Test Equipment Anonymous (TEA) group therapy thread

[srb1954]

Now if you want some fully matured replacements, ebay can help. https://www.ebay.co.uk/itm/165856547320



David

Interestingly the item description says "New other (see details): An item in excellent, new condition with no wear."

So, these capacitors are just as they were when they came off RIFA production line: pre-cracked and primed for explosion. Just add a little moisture.

[Vince]

MicroPross ROM 3000U Vintage EPROM Programmer

Oh, making progress here, as of this afternoon we now have.. PICTURE !!!
We even have a full video of the boot process, for your enjoyment !

Yes this thing is not deaf and mute anymore, it can now talk to me ! It's not a mysterious black box anymore !

SEE HERE -->

https://www.eevblog.com/forum/repair/vintage-chip-programmer-micropross-rom-3000u/msg4606651/#msg4606651

[factory]

Now if you want some fully matured replacements, ebay can help. https://www.ebay.co.uk/itm/165856547320



David

Interestingly the item description says "New other (see details): An item in excellent, new condition with no wear."

So, these capacitors are just as they were when they came off RIFA production line: pre-cracked and primed for explosion. Just add a little moisture.

Well in my opinion they were never fit for purpose when they left the production line, so they are as good as new  , same as every other metallized paper capacitor that is still made by RIFA/Kemet & WIMA, they all need to become extinct and be found only in history books. 

Pity they are £10, could have made a rather nice string of random smoke devices to hang up outside, instead of the boring LED light strings everyone has this time of year.

David

[factory]

...snip...

Here is what I ended up doing in the evenings without the web to distract me, I've plagiarised the design for the A14 board that is missing from my HP 5327B (the second one, part of HP 9500D-221 computer setup), luckily I had a board to measure & copy, just need to figure out what to replace the Motorola MDTL ICs with, the MC1806P, MC1808P & MC846P are available, but not at sensible prices, some MC1806P, MC1808P outputs are joined together too.
The custom 1820-0227 aka 1820-0413 divider IC will need replacing with a 7490 or similar, on an adaptor board Note: the ones of the A5 timebase board would need extra logic gates, as they use the gated output function.



P.S. the JCBPCB quick quote gives around $25 for this board with ENIG plating, anything other than green or purple solder mask is $$$ extra.

David

Got the last piece of this puzzle reverse engineered, the HP 1820-0413 divider IC, ended up using a 74LS90, with two 74LS14 inverter gates, one 74LS02 NOR gate and a random used HP 1854-0071 trannystor providing the open-collector gated output.









Will do a KiCad layout one evening, sometime in the new year, this is not a urgent project, just wanted to test all the functions, including those not used on the A14 board.

I also tested a plain 74LS90 (in both BCD and symmetrical divide by 10 modes) in the 5327B and it does work fine, it's only used to divide by 10 on the A14 DVM control board, the resets & gated output not required. The 5326B has of course stopped working. 







David

[Vince]

Well done, that's dedication / love we have here ! 

Can't wait to see the new board.

Too bad for the 5326B but well...  in some way it's kinda cool as it will give us some cool TEA content at some point ! 

[tautech]

Yaaaayyy!

I hacked my Siglent SSA3021X Plus to a SVA1032X, all working, all options unlocked and serial number is back in too.

Great success!
Now to learn how to use a VNA...

Cool, enjoy. 
Did I mention I really like these and although I have a fancy 4 port $ $ one too SVA1032X is always the first I get out to do something.

What Cal kits do you have ?

[TERRA Operative]

I have a HP 85032B Type N cal kit that I picked up really cheap, and I am working on finding a BNC cal kit to use for BNC stuff too.

[tautech]

I have a HP 85032B Type N cal kit that I picked up really cheap, and I am working on finding a BNC cal kit to use for BNC stuff too.

IMO the future lies with SMA. Especially for price and convenience and ease of knocking up your own Cal kit.
There's some good other stuff in this thread:
https://www.eevblog.com/forum/rf-microwave/diy-short-open-and-load-for-vna-calibration/msg3305756/#msg3305756

If ya like we can do some Cal comparisons as I have the 4.5 GHz Siglent kit F603FE
https://siglentna.com/product/f603fe_female_3_5mm_vna_cal_kit/

More stuff here:
https://siglentna.com/wp-content/uploads/dlm_uploads/2021/04/Mechanical-Calibration-Kit-Datasheet-v1.4.pdf

Your call, here or the SVA thread.

[TERRA Operative]

I'll have a look into it all. I'll have to settle on a connector type, SMA does look good as it's easy to get and seems cheaper for connectors than N.

I'll have to figure out how this VNA works then some comparisons could be interesting.



On another subject, it turns out busted HP power sensor heads can be used to build a calibration rig for HP power meters. I have a couple busted ones and a 437B that needs a tweak, so I know what my next project will be....

[Robert763]

Don't bother with BNC calibration on the VNA. They are so inconsistent it's a waste of time. Just calibrate with the N kit and use the best adaptors you have. Anything with BNCs is unlikely to be critcal. If the size is important use TNCs.

Robert.

[TERRA Operative]

I have HP 1250-0780 N-to-BNC adapters, so pretty decent ones I guess.

[wkb]

Don't bother with BNC calibration on the VNA. They are so inconsistent it's a waste of time. Just calibrate with the N kit and use the best adaptors you have. Anything with BNCs is unlikely to be critcal. If the size is important use TNCs.

Robert.

Yep. For entertainment sake: wiggle a BNC cable on a VNA in 'thru' mode and watch the plot on the screen...

Wilko

[Vince]

Micropross 3000U Vintage EPROM programmer saga

Sorry to bother you again people, but the day has been very productive on that front... we have excellent news to report, have a look :

https://www.eevblog.com/forum/repair/vintage-chip-programmer-micropross-rom-3000u/msg4607824/#msg4607824

We have found the MSDOS program that's required to talk to that programmer and make it a useful device, and got it to run after some work, fantastic... however it's not all rosy as you will see.

Much work needs to be done still, it could still be a dead end...

I am posting about it now since the work required to take it further will take a few weeks / months.

I call it the end of Season #1  ... see you later for Season #2 !!! 

[Vince]

Question about serial ports .....

I thought it would be useful to spy on the serial link while the MSDOS program is talking to my EPROM programmer.

Computer is running windows 95. Since the MSDOS S/W runs in windowed mode, that means I can run other programs alongside it.

So I am thinking... does anyone know if there exists some kind of Windows 95 software that could "spy" the traffic on the serial port, and show it to me in a terminal (if that's plain text), or if it's binary, save it to a file that I could then open in a HEX file editor to see if I can find interesting things in it ?

That's one for the grey beards I guess...

EDIT : I think that's probably not possible eh ? I mean IIRC once a program asks the OS to "open" a COM port, it's not available to other programs, right ? 

EDIT #2 : may have found a solution... when I was sorting all my boxes of random cables a while back, I ran into some weird looking serial cables....I now remember one of these weirdos you were able to tell me what it is : it replicates a serial port onto another serial port, precisely so you can spy on it....  OK I will try to find that cable....

EDIT #3 : nope that won't work, I have only one COM port on that machine.... need to see if I have some old ISA or PCI serial port extension cards in my junk boxes... crossing fingers....

Still, would be so much more practical if it could be done in S/W....

[DC1MC]

Question about serial ports .....

I thought it would be useful to spy on the serial link while the MSDOS program is talking to my EPROM programmer.

Computer is running windows 95. Since the MSDOS S/W runs in windowed mode, that means I can run other programs alongside it.

So I am thinking... does anyone know if there exists some kind of Windows 95 software that could "spy" the traffic on the serial port, and show it to me in a terminal (if that's plain text), or if it's binary, save it to a file that I could then open in a HEX file editor to see if I can find interesting things in it ?

That's one for the grey beards I guess...

EDIT : I think that's probably not possible eh ? I mean IIRC once a program asks the OS to "open" a COM port, it's not available to other programs, right ? 

EDIT #2 : may have found a solution... when I was sorting all my boxes of random cables a while back, I ran into some weird looking serial cables....I now remember one of these weirdos you were able to tell me what it is : it replicates a serial port onto another serial port, precisely so you can spy on it....  OK I will try to find that cable....

EDIT #3 : nope that won't work, I have only one COM port on that machine.... need to see if I have some old ISA or PCI serial port extension cards in my junk boxes... crossing fingers....

Still, would be so much more practical if it could be done in S/W....

I'm bald as a pumpkin, but try this one Vince:
 https://www.heise.de/download/product/advanced-serial-port-monitor-34804

or this from the belly of the beast:

https://learn.microsoft.com/en-us/sysinternals/downloads/portmon

Cheers,
DC1MC

[Vince]

Thanks a lot, will have a look at that ! 

[factory]

Well done, that's dedication / love we have here ! 

Can't wait to see the new board.

Too bad for the 5326B but well...  in some way it's kinda cool as it will give us some cool TEA content at some point ! 

As the IC in question is used in many of the HP counters designed in the late 60s & early 70s, I thought it best to try replicate all the functions in case they are needed (only one I haven't seen yet is the reset 9), even though I only needed a simple divide by 10 function for the DVM control board. Next challenge is to design a small enough board to fit various counters.

The 5326B lights up but no one is home, no counting or voltage measurement possible, I'm sure it worked before it was in storage.

Didn't have as much time today to work on the other projects, due to hospital IV infusion in the morning, thankfully I didn't malfunction this time.

This afternoon I got the 74HCT390 ICs assembled onto veroboard, for the Advance DMM2 repair attempt number 2.

Almost two weeks ago I had a look at the various inputs/outputs of the counter/latch/multiplex board with 74LS90s, with the Yokogawa DSO, everything seemed OK apart from the number of count pulses before the carry output changes, this should be approx 1000 (or 999 from the logic analyser result taken by another forum member).
Measuring ten of the count pulses and the width of the carry pulse between start & stop of counting, I calculated I was getting approx 2400 pulses before it changed back to zero, no wonder things were very wrong, looking more closely with the zoom it is most probable the overloaded 5V rail might be the problem, many pulses were quite low.



The last week before Christmas I put in a order with Mouser for some 74HCT390 and a few other bits (some other common 74 series IC in HCT version, a few extra 74LS90s & those SIL leadframe contacts Terra found  ), these arrived in four days, much quicker than anything sent from within the UK ATM. 


This afternoon I built up a vero board with the 74HCT390 counter stage & attached it to the latch/multiplexer board and fitted HCT version of the 74LS74 & 74LS14, I now have a more healthy +5.1V on the supply rail  , but seem to have created another problem, need to check if it's a wiring error on my part, or maybe a dead IC, or a dead digit readout bulb (might not be as the decimal point still lights). 





P.S. Here are some pictures showing the condition of the original custom IC from the DMM2, slight issue with iron oxide, mixed with corroded copper.





David

[Vince]

As the IC in question is used in many of the HP counters designed in the late 60s & early 70s, I thought it best to try replicate all the functions in case they are needed (only one I haven't seen yet is the reset 9), even though I only needed a simple divide by 10 function for the DVM control board. Next challenge is to design a small enough board to fit various counters.

Time to go surface mount eh ?! 

[factory]

As the IC in question is used in many of the HP counters designed in the late 60s & early 70s, I thought it best to try replicate all the functions in case they are needed (only one I haven't seen yet is the reset 9), even though I only needed a simple divide by 10 function for the DVM control board. Next challenge is to design a small enough board to fit various counters.

Time to go surface mount eh ?! 

Yes, all the ICs are available in surface mount, 74LS90 in SOIC, tiny single & double gate versions are made for some, like the HCT14 & HCT02.

Here are some of the counter boards that use the 1820-0412/413, the 0413 is an 0412 rated for at least 12.5Mhz operation, 16 pin dual versions were also made to add to the confusion, yet to find any though and some older counters use a can IC version too. 

Timebase board from 5325B



Timebase board from 5327B



Mystery board from one of HPs automated measurement systems.





David

[tautech]

I have HP 1250-0780 N-to-BNC adapters, so pretty decent ones I guess.

Thing is, most measurements are on the far end of some cable and generally not hanging on the adapter so we need to measure just the DUT not the cabling and adapters to it so our Cal kits need be in a form that suits our DUT and/or cabling.
Cal kit gender and form should drive your needs otherwise you'll end up with a selection of different type Cal kits.

When I looked at this a while back not wanting to throw a fortune at it SMA was selected as all manner of reasonably priced cabling and connectors were readily available and in reality you Cal any influence they have on measurements out anyways.
You just need a reasonable SOLT collection and for the quality of measurements the novice does the don't need be super flash to obtain real world results.

Go have a look here for the result I got with just a bit of mucking around and not a Cal kit in sight and only a N-BNC adapter uncertainty as the feedline coax was an integral part of the DUT.
https://www.eevblog.com/forum/rf-microwave/antenna-project-log/

[Neomys Sapiens]

Question about serial ports .....

I thought it would be useful to spy on the serial link while the MSDOS program is talking to my EPROM programmer.

Computer is running windows 95. Since the MSDOS S/W runs in windowed mode, that means I can run other programs alongside it.

So I am thinking... does anyone know if there exists some kind of Windows 95 software that could "spy" the traffic on the serial port, and show it to me in a terminal (if that's plain text), or if it's binary, save it to a file that I could then open in a HEX file editor to see if I can find interesting things in it ?

That's one for the grey beards I guess...

EDIT : I think that's probably not possible eh ? I mean IIRC once a program asks the OS to "open" a COM port, it's not available to other programs, right ? 

EDIT #2 : may have found a solution... when I was sorting all my boxes of random cables a while back, I ran into some weird looking serial cables....I now remember one of these weirdos you were able to tell me what it is : it replicates a serial port onto another serial port, precisely so you can spy on it....  OK I will try to find that cable....

EDIT #3 : nope that won't work, I have only one COM port on that machine.... need to see if I have some old ISA or PCI serial port extension cards in my junk boxes... crossing fingers....

Still, would be so much more practical if it could be done in S/W....

In addition to the tools named by DCMC, I use a SW called Serialtest, which needs to be run in a DOS window.
There is an additional newer one, but that name I can only give you after the 9th.

I keep cables with IEC625--style connectors for Sub-D25 and Sub-D9 RS-232 connections. then I have a Sub-D37 on the other end, and a cable re-splicing from Sub-D37 to two serial ports. With this setup, Serialtest can monitor both sides of a communication including the important auxiliary signals with high time resolution.

[Vince]

Thanks for that MSDOS program, I am very interested in it because I am realising the Windows based monitors probably can't work in my case : I understand they spy the windows coms "infrastructure"  that windows programs have to use to get hold of the serial port. However MSDOS programs obviously don't use the windows infrastructure, they talk to the serial port directly or via the BIOS at most, but that's it.
So a windows based monitor can't see that, it flies under its radar...

So an MS-DOS monitor probably might work....


Yeah if just for fun I will try to find that splicing cable, hoping I kept it, and play with it a bit...


But, on the serial spying front.... I have great news already, hot from the oven !!! 

It's super exciting... go have a look :

https://www.eevblog.com/forum/repair/vintage-chip-programmer-micropross-rom-3000u/msg4608829/#msg4608829


I know I just said that Season #1 of that programmer was over but... well, let's just say it's a bonus track or something ?! Or a teaser / trailer for Season #2 ?! 

[Zoli]

Here are some of the counter boards that use the 1820-0412/413, the 0413 is an 0412 rated for at least 12.5Mhz operation, 16 pin dual versions were also made to add to the confusion, yet to find any though and some older counters use a can IC version too. 
...

Can you point me out to some documentation regarding those IC's? Maybe I can figure out something during my wrestling with the fast response bench supply...

[factory]

Here are some of the counter boards that use the 1820-0412/413, the 0413 is an 0412 rated for at least 12.5Mhz operation, 16 pin dual versions were also made to add to the confusion, yet to find any though and some older counters use a can IC version too. 
...

Can you point me out to some documentation regarding those IC's? Maybe I can figure out something during my wrestling with the fast response bench supply...

For the DIL versions of the IC, there is no information apart from the internal diagram I posted yesterday, which is from the pinout table provided in the HP 5326B/5327B manual (good pdf can be found on the web) and some test waveforms with the circuit diagrams, it's a custom part for HP as far as I know.
The 12.5MHz ref was mentioned in the parts list for the 0413, the older counters used one 0413 & seven 0412s, they must have improved the design as newer ones used all 0413s.

Tests have been done to determine operation of an original part and then I experimented with 74series TTL to get the same functionality, including reset timings and the open collector "gated" output.

Will have to look at some dead tree manuals for the can IC version.

David

[Zoli]

For the DIL versions of the IC, there is no information apart from the internal diagram I posted yesterday, which is from the pinout table provided in the HP 5326B/5327B manual (good pdf can be found on the web) and some test waveforms with the circuit diagrams, it's a custom part for HP as far as I know.
The 12.5MHz ref was mentioned in the parts list for the 0413, the older counters used one 0413 & seven 0412s, they must have improved the design as newer ones used all 0413s.

Tests have been done to determine operation of an original part and then I experimented with 74series TTL to get the same functionality, including reset timings and the open collector "gated" output.

Will have to look at some dead tree manuals for the can IC version.

David

Thanks for the info, I've missed that post; it looks like the(dual) 0413/0412 can be replaced with an 16V8 PLD; If I need some distraction to work on, I will let you know.