Test Equipment Anonymous (TEA) group therapy thread

[W6EL]

Hmm.

A little research shows that these are very likely sprague solid and wet tantalum capacitors. Go on google and type "vintage sprague capacitor with plus indicated tantalum" and you basically see these capacitors. (No, I am not buying any magical vintage caps! Save those for the hifi heads, maybe they can get a discount with some vintage carbon composite resistors too... )

I don't really mind if the leads dress differently. I'm more about function than trying to keep it looking vintage inside. I know there are folks out there that would open up the old caps, put new caps inside them, and then seal them up. Not for me! I just want the unit working and meeting original specifications, more or less.

Now, obviously, the caps inside are bound to be out of spec in a variety of ways and would probably be best replaced with any modern capacitor. But I wonder how much worse off it would be using aluminum electrolytic capacitors? People often site the ESR of tantalum capacitors, but let's be clear, the ESR is not necessarily lower with tantalum. And sometimes tantalum capacitors are chosen specifically for their higher-at-times ESR over aluminum electrolytic caps. What I have learned over the years is that the tantalum capacitor is a more stable ESR over temperature, whereas a typical aluminum electrolytic can vary considerably over typical lab temperatures.

My guess, looking over the circuit, is that most of the caps will not matter at all.

--E
de W6EL

[factory]

med6753,

I think you are correct. I don't see nearly as much of this stuff on the other components.

I assume the capacitors with the "+" indicated are tantalum, is that a safe bet? Is it going to matter much if I put aluminum electrolytics instead?

Thank you all for your ideas,

--E
de W6EL

The ones marked 30D are aluminum electrolytics from Sprague (now branded Vishay) and those marked 150D are solid tantalum. The 150Ds rarely give trouble, YMMV.
Both 30Ds & 150Ds are available from Mouser & other places, along with other brands of axial caps. The ordinary electrolytics can be replaced with heatshrink sleeved radial caps, TERRA has shown pictures of this method a few times.
The tants could also be replaced with cheaper bead types, with appropriate voltage de-rating if your determined to replace them, low leakage types are available too, take note of polarity markings from the datasheet too.

David

[TERRA Operative]

Here's a picture of the radial to axial capacitor trick. This is in my HP 334A but I've done it a number of times and it works great.
You get the latest capacitor types in a neat radial shape.

[W6EL]

Re: HP 344A, the more you look the better it gets! Very nice.

[factory]

also one part came off with the tape.

That is not nice.

You must expect similar stuff everywhere.

Dataman Datamess S3 after vinegar treatment, a bit cleaner, two SOT23-3 parts fell off, one never to be seen again, the reset switch also fell off, at least one fixed contact in the ZIF socket is fell off and the write lead socket is barely hanging on. A quick check there are some open traces, some resistors near the emulator plug have dropped in value too.

David

[EE4all]

Regarding radial to axial, I've done that several times also. It works fine. I often use a dab of adhesive to make sure it doesn't get pushed and possibly cause a short.

[factory]

A update on the Toshiba T4400C laptop from 1992.
The new flat flex cables for the replacement direct drive FDD arrived from JLC, as always they didn't work first time, first it complained of "no floppy error", this was the motherboard connector, it needed a thicker stiffener, some thin card got added. Then the FDD would only read 720kB floppies (yes I found one). The replacement FDD is a tri mode type, I had to cut track 13  to the FDD connector to disable the 1.2MB? mode, there is a pull up resistor in the FDD to set it to 1.44MB, this was being grounded by pin 14 at the motherboard end (Toshiba used reversed numbering at the other end of the cable  ). It now reads both types of disk I have, it writes to 1.44MB no problem, but formatting gives an error every time. 





The Dell monitor of shame being used again  , see below for the reason.



Other problem to fix is the TFT screen PCB, this was recapped with solid polymer aluminium caps, the problem it worked for a few minutes and has been intermittent ever since, opening & inspecting the board reveals some electrolyte that has appeared from somewhere (near the fixing holes), despite my best cleaning efforts, hoping cleaning again will fix this, as the TFT screen in this is much nicer, than the STN crap  in the slightly newer one. Adjustment pots seem OK, can't see any obvious failed traces.





David

[Vince]

Wow you are quite motivated with that Toshiba, keep at it 

The Dataman well OK, I admit now... it doesn't look too promising.... with a lost and unidentifiable  SOT23, and other things.... argh....
There are so many components in that Dataman that this SOT23 maybe is not essential to the operation of the device.

So maybe try to clean and reflow the mes here and there and see if it can power up and display something on the screen, even if just "Please let me die, this is agony! "
If it can "boot" and display stuff, might be worth Googling to see if there is a forum somewhere about Dataman stuff, where someone somewhere could tell you what this component is... if no luck with that, just put a 2N2222 in there ! 

[factory]

Missing SOT23 parts only have two connections, looking at the others they are marked A6, this seems to be BAS16 from Philips (now NXP), a switching diode, nothing fancy.
Only thing that concerns me about the Datamess, is what I can't see under all the ICs, on the plus side there are only two layers to the PCB.

Thanks, I will keep working on the Toshiba, once the screen is fixed I can look at the SSD CF card conversion and make it silent  , perfect for datalogging from TEA, without family complaining about fan noise of more modern machines.

David

[Vince]

Wow cool that you could identify the missing SOT23 

As for corrosion underneath the IC's, well just remove a couple here as a sample, in whatever area looks in the worst condition to you....
Other than the quad flat package, the other IC's should be easy to remove with even the cheapo 858D hot air station. Works for me at least.
There is just one mistake to avoid, which got me into trouble once, whjen I was very inexperienced... the old / first gen SMD boards, oftne they used a tiny drop of red glue under the IC to keep them in place during soldering..... so, I was heating up the IC's, heating up, heating up... they would not budge, until they did, once the glue was so cooked that it let go finally. By the time this happens the chip is completely cooked and smoking... So now I gently pry the chip while heating it up, to be able to break the glue once I estimate that the solder is melted.

Good luck with it !

Here it's been British weather all the way for a week and forecast is the same for at least the next two weeks, no rest ! Reminds me of my years in the UK ! 
 ---> Perfect weather to stay warm at home, working in the lab !! 

[McBryce]

Here's a picture of the radial to axial capacitor trick. This is in my HP 334A but I've done it a number of times and it works great.
You get the latest capacitor types in a neat radial shape.

That's not exactly an unusual type or value. Why didn't you just order axial versions? Like this: https://www.mouser.de/ProductDetail/EPCOS-TDK/B41689K7278Q001?qs=gvwIECp2jlnZKbZkYzNRZg%3D%3D

McBryce.

[AVGresponding]

I have what appears to be an older model not halogen bulb and two cells unlike any I've ever seen before.. I guess NiCd but honestly have no clue what we're talking about LOL
Any idea what I can use to replace those with?

Yes, horrid horrid Ni-Cd cells, similar to ones found in Time Electronics voltage standards.

You could replace them with Ni-MH, if you can find them the right size, but that's an old industrial cell shape, I doubt you could get them new. You might have to use a different cell size and shape, and adjust the wiring to fit. Anything over 400mAh or so should be ok, though it's difficult to accurately judge the size from the pics. Any chance you could get pics of the writing on the cells?

[factory]

Here's a picture of the radial to axial capacitor trick. This is in my HP 334A but I've done it a number of times and it works great.
You get the latest capacitor types in a neat radial shape.

That's not exactly an unusual type or value. Why didn't you just order axial versions? Like this: https://www.mouser.de/ProductDetail/EPCOS-TDK/B41689K7278Q001?qs=gvwIECp2jlnZKbZkYzNRZg%3D%3D

McBryce.

Non-stocked and crazy minimum order amount (3000 pieces), maybe the higher cost too. Axial are often on back order, I'm still waiting for some Kemet PEG caps for my HP 6920B, ordered back in March from Mouser.

David

[McBryce]

Here's a picture of the radial to axial capacitor trick. This is in my HP 334A but I've done it a number of times and it works great.
You get the latest capacitor types in a neat radial shape.

That's not exactly an unusual type or value. Why didn't you just order axial versions? Like this: https://www.mouser.de/ProductDetail/EPCOS-TDK/B41689K7278Q001?qs=gvwIECp2jlnZKbZkYzNRZg%3D%3D

McBryce.

Non-stocked and crazy minimum order amount (3000 pieces), maybe the higher cost too. Axial are often on back order, I'm still waiting for some Kemet PEG caps for my HP 6920B, ordered back in March from Mouser.

David

You could try Reichelt instead. A bit more expensive, but no minimum orders and usually have most things in stock.

McBryce.

[TERRA Operative]

I didn't use axials as the development of better types has essentially ceased, also availability and cost is far less favourable than radial caps.

So by doing the thing I did to convert radial caps to axial meant I got much cheaper caps with the latest technology and chemistry and they were all in stock.

I can still get the now discontinued Nichicon VX series (and a small selection of Sprague Atom too) axial caps in Akihabara, but even those are a bit limited in value and some sizes are going out of stock.
I often use the VX series in old power supply sections of equipment (for example, the Korg guitar synth I was just working on).


As for tonights TE fun, I assembled the test PCB for the 3GHz Channel 3 option I designed for my HP 53132A counter.
After realising I accidentally ordered an old version of the PCB that required surgery to fix some shorts and make the final chosen components fit on the wrong footprints (because I'm an idiot...)... It is actually working!

I am still getting spurious counts when no input is applied (as I expected I would), so tomorrow I'll play with biasing the prescaler chip inputs to help prevent that.

As it stands, I can test to 1.05GHz using my Tek SG503 and SG504 levelled sinewave generators, and I get full operation from 15MHz up to 1.05GHz across the full input range of 0.5V to 5.5V peak-to‐peak (terminated with a 50ohm pass-through terminator at the counter input)
(HP specify max 5V P-P from 100MHz to 3GHz)

I'll need to figure out a way to get up to a bit above 3GHz (maybe using the tracking generator output from my Siglent VNA?) and run a bunch of tests once the prescaler bias is set, but so far it's looking good.


Pictures tomorrow, I'm off to sleep now....

[EE4all]

I think there are some who are more purists with restoring gear, and always try to use the originally designed parts. I am more in your camp, and would just adapt a newer type capacitor.

[TERRA Operative]

Yeah, I like to keep it original too, but functionality trumps looks for me.

[Robert763]

Missing SOT23 parts only have two connections, looking at the others they are marked A6, this seems to be BAS16 from Philips (now NXP), a switching diode, nothing fancy.
Only thing that concerns me about the Datamess, is what I can't see under all the ICs, on the plus side there are only two layers to the PCB.

Thanks, I will keep working on the Toshiba, once the screen is fixed I can look at the SSD CF card conversion and make it silent  , perfect for datalogging from TEA, without family complaining about fan noise of more modern machines.

David

The problem with the level of corrosion on the S3 is that it starts to get into the semiconductors along the lead / encapsulation interface. This can affect the long term reliabilty. While I like Dataman products (I have have a S4 and have had a Softy, S2 and S3) and they have some value I really don't think this one is worth any more effort.

Robert.

[factory]

I agree, it's more of an experiment, to try out cleaning methods that might be needed for something less common, or higher value, of course in those replacing jelly-bean ICs & other components wouldn't be a problem. For the Dataman S3 these parts including the display, ZIF socket & other bits would soon exceed the value of a good untested unit, will probably just wire it back together & see if it does anything, or is completely dead, as Vince commented.

The S4 seems far more in demand, someone even offering 4MB ram ICs for that one for near £40  , are they really that hard to find.
Someone has even made a replica of the Softy 1, over on the UK vintage radio forum.
https://theoddys.com/acorn/replica_boards/replica_softy_1/replica_softy_1.html

Think TERRA was after a HP crystal oven, some in the US, might be too expensive with shipping to Japan. 
https://www.ebay.com/itm/404579316631

David

[factory]

Non-stocked and crazy minimum order amount (3000 pieces), maybe the higher cost too. Axial are often on back order, I'm still waiting for some Kemet PEG caps for my HP 6920B, ordered back in March from Mouser.

David

You could try Reichelt instead. A bit more expensive, but no minimum orders and usually have most things in stock.

McBryce.

They ship nearly everywhere in the world except where I live  , will forgot about them and continue with Mouser.

I think there are some who are more purists with restoring gear, and always try to use the originally designed parts. I am more in your camp, and would just adapt a newer type capacitor.

Yeah, I like to keep it original too, but functionality trumps looks for me.

For the HP 6920B, seeing radial caps through the perforated covers bothers me, others I have no problem fitting what I have, or buying new parts.

I've also got a lot of NOS axial caps from the ex-work stock, something like 800 of 10uF at 40V for example, less of other values, these will get used before buying new parts. Some known leaking brands have been thrown out.
The older HP stuff I don't replace every cap anyway, I don't have enough free time or funds to do this, most are usually OK, the smallest diameter axial caps are usually always dead though.

The newer 80s & 90s stuff worry me more, almost all brands seems to have problems with peeing radial caps in this era of electronics, the early 90s Toshiba laptop was a perfect example, Nichicon, Elna, Panasonic, all leaking junk. Tek 2445x 2465x scopes are another example, where leakage is a major problem.

David

[EggertEnjoyer123]

I now have two more 5.5 digit multimeters.
Someone sold me two Fluke 8840s for $30 at the W6TRW swap meet. One of them had bad calibration data and shows error 29, which is related to calibration memory. I cleared all the calibration data and that seemed to fix it. The other one has a bad solder joint somewhere - the display doesn't work unless you wiggle one of the cables a few times. Other than that it works fine. Going to have to figure out how to remove the display assembly, which seems riveted
I think he had three more that he was selling... Maybe I'll get them next month.

I also picked up a HP 8406A comb generator for $20. Unfortunately that one has a lot of screws, so I'll take it to the lab and then I'll take it apart.

[W6EL]

Holy crap I think I saw you there, I remember someone walked by my table with two of these. You did well with those Fluke 8840 meters!

[W6EL]

And with that, I will now add my own TRW Swap Meet find!

Hot on the heels of buying an HP-3400A on ebay for $77+tax recently (in "it worked 20 years ago" condition), I came upon two HP-3400A meters at the TRW swap meet. They looked much newer than the one I just got, as they were not yellowed and had the modern "IEC" power connector. Serial numbers only a few hundred apart.

The seller wanted $30 each (not bad considering I've seen them sell around $300...) and for whatever reason I asked for $50 for the pair. She couldn't seem to get a hold of her husband on the phone so I just paid the $60. But a few minutes later, he showed up, and she made him agree and gave me back the $10. I felt ...almost... bad about it.

Got home with the meters. The switches needed a spray. One meter works very well, just a half dB off from what I think is correct. The other is 3dB down (suspicious indeed) and sometimes exhibits syndromes of neurological disorders (random meter movement here and there). Likely a bad cap on the meter movements. The 3dB down thing could be a bad contact on the step attenuator knob perhaps. I will see. The nuvistor board was also dislodged from its holders, so maybe this unit saw some action. There's a bit of dust too, although not much really.

But let me tell you. The good one (on the right in the photos), when I opened it up to clean it, it was SO clean. You could eat your dinner off it. Not a scratch anywhere, no dust, no sticky goo, capacitors looking NICE. Wires bent like some sociopath put it together with perfect right angles. You could not ask for a more clean example of HP construction. Or maybe I've been buying too much old yellow HP... I think with a quick cal it will be good to go. Even the foam was intact around the nuvistor board, although it did shed some when I touched it.

The swap meet had all sorts of great test gear. I also scored rolls of Kapton tape for $1.50 each, loads of heat shrink tubing (including some low-outgassing "flight" tubing, didn't ask where that came from), a GHz-friendly directional coupler ($10), probably more I'm forgetting. I saw some programmable HP power supplies and loads of scopes and analyzers, just too bad I already have that sort of thing otherwise I could have dumped even more money. Additionally, I sold my IC-736 (the subject of many videos on my youtube channel) for $350 and my Yeasu VHF HT ($15). Prices might seem low but I'm just here to have fun, what can I say.

Now, what to do with the meters... I could get all three that I suddenly poses all cleaned up and calibrated (as well as I can), and then pick one to keep and two to sell. I should easily make what I put in to them, and that's all I need to do.

Oh, and my HP-3490A arrived. After it blew out a RIFA cap (flames like a torch, loud noise, awful smell, you know it), it's working fine. I am considering adding a true RMS circuit of some kind or another inside it, because.. why have a 6 digit meter without actually reading RMS... We can do this, it's 2023, there's dozens of ways to measure RMS and it'll be fun to experiment. This thing has the most satisfying switches and rolling unit selector you've ever felt. Click!

[factory]

Some good finds at the swap meet, sometimes you do find stuff at low prices, they just want it gone quickly without the hassle of online selling & packing/shipping. The same can happen with collection in person online items, this happened a few weeks ago, went to get a 5340A & 183B and ended up with lots more. 

You will still need to replace the foam in those 3400A, it'll be doing nothing useful, as it will be hard & crumbly. You may find Spraque 109D wet tant caps on the chopper board, these sometimes leak & make a mess, when the rubber seal fails, replacement would be a solid tant, as wet tants are crazy expensive.

The 3490A looks good too, mine are slightly older, have the dotty displays, neither work & the plastic rails at the front are falling apart, never checked for RIFA/Kemet/Yageo madness metallized paper caps, but have found them in many other HP, Tek & others, replace on slight with something not made with paper.

David

[factory]

Oh, and my HP-3490A arrived. After it blew out a RIFA cap (flames like a torch, loud noise, awful smell, you know it), it's working fine. I am considering adding a true RMS circuit of some kind or another inside it, because.. why have a 6 digit meter without actually reading RMS... We can do this, it's 2023, there's dozens of ways to measure RMS and it'll be fun to experiment. This thing has the most satisfying switches and rolling unit selector you've ever felt. Click!

The earlier HP 3480 series with 3484A plug-in, did have a TRMS option (043), based on similar thermocouple to the 3403 series, strange they never offered this option for the 3490A. 

David