Test Equipment Anonymous (TEA) group therapy thread

[ch_scr]

Well I can't see the fluff with a tester like this, especially with the price and UI 
The TL866II Pro "eeprom" programmer that also does flash, bios, etc. also has a chip tester as well for 300 different logic IC.

Totally different devices to support different use cases. For example, the first thing I looked up after reading your post were the RAM chips for a Tandy/Radio Shack Color Computer 1: 2104 (4Kx1) and 4116 (16Kx1). Both are supported by the Retro Chip Tester, neither by the TL866II Pro. I'm sure there will be other examples, because the focus of the products is different.

Yes, the UI leaves a lot to be desired, as I eluded to previously. Had I the free time available, I'd probably just make my own retro chip tester. It's not like I have that many retro computers, thankfully.

It can't test DRAM in general, yes. It can test some SRAM. You are right, how often do you need that depends on use case.
I thought more along the lines of testing logic IC's. But if that's your only concern, there are a lot of Arduino DRAM / SRAM testers out there, building one you still might end up at 100€ total for the TL866 and that, but IMHO that would be more favorable still than the other tester 

[nfmax]

Just recently, I've been playing with some PT100 temperature sensors and conditioning circuits for them. I got fed up with the shortcomings of my resistance decade box (1Ω resolution, over-cooked low value resistors) so resorted to eBay. Picked up a JJ Lloyd 5-decade box with 0.1Ω resolution, which has cleaned up nicely. A couple of resistors needed trimming down with a parallel R, but after that it is not bad at all. Well within 0.5% almost everywhere.

[Robert763]

Small added Rabbit Burrow  Getting square and flat.

Now I just need a better test indicator, smaller holder and or a larger surface plate and a .....Laser Interferometer. 

FTFY

[beanflying]

Small added Rabbit Burrow  Getting square and flat.

Now I just need a better test indicator, smaller holder and or a larger surface plate and a .....Laser Interferometer. 

FTFY

I am already in trouble enough as I had been re-watching some RobRenz content over the last few weeks https://www.youtube.com/user/ROBRENZ

[dl6lr]

Anyway - I got a magna cum laude and I am supposed to be satisfied with it. I can still publish some extra chapters on my webpages ... 

Congratulations from my side! 
Being an old fart, I have difficulties to understand everything but I am just reading it and I find it very interesting. Although I will never reach that level nor will I directly or indirectly profit from your research with my hobby.

Bernd

[ch_scr]

...

whereupon he's all like:

ch_scr  — Today at 6:05 PM
fuck
you know I have to make me one as well?
thanks I guess    

And he means from scratch out of his parts bins, just because... 

But he posted a vid showing how to do what I wanted with the nanoVNA; and damn if he didn't get me back: Next in my yoobToob queue was this little gem, which sucked up another half-hour of my life...
...

Found two reasonable "primary" source for info on the dg-mos fet probe:
T. Hirschbüchler Bachelor Thesis on the "Design of a low-cost 1 GHz Active Probe" with "conception, construction and evaluation"
https://content.instructables.com/ORIG/FQZ/1QZP/IPJTFO82/FQZ1QZPIPJTFO82.pdf
and the layout from elektor with the gimmick-cap on the pcb:
https://www.elektormagazine.com/files/magazine/2004/dolo/040108-PCB.pdf

[bd139]

Some more work on the HP 5381A. The thing uses a good old SN7446 as the display decoder and this thing appears to eat displays. The question is why. Well looking into it in detail of the design, the SN7446 is the digit driver and is open collector. The digit mux is done using some discrete generic PNP transistors. The whole thing is current limited by a single 47 ohm resistor per segment.



Throwing all this into LTspice results in this with a crude approximation of a 2.36v drop LED:



So what does this tell you? Well the standing current is 53mA through each of these segments! Even if pulsed that's a bit much for vintage LEDs so no wonder they were blowing up.

So what do we do here? Well I've ordered a bunch of new Kingbright ones which are significantly more efficient so I'm replacing all the displays and the 47 ohm resistors with 220 ohm ones. As the rest of the counter runs luke warm but the display board runs hotter than satan's bollocks, this should reduce the stress considerably!

Edit: considering bagging a Duratool desolder station for this one as I don't fancy having to braid all these displays out https://cpc.farnell.com/duratool/d00672/desoldering-station-uk-eu-plug/dp/SD01384

[Brumby]

Bean should be pleased........

I had a delivery today - SUNDAY!! - from Australia Post!!  This is unheard of.  We don't get Sunday deliveries unless they are absolutely overrun - which might happen around Christmas.  I am still bewildered.

However - down to business.

Arriving today was one HP 5328A counter - ex eBay.  While I would normally prefer picking such items up, the current Covid restrictions forbid such a trip - so Australia Post it was.  I sent the seller a message asking for it to be packed well ... and I must say, it was.  There was a piece of foam carved out to cradle the knobs on the front panel - a nice attention to detail.  All in all, it survived the journey totally unscathed.

Front panel looks pretty good (and it came with the rack ears) but there's a lot of blank space in the middle that's just begging to be populated.  The gauntlet is at your feet, Bean.  (This could be a dangerous challenge.)



SO ... true to the tradition of the EEVblog, I didn't turn it on, I took it apart!

Inside it was quite clean and tidy - with a lot of unpopulated space on the main PCB.  You can tell this board was destined for greater things - some under the 5328 badge - and others one might expect from such a layout.





Now ... where are they ....?

AHA!  Very easy to get to - and not as bad as some I've seen, but they will be given a retirement party soon enough.
 

Nice oscillator can....



There are also tell-tale signs there's been solder flung around in the vicinity (outside, thankfully) as I found little flakes of solder caught in a couple of corners.  Easily removed with some gentle assistance from a carefully wielded hobby knife.



I've done a quick reorg of my rack to house this new bit of kit (To be honest, I HAD to get this out of the way very quickly.  You should have seen "The Look" I got when I brought the box in.)

Fired it up - along with my 8656B - and had a bit of a play. 



Here, they've warmed up for 20 minutes.  I know my 8656B is well outside calibration and there is nothing on the 5328A - but a 6 digit match is not a bad start.

[bd139]

I had two 5328A's a looong time ago. Nice counters but absolute bastards to debug if they go wrong and are noisy as hell. I managed to piece together one working one from two dead ones. I had a DVM option as well but that was completely borked . But you got the OCXO there which is a win on its own!

Only surviving image of them...



Oh and what's that lurking on top? A 475 and an E3630A. Deja Vu here 

[Specmaster]

Some more work on the HP 5381A. The thing uses a good old SN7446 as the display decoder and this thing appears to eat displays. The question is why. Well looking into it in detail of the design, the SN7446 is the digit driver and is open collector. The digit mux is done using some discrete generic PNP transistors. The whole thing is current limited by a single 47 ohm resistor per segment.



Throwing all this into LTspice results in this with a crude approximation of a 2.36v drop LED:



So what does this tell you? Well the standing current is 53mA through each of these segments! Even if pulsed that's a bit much for vintage LEDs so no wonder they were blowing up.

So what do we do here? Well I've ordered a bunch of new Kingbright ones which are significantly more efficient so I'm replacing all the displays and the 47 ohm resistors with 220 ohm ones. As the rest of the counter runs luke warm but the display board runs hotter than satan's bollocks, this should reduce the stress considerably!

Edit: considering bagging a Duratool desolder station for this one as I don't fancy having to braid all these displays out https://cpc.farnell.com/duratool/d00672/desoldering-station-uk-eu-plug/dp/SD01384

Yeah, bag one of those desolder stations, you'll wonder why you stuck with braid for so long, honestly it is a good investment.

[ch_scr]

Oh and what's that lurking on top? A 475 and an E3630A. Deja Vu here 

So totalling it up, did you make or loose money buying and selling the same gear over and over again? 
Also  on getting a desoldering station.

[Saskia]

have an Atten GT 5150 https://www.atten.eu/product/1129431/atten-gt-5150-single-channel-high-end-intelligent-suction-de-solder-station sitting here, as well as a Hakko that I need pressurized air to run ...

But no time for such fiddledoodlerie right now. I'll look back into it after moving.

In other news, I am waiting for a Nakamichi external Dolby C to use with my Revox B77 ...
DHL sez its on its way.

[mnementh]

...

whereupon he's all like:

ch_scr  — Today at 6:05 PM
fuck
you know I have to make me one as well?
thanks I guess    

And he means from scratch out of his parts bins, just because... 

But he posted a vid showing how to do what I wanted with the nanoVNA; and damn if he didn't get me back: Next in my yoobToob queue was this little gem, which sucked up another half-hour of my life...
...

Found two reasonable "primary" source for info on the dg-mos fet probe:
T. Hirschbüchler Bachelor Thesis on the "Design of a low-cost 1 GHz Active Probe" with "conception, construction and evaluation"
https://content.instructables.com/ORIG/FQZ/1QZP/IPJTFO82/FQZ1QZPIPJTFO82.pdf
and the layout from elektor with the gimmick-cap on the pcb:
https://www.elektormagazine.com/files/magazine/2004/dolo/040108-PCB.pdf

That link is borked (at least for me); it may be behind a paywall. I've attached the .pdf below. 

Also, I actually found that article while doing my own research; however I didn't agree with two primary features of his construction technique:

1), his pogo pins are just too long. Everything I've read on the subject aside from his article indicates that ~5mm or less is the ideal length here.

And B) the pin spacing is just ridiculously too wide (yes, I understand the reasons) to actually be a useful probe for most of the kind of stuff you'll use it for.  The really good thing aboot the  60dbmcom probe is the big pads for soldering pins to; you can easily reconfigure to match almost any needed pin spacing.   



https://www.ebay.com/itm/174764316127

There's also this one; it supposedly has specs from the actual device (not just copied from the Elektor article), and I like that the total signal path is as short as possible; however, it was 2x the price and completely unreasonable delivery time. Also it does not use the gimmick capacitor and no pad for GND pin, which is part of the reason the Elektor design uses a dual-gate FET.

mnem
 

[mnementh]

Some more work on the HP 5381A. The thing uses a good old SN7446 as the display decoder and this thing appears to eat displays. The question is why. Well looking into it in detail of the design, the SN7446 is the digit driver and is open collector. The digit mux is done using some discrete generic PNP transistors. The whole thing is current limited by a single 47 ohm resistor per segment.



Throwing all this into LTspice results in this with a crude approximation of a 2.36v drop LED:



So what does this tell you? Well the standing current is 53mA through each of these segments! Even if pulsed that's a bit much for vintage LEDs so no wonder they were blowing up.

So what do we do here? Well I've ordered a bunch of new Kingbright ones which are significantly more efficient so I'm replacing all the displays and the 47 ohm resistors with 220 ohm ones. As the rest of the counter runs luke warm but the display board runs hotter than satan's bollocks, this should reduce the stress considerably!

Edit: considering bagging a Duratool desolder station for this one as I don't fancy having to braid all these displays out https://cpc.farnell.com/duratool/d00672/desoldering-station-uk-eu-plug/dp/SD01384

Well motherfucker... I actually suggested that you might need to replace those ballast resistors if you changed the color... and that really, they should be changed anyways (thinking carbon composite, etc)... which would've put you on that very track. 

But then I thought I was just getting too outre' in my random "stream of consciousness" and almost immediately edited it out for brevity's sake.   

mnem
Even when I'm right, I can find a way to be wrong...

[bd139]

Oh and what's that lurking on top? A 475 and an E3630A. Deja Vu here 

So totalling it up, did you make or loose money buying and selling the same gear over and over again? 
Also  on getting a desoldering station.

There's a nice story behind this. If I go back to 2015 when I started this again, I was actually financially completely up shit creek. Buying and selling TE was a good way of making low investments with good returns to keep me afloat while I sorted everything out and it was something I found interesting at the same time. It took through to 2018 to get properly sorted and I was running on a 200% profit overall on arbitrage and value-add repairs which is a pretty amazing ROI. Quite frankly TE saved my ass for about 3 years of my life. I do this for love now not for survival, plus if I don't do it the devil might make use of my idle hands 

[bd139]

Some more work on the HP 5381A. The thing uses a good old SN7446 as the display decoder and this thing appears to eat displays. The question is why. Well looking into it in detail of the design, the SN7446 is the digit driver and is open collector. The digit mux is done using some discrete generic PNP transistors. The whole thing is current limited by a single 47 ohm resistor per segment.



Throwing all this into LTspice results in this with a crude approximation of a 2.36v drop LED:



So what does this tell you? Well the standing current is 53mA through each of these segments! Even if pulsed that's a bit much for vintage LEDs so no wonder they were blowing up.

So what do we do here? Well I've ordered a bunch of new Kingbright ones which are significantly more efficient so I'm replacing all the displays and the 47 ohm resistors with 220 ohm ones. As the rest of the counter runs luke warm but the display board runs hotter than satan's bollocks, this should reduce the stress considerably!

Edit: considering bagging a Duratool desolder station for this one as I don't fancy having to braid all these displays out https://cpc.farnell.com/duratool/d00672/desoldering-station-uk-eu-plug/dp/SD01384

Well motherfucker... I actually suggested that you might need to replace those ballast resistors if you changed the color... and that really, they should be changed anyways (thinking carbon composite, etc)... which would've put you on that very track. 

But then I thought I was just getting too outre' in my random "stream of consciousness" and almost immediately edited it out for brevity's sake.   

mnem
Even when I'm right, I can find a way to be wrong...

Yeah good point. I did actually read that post and didn't put two and two together until you just mentioned it. I shall go and sit in the corner with my dunce hat on. You can scowl at me at your pleasure 

[mnementh]

LOL.... if I'd left my post alone so it was just staring you in the face, you'd have got there in no time. I'm still at least partly guilty. 

mnem
 

[mnementh]

Edit: considering bagging a Duratool desolder station for this one as I don't fancy having to braid all these displays out https://cpc.farnell.com/duratool/d00672/desoldering-station-uk-eu-plug/dp/SD01384

Yeah, bag one of those desolder stations, you'll wonder why you stuck with braid for so long, honestly it is a good investment.

The Duratool uses the same guts as my Aouyue 2702; tho the handset is different. On the 2702, it is definitely high-maintenance; sometimes requiring disassembly/cleaning after only a couple-three desolderings. We spoke aboot using "alternate" filter materials in the handset and hose a while back... if that one Saskia offered can be had at a price you can live with, I'd think aboot going with it both for you having a better tool and for giving her the business.

mnem
Well, not giving her the business that way.... jeez, you pervs...

[med6753]

Lots of interesting TE today. Keep it coming. 

No TE for me today, got other things going on. Those of you interested in the Type 547 saga I'll be back at it tomorrow.

Oh and started initial buy of #2 one inch pine for the new scope cart. One 10 inch by 6 foot, two 6 inch by 4 foot, and two 2 inch by 6 foot boards cost $43 USD.    The KY didn't help. Ouch. 

[mnementh]

...I do this for love now not for survival, plus if I don't do it the devil might make use of my idle hands 

So that's my problem... 

mnem

[Specmaster]

...

whereupon he's all like:

ch_scr  — Today at 6:05 PM
fuck
you know I have to make me one as well?
thanks I guess    

And he means from scratch out of his parts bins, just because... 

But he posted a vid showing how to do what I wanted with the nanoVNA; and damn if he didn't get me back: Next in my yoobToob queue was this little gem, which sucked up another half-hour of my life...
...

Found two reasonable "primary" source for info on the dg-mos fet probe:
T. Hirschbüchler Bachelor Thesis on the "Design of a low-cost 1 GHz Active Probe" with "conception, construction and evaluation"
https://content.instructables.com/ORIG/FQZ/1QZP/IPJTFO82/FQZ1QZPIPJTFO82.pdf
and the layout from elektor with the gimmick-cap on the pcb:
https://www.elektormagazine.com/files/magazine/2004/dolo/040108-PCB.pdf

That link is borked (at least for me); it may be behind a paywall. I've attached the .pdf below. 

Also, I actually found that article while doing my own research; however I didn't agree with two primary features of his construction technique:

1), his pogo pins are just too long. Everything I've read on the subject aside from his article indicates that ~5mm or less is the ideal length here.

And B) the pin spacing is just ridiculously too wide (yes, I understand the reasons) to actually be a useful probe for most of the kind of stuff you'll use it for.  The really good thing aboot the  60dbmcom probe is the big pads for soldering pins to; you can easily reconfigure to match almost any needed pin spacing.   



https://www.ebay.com/itm/174764316127

There's also this one; it supposedly has specs from the actual device (not just copied from the Elektor article), and I like that the total signal path is as short as possible; however, it was 2x the price and completely unreasonable delivery time. Also it does not use the gimmick capacitor and no pad for GND pin, which is part of the reason the Elektor design uses a dual-gate FET.

mnem
 

Surely such a probe should be in a shielded body of some sort?

[tggzzz]

I don't think I mentioned this here, but those in the UK might be interested in:

TEKTRONIX 561A Vintage Oscilloscope - Fantastic Condition, Shenley, Hertfordshire £100.00 Collection only at Shenley Herts (J22 on M25)
https://www.gumtree.com/p/hobbies-collectibles/tektronix-561a-vintage-oscilloscope-fantastic-condition/1411056157

The photos do look good.

[bd139]

That's about 45 mins from here. Nope not got the room! 

[Specmaster]

That's about 45 mins from here. Nope not got the room! 

And you don't want or need the hernia that comes with it, even if it is free 

[med6753]

I don't think I mentioned this here, but those in the UK might be interested in:

TEKTRONIX 561A Vintage Oscilloscope - Fantastic Condition, Shenley, Hertfordshire £100.00 Collection only at Shenley Herts (J22 on M25)
https://www.gumtree.com/p/hobbies-collectibles/tektronix-561a-vintage-oscilloscope-fantastic-condition/1411056157

The photos do look good.

Wow. That thing is cherry.  The Type 561A was all vacuum tube PSU compared to my Type 561B which is completely solid state.