Author Topic: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)  (Read 59688 times)

0 Members and 1 Guest are viewing this topic.

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #175 on: February 07, 2021, 06:18:52 pm »
Actually, it made sense to concatenate the two EPROMS before disassembling, because the code runs across both EPROMS and makes subroutine calls from one EPROM to the other...

Attached:  disassembly of entire firmware across both EPROMS.  This one actually seems to make some sense, but would still need an amazing amount of work to correct and make sense of...


« Last Edit: February 07, 2021, 06:20:54 pm by SilverSolder »
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #176 on: February 07, 2021, 06:42:34 pm »
I pulled my controller board last night and when I attempted to use my programmer, it was dead.  Looking into it, the parallel port board is dead.   A bit of probing and of course, the main IC has a low impedance pin.   I am not sure about the programmer.   It's rare I program parts like this anymore and suspect the it was damaged during a lightning strike we had a while back that damaged some of my test equipment.   

I scrapped my old XP computer over the summer assuming I would never need it.   That leaves the old dual Pentium 200MHz PC.  lol.   

Anyway, thanks for capturing the PROMs.   It looks like  my controller is revision J.

In the mean time, I am still very interested in what others are finding for times.  Again, no need for GPIB and PC software.  Just use the sync and a scope. 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #177 on: February 07, 2021, 07:15:00 pm »
8505A with scope on 'Scan Advance' BNC on back.

Power on, no changes except selecting AUTO range: 1.87Hz

Press AVG button: 232mHz

Press FILTER button, no change.  Turned filter off again.

Press SAMPLE button, light flashes faster and 6.41Hz

Press "STORE 6 SAMPLE" or any number over 6 and the rate goes down, I didn't record them.

Press "STORE 5 SAMPLE" and I get 6.41Hz again.

Press "STORE 4 SAMPLE": 9.62Hz

Press "STORE 3 SAMPLE": 12.7Hz

Press "STORE 2 SAMPLE": 15.5Hz

Press "STORE 1 SAMPLE": 17.6Hz

Press "STORE 0 SAMPLE": 18.6Hz

All numbers are a bit approximate and weren't perfectly stable.  I have a National Instruments GPIB-to-USB adapter but my lab labtop (labtop?) is being used elsewhere at the moment.  I suspect that the scan advance doesn't happen at the same rate as the rate that you might get readings over GPIB, but someone will have to try that.

AFAIK, it is 27 samples at power on and 25 when you press SAMPLE one time.

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.
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #178 on: February 07, 2021, 08:57:01 pm »
Thank you very much for running that.  18.6Hz  is in the ballpark of what I was seeing (21Hz).  What version of firmware does your meter have?

I found the exact same printer port card and thought I should test the programmer first.   Pulled out my old Dec 486 PC  and sadly, the programmer also appears to have been damaged.   It's a very old EETOOLs ALLMAX.  I've hacked their software a few times to add support for odd devices.    They used some PALs right on the printer port interface so I doubt there is anything I can do to repair it.  Someone has one on ebay but they want $200+ for it.   :-DD  Time to look for a new programmer. 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #179 on: February 07, 2021, 08:59:20 pm »

STORE 0 SAMPLE ---> ~20Hz on the scan advance output.

8506A rev. 6.0.7
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #180 on: February 07, 2021, 09:08:25 pm »
That being the case, whats all that hype in the manual?  I am not understanding that <5ms should yield > 200Hz yet we are seeing 10X slower results.     

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #181 on: February 07, 2021, 09:36:12 pm »
It says 'HI 5.0.4' on power up.  Is that the firmware version? 
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 SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #182 on: February 07, 2021, 09:42:20 pm »
That being the case, whats all that hype in the manual?  I am not understanding that <5ms should yield > 200Hz yet we are seeing 10X slower results.     

Hmm yes, the spec sheet does mention 500 readings per second at 6.5 digits.

First, can the meter do that internally, before we talk about outputting the results to the outside world?

Each reading is normally synced to the AC mains waveform.  The circuitry supports 4 trigger points per mains cycle (that seems a lot - is it not normal to trigger just twice per cycle?).  The 4 points mean that each reading has to take 1/4 mains cycle at best, which in a 60Hz country means 4ms (stated in manual) or 4.16ms if you check the math.

You can turn off the line sync mode and let the meter read asynchronously instead (GPIB command T).  Here, the meter no longer syncs to the AC waveform and just runs as fast as it can, held back only by its own processing capability.  This is where the meter can do a reading in just 2ms (stated in manual...  with dc zero, offset, limits, and calibration factors turned off!) which works out to...   500 readings per second, as claimed by Sales!

What the sales department also neglects to mention is that it is a bit of a challenge to get the individual readings out of the instrument at high rates!

We know from the scan advance scope tests that the internal display cannot go any faster than 20Hz, but there is no reason the serial card (and GPIB) shouldn't be able to run significantly faster than 20Hz.

It is actually possible to de-activate the internal display when operating in remote mode (command "D") - perhaps this is the secret for getting the speed up?   

So we have three things to try: 
1) disable internal display,
2) mains asynchronous readings,
3) binary mode comms

I suspect if all three are in place, we will see rates closer to what Sales & Marketing are saying.

For most purposes, it is probably easier to just let the meter do the averaging internally!
« Last Edit: February 07, 2021, 09:43:59 pm by SilverSolder »
 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #183 on: February 07, 2021, 09:43:29 pm »
It says 'HI 5.0.4' on power up.  Is that the firmware version?

Yes, and - the evidence so far - indicates that version number starting with 5 means 8505A, starting with 6 means 8506A.
 

Offline srb1954

  • Super Contributor
  • ***
  • Posts: 1124
  • Country: nz
  • Retired Electronics Design Engineer
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #184 on: February 07, 2021, 09:49:53 pm »
Apparently the 8080A is only a "up to 4MHz" processor, so it is not picky about EPROM speed.

I wonder why Fluke chose a 15.36MHz crystal to run this board...  would that not run the CPU at 1.92MHz?


8080A     2.0MHz
8080A-1 3.125MHz
8080A-2 2.67MHz
8080B     4.0MHz


[edit: CPU speed]
15.36MHz is a good frequency for generating the baud rate clock for a UART although this doesn't seem to have been used for the 8506A serial interface as this has its own crystal on board.

15.36MHz/100 = 16 x 9600 Baud

Perhaps Fluke just used this crystal frequency because they had a large stock of them they wanted to use up.
« Last Edit: February 08, 2021, 08:58:02 pm by srb1954 »
 
The following users thanked this post: SilverSolder

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #185 on: February 07, 2021, 10:55:04 pm »
As I mentioned, I had been running with the display turned off both asynch and synch and both ascii and binary.   I've tried them all.   As I understand it, the sync output should pulse on every conversion.  You can see that from the data I have collected, I never see it get anywhere near the claimed conversion rates.   

Quote
I suspect if all three are in place, we will see rates closer to what Sales & Marketing are saying.
This is where I saw the fastest numbers but again, no where near the claimed numbers. 

Quote
For most purposes, it is probably easier to just let the meter do the averaging internally!

And again, just to be clear, I was running averages as well while monitoring the sync output assuming that when the meter would average, it wouldn't throttle the conversion speed.  It wasn't even close.   



I looked in my stock of ICs and have some brand new 128s.   So I looked at the schematic and I notice U24 routes A13.  Then we see U23 ties pin 26 high.  Looking at my original manual, it is also routed like this.   They must have planned on adding some features to U24 at one point and decided to support the 27128.  More odd is U18 (74S287) which is a PROM that they are using as part of their decoder logic.   Then they have jumper W3 (mine is strapped 1-2) which would change the decoding.   

I started to look at the TL866CS programmer.  It's cheap enough but after seeing the comment about reading the parts three times to make sure they are right and then there were several posts about the company having malware in their software.   But, then again, you get all these crazy adapters and such for under $100.  What's been your experience with it?

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #186 on: February 07, 2021, 11:49:18 pm »
As I mentioned, I had been running with the display turned off both asynch and synch and both ascii and binary.   I've tried them all.   As I understand it, the sync output should pulse on every conversion.  You can see that from the data I have collected, I never see it get anywhere near the claimed conversion rates.   

OK, perhaps the explanation for Sales & Marketing's 500 reading per second claim is that the speed of 500 readings per second is only possible inside the instrument itself, as part of an (internal) averaging process?  Perhaps this instrument NEVER returns a result that is not the outcome of an averaging process of some kind?

Maybe it is possible to put the scope on the ACK line of the digital part of the R2 board, to get an idea of how many A/D conversions per second are actually occurring? - that has to be the ultimate limitation of how fast it is running, right?



I started to look at the TL866CS programmer.  It's cheap enough but after seeing the comment about reading the parts three times to make sure they are right and then there were several posts about the company having malware in their software.   But, then again, you get all these crazy adapters and such for under $100.  What's been your experience with it?

I just installed the latest software yesterday (which I got from the company's web site (http://autoelectric.cn/MiniPro/minipro_setup685.rar) and my antivirus did not balk at anything, nor do I see any evidence of phoning home - there is no need to register or sign up for an account or any of the BS we typically get from Big Tech nowadays. Considering how popular this programmer is, and the amazing amount of devices it supports, it seems a no brainer. 

I originally bought it to program Atmel ATTiny chips, but never actually used it for that...  so I am dusting it off for this Fluke Boat Anchor Project!  :D

I'll have a go at extracting some EPROMs with it maybe later tonight and let you know how that goes.

[Edit] It was very easy - this thing is a winner (for this job, anyway)!
« Last Edit: February 08, 2021, 01:06:31 am by SilverSolder »
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #187 on: February 08, 2021, 12:14:03 am »
If you look at the line synchronous spec of 4ms, that is 250 per second.  If you take the amount of time it takes to average the max number of samples, 217, divided by 250, you get about the right number of seconds to match the 9+ minute acquisition time or output rate at that setting. 
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 SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #188 on: February 08, 2021, 12:39:44 am »
If you look at the line synchronous spec of 4ms, that is 250 per second.  If you take the amount of time it takes to average the max number of samples, 217, divided by 250, you get about the right number of seconds to match the 9+ minute acquisition time or output rate at that setting.

You get even closer if you divide the 60Hz period by 4.  I think we have this right!

It is pretty cool that it samples 4 times per mains cycle, on the flanks.  Maybe it is less noisy there than at the peaks (not to mention, 2x more of them!).
 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #189 on: February 08, 2021, 01:00:08 am »
I was able to use the TL866CS to read the EPROMS from a controller out of an 8505A.  The firmware is revision 505/606.  A binary comparison with the files posted by @garrettm showed zero differences.

The cover says part no. 660563 but inside, the PCB had no. 638544 printed on it...  not sure if some replacement/swapping has occurred in the meter's life. 

It is a Rev W, not sure we have seen one of those before!







 
 
The following users thanked this post: garrettm

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #190 on: February 08, 2021, 01:01:31 am »

By the way, what is the recommended method for pulling and inserting these kinds of chips out of / into their sockets?  It seems a challenge to avoid bending pins...
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #191 on: February 08, 2021, 01:05:29 am »

By the way, what is the recommended method for pulling and inserting these kinds of chips out of / into their sockets?  It seems a challenge to avoid bending pins...

U-hook with small jaws

https://www.amazon.com/Jonard-S-340-Insulated-Spring-Extractor/dp/B006C4BCAK/ref=sr_1_1?dchild=1&keywords=Jonard+Tools+S-340+DIP%2FIC&qid=1612746294&sr=8-1
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.
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #192 on: February 08, 2021, 01:06:05 am »
Quote
Perhaps this instrument NEVER returns a result that is not the outcome of an averaging process of some kind?

I suspect that the Sync output does indeed pulse on each measurement sequence, as the manual suggests.  Besides "measurement sequence" the manual also mentions "measurement cycle" "readings" "samples" and "digitizing time".    The manual doesn't come out and define all of these.  At one point they state "reading (display update)" so it seems like that would also include a measurement cycle and measurement sequence.   I don't understand why anyone would lead the reader through such a fog zone.   

I was reading about the programmer's software here: 
https://www.eevblog.com/forum/beginners/tl866ii-plus-tl866acs-open-source-software-(oem-software-has-malware)/msg2799742/#msg2799742
Hopefully this was resolved.

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #193 on: February 08, 2021, 01:29:54 am »
[...]
I was reading about the programmer's software here: 
https://www.eevblog.com/forum/beginners/tl866ii-plus-tl866acs-open-source-software-(oem-software-has-malware)/msg2799742/#msg2799742
Hopefully this was resolved.

VirusTotal can give false positives from some of the many scanners it uses - I have seen that before even with stuff I've written myself.  I would guess the fact that it installs a USB driver for the device might appear "low level" to some scanners.  Nothing in life is certain, but this looks safe to me.

 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #194 on: February 08, 2021, 01:30:47 am »

By the way, what is the recommended method for pulling and inserting these kinds of chips out of / into their sockets?  It seems a challenge to avoid bending pins...

U-hook with small jaws

https://www.amazon.com/Jonard-S-340-Insulated-Spring-Extractor/dp/B006C4BCAK/ref=sr_1_1?dchild=1&keywords=Jonard+Tools+S-340+DIP%2FIC&qid=1612746294&sr=8-1

I have one, but find it hard to use...  suddenly, one side or the other "lets go" and there you are...   bent pins all over the shop.
 

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #195 on: February 08, 2021, 01:42:51 am »
The datasheet states "up to 40 ASCII character readings per second" on the bit serial interface (at 9600 baud) and "up to 500 readings per second" in the binary mode are possible on the 8/16bit parallel and GPIB interfaces.

The 500 readings/s mode assumes you copy the cal data off the meter and perform the zero and gain corrections yourself. It also says that the Line Asynchronous mode is a 3-byte binary format, rather than 6-byte. Assuming <CR><LF> termination characters, that's 5-bytes per transfer. Using command J should drop that to 4-bytes.
« Last Edit: February 08, 2021, 04:36:58 am by garrettm »
 

Offline garrettm

  • Frequent Contributor
  • **
  • Posts: 319
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #196 on: February 08, 2021, 02:04:06 am »
I have one, but find it hard to use...  suddenly, one side or the other "lets go" and there you are...   bent pins all over the shop.

For large pin count ICs it's a bit tricky.

You generally have to "see-saw" the IC out slowly and with decreasing force as it starts to come loose. Steady pressure will result in an explosive release that usually bends the leads since the applied force is invariably at a sight angle. You can also use a flat-head screwdriver to pry up each edge in a similar see-saw action, though it might take two screw drivers on the final leg of the release to prevent pins from being bent. I would apply a small amount of masking tape on the bare metal of the IC grabber and/or screwdriver to prevent scuffing the ceramic.
« Last Edit: February 08, 2021, 02:06:41 am by garrettm »
 
The following users thanked this post: SilverSolder

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #197 on: February 08, 2021, 02:12:27 am »

Perhaps those plastic tools used to pry cell phones apart might work in this application...
 

Offline SilverSolderTopic starter

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #198 on: February 08, 2021, 02:15:56 am »
Quote
Perhaps this instrument NEVER returns a result that is not the outcome of an averaging process of some kind?

I suspect that the Sync output does indeed pulse on each measurement sequence, as the manual suggests.  Besides "measurement sequence" the manual also mentions "measurement cycle" "readings" "samples" and "digitizing time".    The manual doesn't come out and define all of these.  At one point they state "reading (display update)" so it seems like that would also include a measurement cycle and measurement sequence.   I don't understand why anyone would lead the reader through such a fog zone.   

I was reading about the programmer's software here: 
https://www.eevblog.com/forum/beginners/tl866ii-plus-tl866acs-open-source-software-(oem-software-has-malware)/msg2799742/#msg2799742
Hopefully this was resolved.

By the way, it looks like the current model is the TL866II Plus, the TL866CS is no longer made.
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: Some old school instruments showing how it's done (HP 3325A and Fluke 8506a)
« Reply #199 on: February 08, 2021, 02:35:06 am »
I used to have a tool to extract DIPs but it seems to have been lost or pitched.   I have a few of the PLCC pullers still.  I used to have a DIP insertion tool that would form the leads.   Days gone by but looks like Grainger still offers some of them.     
https://www.grainger.com/category/tools/hand-tools/electrician-s-hand-tools/communications-tools/ic-insertion-extraction-tools   

VirusTotal can give false positives from some of the many scanners it uses - I have seen that before even with stuff I've written myself.  I would guess the fact that it installs a USB driver for the device might appear "low level" to some scanners.  Nothing in life is certain, but this looks safe to me.

I was reading the whole thread,  where others chimed in. Looks like it wasn't just one scanner.   
https://www.eevblog.com/forum/beginners/tl866ii-plus-tl866acs-open-source-software-(oem-software-has-malware)/msg2802468/#msg2802468

They were talking about both models.  It's certainly low cost.   Are the pins protected against shorts to ground and adjacent pins?   Is the reason for reading the parts several times because of the quality of the socket?  Are they using a name brand socket? 


40 ASCII character readings per second would be 2.8 readings a second with 14 characters but I was seeing data much faster than this.   It's odd they would use the term Readings as they defined that to mean display updates in the manual.   I wouldn't be surprised it they turn off the display in these high speed modes.    Still, I don't buy it.  The reason is because I never see the sync running that fast with no data being transferred.   Then again, maybe reading in this context is that rate the ADC is making decisions, not how fast the display is being updated.   

Easy enough to see if using the GPIB will magically speed it up.   If you start seeing actual full packets of data showing up at 2ms like they suggest, that would be very good to know.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf