the bias spec I copied came from),  it routes to the DC signal conditioner, then to the filter and finally to the ADC.   Not sure what peaks you are expecting to see.

We can't rule out that the CPU switches the signal conditioner in and out of the high quality bus for each measurement, which could affect things at the margins?

Time to take a look...

 

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us
Interesting thread!

I've got one of the late model 8506As manufactured around 1992. My favorite feature is the 200% overrange on the 10V range and below as well as the permanently stored zero corrections for each range. My unit has the laser trimmed glass/ceramic resistor package in the DC Signal Conditioner module. So I'm curious how well mine performs now that everyone is sharing data from their units. If anyone is curious, I can take photos of the DC Signal Conditioner or other boards. My unit has the Current Shunts installed, rather than Ohms.

What software are people using to collect and process the data? I have MATLAB and could write my own code to collect and plot the data, but if there is a turn-key solution I wouldn't mind using that to start with.

The low input bias current, at least for the late model units is from using an Analog Devices OP97 (LT1012 equivalent) to buffer the input. The BOM doesn't list the part suffix for the OP97 as it was a selected part for <5pA Ib. I'm not sure if there is a better OPAMP available. The chopper stabilized OPAMP is an LT1052CN8, which looks like it could be upgraded to something a little more modern, possibly an OPA189?

I just finished replacing all the electrolytics in the power supply. Almost all of the tiny 5mm caps were dead. The larger diameter parts surprisingly looked okay. The beefy 15000uF snap-in part was a pain to unsolder and I ended up pulling a trace, but some Kapton tape and soldering the pad back to the plated through-hole tidied that up. When installing the large snap-in style replacement caps I had to bend the leads so they would fit into the holes, otherwise the repair went smoothly.

Fluke 850xA PSU Module Radial Aluminum Electrolytic Capacitors

Qt C       WVDC DxHxP       Part
1x 15000uF  16V 30x42x10mm  (C8)
1x  3300uF  16V 22x26x10mm  (C5)
4x   470uF  50V 16x26x8mm   (C3,4,6,7)
2x   330uF 100V 16x26x8mm   (C1,2)
4x   5.6uF  35V 5x11x2.5mm  (C9,10,11,17)
3x   2.2uF  50V 5x11x2.5mm  (C18,19,20)


Notes: D=diameter, H=height, P=lead pitch
« Last Edit: February 02, 2021, 05:23:40 am by garrettm »
 
The following users thanked this post: guenthert

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00

That looks very interesting!

What manual are you finding that diagram in, I don't seem to be able to find that diagram in any of the ones I have?  (One found online, another purchased from Artek).

I guess it's time to open up the DC conditioners and see what revision level they are...   do you happen to know the rev letter on yours?
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
For DCV (where the bias spec I copied came from),  it routes to the DC signal conditioner, then to the filter and finally to the ADC.   Not sure what peaks you are expecting to see.

We can't rule out that the CPU switches the signal conditioner in and out of the high quality bus for each measurement, which could affect things at the margins?

Time to take a look...

1) there's no reason it would
2) I would hear it since these units use relays
3) My 8846A would notice, especially on high-impedance sources like the 1V range of the 731B.  It picks right up on other meters that have charge pulses due to autozero and I can plot a nice graph of it happening.  In this case, there's no such noise.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 
The following users thanked this post: SilverSolder

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00

A crude scope shot of the inputs doesn't show any immediate signs of any switching at the reading speed.

I agree there is no reason to do any switching, but maybe some digital noise makes it back out in a periodic way as a side effect of the readings?

 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
...
 My unit has the laser trimmed glass/ceramic resistor packages in the DC Signal Conditioner module. So I'm curious how well mine performs now that everyone is sharing data from their units. If anyone is curious, I can take photos of the DC Signal Conditioner or other boards.
...

What software are people using to collect and process the data? I have MATLAB and could write my own code to collect and plot the data, but if there is a turn-key solution I wouldn't mind using that to start with.

I would like to see the signal conditioner inside of yours.   Guessing it's a fair bit different than the one shown on the XDEV site.  I only opened the ADC and ohms clam shells.   

Which interface is your meter using?   I use LabView.

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us

That looks very interesting!

What manual are you finding that diagram in, I don't seem to be able to find that diagram in any of the ones I have?  (One found online, another purchased from Artek).

I guess it's time to open up the DC conditioners and see what revision level they are...   do you happen to know the rev letter on yours?

The diagrams are errata notes tacked on at the end of the manual. Apparently Fluke was too lazy to actually update the manual (though HP was often just as lazy). The pdf on xDevs for the 8505A (with 1991 revision date) has them at the end of the document. It looks like the same pdf I got from Fluke's website. So sadly there isn't a better quality scan of the circuit diagrams unless your Artek manual has them in it.

I don't remember the rev. code for my module, but I'll post some photos of it here in the next day or so.
« Last Edit: January 31, 2021, 03:05:26 am by garrettm »
 

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us
I would like to see the signal conditioner inside of yours.   Guessing it's a fair bit different than the one shown on the XDEV site.  I only opened the ADC and ohms clam shells.   

Which interface is your meter using?   I use LabView.

Looking back at the photos of your unit, someone has replaced your DC Signal Conditioner with a newer module. The older modules used two metal stickers while later units switched to a single plastic sticker and molding the warning text into the lid. What number does your module show on the label (I can't see it in the photo)?

I have the standard IEEE488 interface on mine and a GPIB to USB adapter floating around somewhere that I need to find...
 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
[...]
The diagrams are errata notes tacked on at the end of the manual.
The errata are missing from the Artek 8506A manual, which also happens to be from 1988, so a few years older.  Hopefully no major changes in the Thermal RMS part in that time...

 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Just for fun, here is some price information for these units, and the options!  Don't forget to multiply prices by 1.79 to adjust for inflation until today!


« Last Edit: January 31, 2021, 04:19:09 am by SilverSolder »
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Just for fun, here is some price information for these units, and the options!  Don't forget to multiply prices by 1.79 to adjust for inflation until today!

While I'm glad to have found one to play with for $50, I can't imagine buying one in 1993 for more than an HP 3458A, once you factor in the appropriate options. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us
I'm pretty sure the 8506A and 8505A were on some sort of weird military contract. Otherwise it wouldn't have made sense to continue manufacturing these units all the way up to 1993--where they were long obsolete.

The unit I currently have came from the US Air Force, and the CT variant I had before it was also surplus Air Force gear.

Prior to the release of the HP 3458A, the 8506A wasn't a bad metrology grade meter. Though there were still arguably better meters from Solartron at the time. And while HP's 3456A had better DC volts spec below 100V as well as more comprehensive math functions, the 8506A's AC volts spec was pretty impressive for a 6.5 digit meter.
 
The following users thanked this post: SilverSolder

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
I would like to see the signal conditioner inside of yours.   Guessing it's a fair bit different than the one shown on the XDEV site.  I only opened the ADC and ohms clam shells.   

Which interface is your meter using?   I use LabView.

Looking back at the photos of your unit, someone has replaced your DC Signal Conditioner with a newer module. The older modules used two metal stickers while later units switched to a single plastic sticker and molding the warning text into the lid. What number does your module show on the label (I can't see it in the photo)?

I have the standard IEEE488 interface on mine and a GPIB to USB adapter floating around somewhere that I need to find...


881722 with a blue dot next to it.   You may also notice that many of the guides have been replaced as well.  There are at least three different colored rails used.   Guessing it was maintained by Fluke and the upgraded plugin was part of their service.

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Just for fun, here is some price information for these units, and the options!  Don't forget to multiply prices by 1.79 to adjust for inflation until today!

My first Fluke meter was only $299 in 1974.  It's on par with the free meters from Harbor Freight.   

https://youtu.be/ObKomuLLqU8?t=220

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
I would like to see the signal conditioner inside of yours.   Guessing it's a fair bit different than the one shown on the XDEV site.  I only opened the ADC and ohms clam shells.   

Which interface is your meter using?   I use LabView.

Looking back at the photos of your unit, someone has replaced your DC Signal Conditioner with a newer module. The older modules used two metal stickers while later units switched to a single plastic sticker and molding the warning text into the lid. What number does your module show on the label (I can't see it in the photo)?

I have the standard IEEE488 interface on mine and a GPIB to USB adapter floating around somewhere that I need to find...


881722 with a blue dot next to it.   You may also notice that many of the guides have been replaced as well.  There are at least three different colored rails used.   Guessing it was maintained by Fluke and the upgraded plugin was part of their service.

I think the blue dot is just a sticker used by the technician to indicate something about the module.   By complete coincidence, I have a sheet of exactly the same kind of little colored circles, and you will see those on my modules too, when I get around to taking some photos!

 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Just for fun, here is some price information for these units, and the options!  Don't forget to multiply prices by 1.79 to adjust for inflation until today!

My first Fluke meter was only $299 in 1974.  It's on par with the free meters from Harbor Freight.   

https://youtu.be/ObKomuLLqU8?t=220

Progress!  :D
 

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us
881722 with a blue dot next to it.   You may also notice that many of the guides have been replaced as well.  There are at least three different colored rails used.   Guessing it was maintained by Fluke and the upgraded plugin was part of their service.

Well, you're in luck! That's the same number as mine, so you have the latest DC signal conditioner that I am aware of.

Here's a picture of my modules if anyone is interested.

1162014-0
 

Offline dietert1

  • Super Contributor
  • ***
  • Posts: 2473
  • Country: br
    • CADT Homepage
I am using one of the USB-GPIB interfaces we made 12 years ago. Send '*' to reset the instrument, wait 3 seconds, then '%' to go remote. Then exercise the DC conditioner relays sending "R2?" and "R3?" several times, each time with a 1 second delay. Then data taking starts by setting range "R2" and a loop with trigger "?" and reading the result. Timeout is set to one second and reading terminates when result is available.
This works very welll until somebody touches the USB cable and the FTDI driver hangs. Error recovery requires pulling the USB cable and enumeration. Sigh.

Regards, Dieter
 
The following users thanked this post: SilverSolder, garrettm

Offline srb1954

  • Super Contributor
  • ***
  • Posts: 1124
  • Country: nz
  • Retired Electronics Design Engineer
Prior to the release of the HP 3458A, the 8506A wasn't a bad metrology grade meter. Though there were still arguably better meters from Solartron at the time. And while HP's 3456A had better DC volts spec below 100V as well as more comprehensive math functions, the 8506A's AC volts spec was pretty impressive for a 6.5 digit meter.
At the time the 8506A was pretty much the best AC accuracy you could get in a DVM. To get better accuracy you had to move up to an AC-DC transfer standard.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Quote
Then exercise the DC conditioner relays sending "R2?" and "R3?" several times, each time with a 1 second delay.
Why?  Are the relays in yours starting to fail?

With mine using RS232, afraid I can't be of much help on the software side of things. 

881722 with a blue dot next to it.   You may also notice that many of the guides have been replaced as well.  There are at least three different colored rails used.   Guessing it was maintained by Fluke and the upgraded plugin was part of their service.

Well, you're in luck! That's the same number as mine, so you have the latest DC signal conditioner that I am aware of.

Here's a picture of my modules if anyone is interested.
If you like, I can take some decent pictures of this board to compare with yours and see if there are any differences.  Perhaps the performance was improved to the point that Fluke would just upgrade them as part of their service.   

Offline dietert1

  • Super Contributor
  • ***
  • Posts: 2473
  • Country: br
    • CADT Homepage
Had some trouble with the relays of one unit (after not having used the Flukes for two or three years. Same as the front/back switches in some other DVMs.

Regards, Dieter
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
The front rear switch on this meter had also oxidized.  I doubted I would find a replacement so  I ended up flushing it, then running about 50mA at a 1V through each contact while I cycled the switch assembly 10-20 times.  Then flushed it once more.   Took several minutes to go through them all but the switch is now very quite. 

I was a bit surprised that was the only mechanical issue with the meter.

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Quote
Then exercise the DC conditioner relays sending "R2?" and "R3?" several times, each time with a 1 second delay.
Why?  Are the relays in yours starting to fail?

With mine using RS232, afraid I can't be of much help on the software side of things. 

881722 with a blue dot next to it.   You may also notice that many of the guides have been replaced as well.  There are at least three different colored rails used.   Guessing it was maintained by Fluke and the upgraded plugin was part of their service.

Well, you're in luck! That's the same number as mine, so you have the latest DC signal conditioner that I am aware of.

Here's a picture of my modules if anyone is interested.
If you like, I can take some decent pictures of this board to compare with yours and see if there are any differences.  Perhaps the performance was improved to the point that Fluke would just upgrade them as part of their service.


Every circuit board has a revision letter written on it.

The manual has a page that explains what revision letters are covered by it.  This particular manual is probably too old to cover the newer model numbers...  but presumably the newer models have revision letters too.

In my collection of parts, I have encountered a wide range of revision letters, so chances are quite good that we don't all have exactly the same modules...







 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf