Test Equipment Anonymous (TEA) group therapy thread

[factory]

A little distraction from Ebay. I never go there, instead I just search Leboncoin.fr. It's like walking around a giant flea market. I like being surprised at what I might find whenever I visit. And before X-max I found this :

https://www.leboncoin.fr/ad/equipements_industriels/2897609579

I have enough hollow-state Tek scopes here, so a couple years ago or so, I started changing the goal, and now enjoy going after samller TE of that same era, that accompanied these scopes. Like special / uncommon plug-ins, special TE to calibrate this or that in the scope. Makes for a more vibrant and interesting "ecosystem", than just having bare scopes, doesn't it ? Yes it does, to me at least.

So in this vein I have managed to find so far :

- A type 107 fast pulse generator, to check B/W and frequency response of vertical amps.
- A type 1121 wide band pre-amp
- A type 180A time mark / comb generator to calibrate the horizontal amplifier of the scope

and today... this.... thing, which I can't even tell you what "type" it is because it doesn't even have one ! 
It's purely a calibration fixture, and often Tek didn't give them a type number, but instead a long sequence of digits that looks very much like the part-number of a... single part, not a complete instrument.

This thing they call a "Amplitude calibrator and comparator".
Looked in the manual, it's designed solely to fine tune the amplitude of the calibrator output of the scope. The scope already has a trimmer to fine tune this, but I guess with a calibrator it's more... accurate. I don't know, I just found this thing cool and cute, so I could not resist buying it. I am very excited to welcome the fourth member of my debuting collection of 500 scope "peripherals" so to speak 

Nice find, correct me if I'm wrong, but doesn't the whole thing slide out of the optional case with the screw on the bottom left on the front panel.

I have another from this series of Tek scope test/cal items, the 067-0532-00, a 500MHz constant amplitude calibrator, really need to give it a service one day.

David

[factory]

Had a bit of trouble with the ESR tester over the holiday, having spent too much recently I'm looking for 2mm connectors to repair it, as I can't seem to get the socket out of the plastic sleeve.
It's a gold plated version of Electro PJP part no. 209, I can only find nickel plated ones in the UK, I can probably cut the old plastic sleeve off, reuse the gold plated socket & use the sleeve from a new Ni plated version.



But thinking about it I would like to make a set of leads with bigger croc clips, does it matter if they have fancy gold plating or not for the 2mm connector? I can't say I've ever seen a gold plated capacitor lead. I already have some fancy Mueller croc clips.

Some pictures of insides of the ESR60 tester below, Rev 2.0B with R4.9 firmware.

David

[Martin.M]

S51B

tubes are very good.

[Robert763]

So I broke my ebay drought. Pulled the trigger on a Tektronix THS720A. Not super cheap at £240 but a late one with FW 1.16 shown working and all the bits including the proper CAT II probes. Looks fairly clean. Main reason for purchase is to upgrade to THS720P because I want something for measuring power harmonics. The 720P is not the most accurate for harmonics but it does work with 400Hz supplies which is of interest to me.
We will see what turns up.
I've had a THS720A before when they were a current product and the upgrade was unknown. I sold it o twhen prices were much higher.

Robert.

[factory]

I ended up buying a THS720A a while ago too, also a late one, still in the original box and only had around 5 power cycles when I bought it.

The end of last year I bought a faulty TDS2014, a recap of the PSU was done and a shorted capacitor array was found on the mainboard, all working now.

And I've sorted the Peak test lead, a small hammer & parallel punch was the answer to getting the socket out of the sleeve, plus a piece of metal with a hole about the same size as the sleeve, it's now soldered back together and back in service.

I'm currently working on an Advance counter from the mid to late 60s, many knackered capacitors found and the odd rusted diode/transistor lead so far. At least the PCBs are much better quality than other British makers, i.e. traces don't peel off if you look at them the wrong way.

David

[Vince]

Nice find, correct me if I'm wrong, but doesn't the whole thing slide out of the optional case with the screw on the bottom left on the front panel.

David

 
Oh.... nooooooooo..... you are perfectly right, it just slides out indeed! There was no need to destroy this cabinet for the  seller to take pictures ! 

To my defense (and his), I have zero knowledge of these calibration fixtures. There was no reason to believe it was a plug-in... for what purpose ? Why was the cabinet optional ? What are we supposed to do with a bare instrument, with all of its guts accessible for them to be damaged, or the operator to get zapped ?

Anyway, you live and learn. At least it made working the thing much easier for sure 

OK, been working on the thing all day and late at night yesterday. There isn't enough material for me to warrant creating a dedicated topic for this thing, so I will just report here as concisely as possible and not bother you with it any more...


CABINET

Spent 2 hours to put it back in one piece... not familiar with this thing. Spent most of it trying to fit a square peg into a round hole so to speak, because as I eventually found out, the seller had dismantled the cabinet even further than what I could see, then tried to partially re-assemble it. Turns out he fitted a part the wrong way around, which made it impossible to put the thing together again.
Once I had figured it all out, realised I was missing a few nuts and screws. Managed to pinch a few from an old Tek scope, not a perfect match but good enough. However overall, in the end I am still missing one screw, and this one I will never find in my Tek scopes as it's not a common screw. It's one that secures the feet at the bottom. Theses screws are very long and have a very tall and cylindrical head.
Luckily the concerned foot is secured well enough with just one of its two screws. So not a show-stopper for now.
The exploded view in the manual, and a couple pics in the Tekwiki, did help some.

I was a bit intrigued by the fact that the cabinet came with TWO tilting bails ? There is a short one and a tall one. I am confused... a bail at the front to raise / angle it towards you, fine.. but why also prop the butt of the instrument ? So my theory is that the short bail at the rear, you are not supposed to use... I think you just fit the bail of the height you prefer, at the front, and the one you don't use you just "store" it at the rear, keeping it folded / retracted at all times. I don't know... that just intrigued me.

CLEANING

Those yellow stickers were a pain, took quite a while to fully remove the glue residue but other than that it was not that dirty. It came out pretty nice to me, I very much enjoy looking at it up close and playing with it 



POWER CORD

I went to replace the hard wired power cord with a Euro one, but I changed my mind mid-way when I realised that I could do much better than that : get rid of the cord altogether and fit a socket instead : I realised that this blue part below, that holds the cable strain relief piece, is actually more of a "plug" that fits where a proper socket would go !



Why did they hard-wire the cord instead of just fitting a socket, when the metal work /chassis was already meant for a socket ?! Madness...
Then I had a X-mas moment when I realised that a modern IEC socket actually fits the old american cut out ... PERFECTLY ! It's literally a drop-in replacement. The squarish shape of the IEC socket fits e-axc-tly into the american round cutout. it's neither lose nor tight, it's perfect. Same for the mounting holes. exactly the same position. I couldn't believe my eyes. It's too good to be true of course, so I guess whenever the IEC introduced these more modern sockets, they purposefully made them mechanically backward compatible with the american sockets. Genius...





So that means I can retrofit all my Tek scopes, marvelous !   
Plus, that means the conversion requires absolutely zero modification or destruction, it is 100% reversible should that be desired later down the line for whatever reason ! 

That really made my day.
The original screws were therefore too long for the new IEC socket, so I put a piece of heat shrink tubing around the now protruding screws, to keep them 100% safe from the live terminals of the socket.  Yes, I only had green tubing of the required diameter, sorry for the eye sore... 





QUESTION : as you can see (and in accordance with the schematic), the earth wire from the socket / cord, is NOT connected straight to the chassis. Instead, it's connected to it VIA a pair of diodes in //, front to back. Never seen that before, what's the reason for this ? To me it's just a safety hazard... imagine if just one of the two diodes goes open-circuit for whatever reason. Though unlikely of course, it's still less reliable than not having diodes there. So if just one diode is open-circuit... for half the sine-wave, the earth connection is GONE and you could get zapped.     Of course if they did it, I am sure they had to, for some reason... just curious what this reason actually is !  I guess it was a requirement to achieve the performance / accuracy of the instrument ? Maybe in the very low voltage settings, the earth was introducing noise that was unmanageable and it was impossible to achieve the desired accuracy ?
Just a shot in the dark... I have no clue whatsoever !

So just in case, I checked the diodes, they are fine.

FUSE HOLDER

Was missing. Sadly I had no perfect replacement. Have a dozen salvaged ones but only one model with the suitable outside diameter. The flat on it is too small compared to the flat in the chassis, so it wiggled quite a bit. Tightening the large retaining nut fixed that, but for how long... luckily one is not supposed to mess with the fuse holder all day long eh...





TRANSFORMER

I checked its wiring. Luckily it was wired for 240V not 120V, despite the US power cord. There was no sign of rework on the transformer pins either. So it clearly was wired for 240V at the factory. Which is strange... so maybe that US plug was not so US after all ? It looks like one, but... scaled down. Maybe it was not american then, maybe it was from some European country that had this style of plug back in the day, and the instrument later. Whatever.... the point is, my biggest worry was now cleared : the prospect of having a fried transformer and fried internals... was now going away fast ! 
I measured the primary winding and it read 36 ohms or something, which was further encouraging for sure.


TESTING

OK so it was looking good now eh ? Cleaned, re-assembled, getting power to it is now sorted, and transformer is wired properly and in good shape.  So time to give it a test ride  ! 

At first I used the dim bulb tester just in case, with just one bulb since it's a low power instrument. Bulb didn't light up at all and no smoke nor fire-works to be seen. So I got rid of the bulb tester.

The pilot light on the front panel does light up but can only be seen if you look straiiiight at it. If you look at any angle, it looks like it's not lit. Normal ? Or maybe a weak bulb ?

Then I heard a little buzzing sound... yeahhhh, the electro-mechanical chopper thingy is alive, what a relief ! 

Then I scoped the output and yeah, a nice clean square wave, it's alive !!! 






Frequency is 1kHz as it should.  Scope measures it at spot on 1.00kHz which is  a miracle given that the manual states the tolerance is 20% !  I guess Tek likes to undersell and over deliver...
I went through all the ranges, and it looked accurate to me on the scope.
It basically works just fine...
But is it as accurate as it's supposed to be ? The whole point of this fixture is to be accurate, so it's important to check if it actually is...

The front panel allows you to output DC voltage rather than a square wave. This means I can measure it with a DMM, much more accurately than the scope. So I did that, and here are the results (I let it warm up for 15+ minutes). I don't have a fancy super accurate bench meter like all of you have.. for now I make do with my best handheld DMM, a Metrix MX56.  0.025% +/- 1LSD , 50,000 counts.





As we can see, the very worst we get, is 0.24%. What are the specs for this instrument ? It's 0.25%  !
Yeah, we have a winner ! 

0.24% is spot on an order of magnitude worse than my MX 56, which means it's likely that all my measurements are actually truthful (enough).

But what does 0.24% mean in practice ? Well that fixture is meant solely to calibrate a scope calibrator, so let's put it into a valid context : a user expanding his signal vertically to fully fit the graticule. Other than the 545A scope which had only 4 vertical divisions, most 500 scopes had 6 divisions, that's 30 minor divisions. So 0.24% of that is...0.07 ! Much less than a tenth of a small division... I don't know about you, but the best my eyes can discern on a graticule, in the best of conditions (sharp thin trace, and my face glued to the screen), would be about a quarter of a small division. Smaller than that is getting difficult to me. I am sure there are super humans on this forum that can see a micron on the screen, but I am not one of these people....

Hell... the WORST... really worst would be a low B/W scope like a 502. These have a 10x10 graticule. So that's 50 minor divisions. Accuracy then falls from 0.072 to 0.12. Barely more than a tenth of a small division. I am sorry but I sure as hell can't discern that !

And that's really the worst, wooorst scenario ! In practice most of the time it will be more accurate than that.

So... I think that we have a winner here what do you think ?! 

I am glad I took that gamble, for 35 Euros + shipping  it was well worth it !!

[Vince]

POWER CORD

[...]



Why did they hard-wire the cord instead of just fitting a socket, when the metal work /chassis was already meant for a socket ?! Madness...

Ah, I might have the answer to my question here ! Was on the phone a minute ago with an old retired EE friend of mine... I told him about it and he said that here in Frog land in the '70s or thereabouts, legislation (local or Europe-wide I don't know) explicitly FORBID the use of detachable power cords ! So manufacturers of TE or any appliance, were forced to hard wire the power cord instead !    That explains why every '70s TE I own drives me nuts with this cord.. I hate it.

So it's all sussed out now. Tek made the cutout in the metal work to fit a socket for its domestic market, but when they shipped it to Frogland they removed the socket and soldered a cord instead, and disguised the resulting gaping hole with this blue "plug". Which further confirms that this unknown to me power cord / plug style was indeed not american at all... wrong assumption...

[Vince]

QUESTION : as you can see (and in accordance with the schematic), the earth wire from the socket / cord, is NOT connected straight to the chassis. Instead, it's connected to it VIA a pair of diodes in //, front to back. Never seen that before, what's the reason for this ? To me it's just a safety hazard... imagine if just one of the two diodes goes open-circuit for whatever reason. Though unlikely of course, it's still less reliable than not having diodes there. So if just one diode is open-circuit... for half the sine-wave, the earth connection is GONE and you could get zapped.     Of course if they did it, I am sure they had to, for some reason... just curious what this reason actually is !  I guess it was a requirement to achieve the performance / accuracy of the instrument ? Maybe in the very low voltage settings, the earth was introducing noise that was unmanageable and it was impossible to achieve the desired accuracy ?
Just a shot in the dark... I have no clue whatsoever !

RTFM !! 

Replying to my question and sharing in case someone else might be wondering as well...

The manual has a paragraph devoted to explaining this to us.
Looks like my guess was right... incredible luck :




I also found this paragraph which explains how critical it is that the load impedance be precisely 1Meg as the instrument is designed solely to feed a scope input.  My DMM of course is 10meg not 1Meg, so might redo all my measurements to see if that makes any difference. Logically I guess the voltage should drop a hair on all ranges. Most ranges are either spot on or on the low side of things, so it will only makes things worse I guess. So I might need to tweak the reference voltage just hair to brings things back up. We shall see...

[factory]

Nice find, correct me if I'm wrong, but doesn't the whole thing slide out of the optional case with the screw on the bottom left on the front panel.

David

 
Oh.... nooooooooo..... you are perfectly right, it just slides out indeed! There was no need to destroy this cabinet for the  seller to take pictures ! 

To my defense (and his), I have zero knowledge of these calibration fixtures. There was no reason to believe it was a plug-in... for what purpose ? Why was the cabinet optional ? What are we supposed to do with a bare instrument, with all of its guts accessible for them to be damaged, or the operator to get zapped ?

Anyway, you live and learn. At least it made working the thing much easier for sure 


So... I think that we have a winner here what do you think ?! 

I am glad I took that gamble, for 35 Euros + shipping  it was well worth it !!

I have to admit I don't know as much about these modular calibration fixtures, but have seen a choice of cases, including the one you have for one unit and a double width case for two units.

The constant amplitude signal generator is taller than the others, it has a simpler box case, I haven't got a picture showing the case.
Edit: I've taken a picture, the covers are completely different, as I said already it's taller than most of the cal fixtures.

You did very well with this one, seems to be a winner. 

David

[Vince]

[...] but have seen a choice of cases, including the one you have for one unit and a double width case for two units.
David

Oh wow... interesting, so it was indeed a plug-in system, they made modular bays for their fixtures !
Must be as rare as hen's teeth, imagine a collector trying to find a double-size bay ! 
Definitely not as common or well known as the usual TM500 / 5000 systems...

[factory]

[...] but have seen a choice of cases, including the one you have for one unit and a double width case for two units.
David

Oh wow... interesting, so it was indeed a plug-in system, they made modular bays for their fixtures !
Must be as rare as hen's teeth, imagine a collector trying to find a double-size bay ! 
Definitely not as common or well known as the usual TM500 / 5000 systems...

Cubdriver has a double width rack mount case for these modular units, might also explain why I have a one unit without any case;
https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/msg2715940/#msg2715940



Picture stolen from ebay, of the one with covers, instead of the fancy case.



David

[Vince]

Thanks for that. So Cub did meet that rare hen with teeth ! 

OK so Tek at least made covers for these fixtures, in case you didn't have a frame to put them in... 

[TERRA Operative]

My Type 284 Pulse Gen uses the same type of case, but half width.

https://w140.com/tekwiki/wiki/284

[Vince]

Half width, normal width, double width.. there is something for everyone ! 

Wow that fixture looks the business and what.... 70ps rise time ?!

I WANT ONE !!!!!

Oh boy if I can find one on Leboncoin.fr one day, I would be thrilled...

[44kgk1lkf6u]

70ps is rather common in this day and age.  There is the Leo Bodnar pulse generator.  I use an NB7VQ14M when I need a fast edge.

[Vince]

Yeah sure, but I meant, of course, it wasn't at all back in the day of this Tek fixture...

I see the Leonardo and the chip you mentionned both are about 30ps.
That's only a factor of two or better than what this Tek fixture managed 60 years prior.
I find it quite embarrassing actually... certainly nothing to brag about I would think.

I might order some of those chips you mentioned to see what I can do.
However I suspect that the actual rise time of this chip might be heavily influenced by the board layout and practical implementation overall.
So I would have no idea what rise time I am actually getting out of a complete, finished product, no ? Or am I being too pessimistic ?!....

Are there "reference" / proven designs and board layouts out there for this chip ?

Still want that Tek fixture. It's just so sexy, and unlike the Leonardo it does not require USB and a Windows computer I don't want, to configure it. The Tek has a built-in human interface so to speak, and only need mains. Much more practical, intuitive and straight forward to me...

The cheapest Leonardo is extremely, extremely compact though, it's  plus for sure. But at 150 Euros it's a lot of money for what it is... I guess you pay for the labour involved in testing each and every unit.

[tautech]

Yeah sure, but I meant, of course, it wasn't at all back in the day of this Tek fixture...

I see the Leonardo and the chip you mentionned both are about 30ps.
That's only a factor of two or better than what this Tek fixture managed 60 years prior.
I find it quite embarrassing actually... certainly nothing to brag about I would think.

I might order some of those chips you mentioned to see what I can do.
However I suspect that the actual rise time of this chip might be heavily influenced by the board layout and practical implementation overall.
So I would have no idea what rise time I am actually getting out of a complete, finished product, no ? Or am I being too pessimistic ?!....

Are there "reference" / proven designs and board layouts out there for this chip ?

Still want that Tek fixture. It's just so sexy, and unlike the Leonardo it does not require USB and a Windows computer I don't want, to configure it. The Tek has a built-in human interface so to speak, and only need mains. Much more practical, intuitive and straight forward to me...

The cheapest Leonardo is extremely, extremely compact though, it's  plus for sure. But at 150 Euros it's a lot of money for what it is... I guess you pay for the labour involved in testing each and every unit.

Unfortunately you are a bit late to the Leo Bodnar Pulser party as just a few months ago a dumb pulser from Leo was available for 50 quid.
They were rated at 10 MHz and with 40ps rise/fall times tested on a 20 GHz sampling scope.
Mine arrived with 28ps rise/fall screenshots.....

I was going to build one with a 7414 IC and had it all drawn up in Altium but knew it wouldn't every get into the 10's of ps.

[Vince]

Yes I did see it on their website, marked as "Obsolete". Was 40ps but more like100 Euros so still very expensive for what it is.
I guess there is not much on that minuscule board so probably someone will reverse engineer it and Gerber files will be available ?
Probably been already done, somewhere on the forum, I guess...

That being said, when you think about it even 30ps is still very slow to check fast-ish scopes.

My fastest scope is 3GHz, a TDDS694C, and if I get my math right that's about 117ps rise time.
So even 30ps is hardly negligible. you can still work out the rise time of the scope but it takes an extra step/calculation.
For all my other scopes though, the 42 of them... it's plenty good enough...

I wonder what kind of voodoo goes into making a pulser for say a 100GHz scope...

[factory]

The mention of 3DP at the end of last year, encouraged me to get some feet made for the snap together HP instruments from the 1970s, eg the 5300 series, 34740/50 series. 3311A use the same style of case.
I really hope they aren't as described on the packaging, we need the opposite. 



They will need some adjustment, for some unknown reason they are not symmetrical and one of the holes doesn't line up quite right, having more made isn't an option, these weren't cheap enough for that. Shouldn't be too difficult to adjust, just attack with a file & drill.





P.S. The design file is from this post; https://www.eevblog.com/forum/testgear/replacement-knobs-feet-and-fittings-for-test-equipment/msg2472369/#msg2472369

David

[44kgk1lkf6u]

I might order some of those chips you mentioned to see what I can do.
However I suspect that the actual rise time of this chip might be heavily influenced by the board layout and practical implementation overall.
So I would have no idea what rise time I am actually getting out of a complete, finished product, no ? Or am I being too pessimistic ?!....

Are there "reference" / proven designs and board layouts out there for this chip ?

Onsemi Motorola provides an evaluation board for NB7L14.  It is more expensive than the Bodnar thing.  I failed to find its design files.

I don't have a 3 GHz oscilloscope.  I only needed something fast enough for my 1 GHz oscilloscope.  I put the chip and an LVDS oscillator on a 2 layer PCB without the fancy controlled impedance stuff.  It only needs a -3.3 V power supply. Measuring on a 7.1 GHz spectrum analyzer showed that its knee frequency is likely faster than what the spectrum analyzer can measure.  It did not feel right to measure an oscilloscope, and then buy a sampling oscilloscope just to measure what measures it.  I did not optimize the board because the first version was already good enough.  I can upload the design if you want it.

[m k]

P.S. The design file is from this post; https://www.eevblog.com/forum/testgear/replacement-knobs-feet-and-fittings-for-test-equipment/msg2472369/#msg2472369

David

Online viewer I tried had no measurements, but the picture was pretty symmetrical.

Maybe somebody has a real CAD and can check the situation.

[TERRA Operative]

If the overall dimensions and angles are correct, I can fix it up.
Let me know if there are any other modifications you want made too.

Unfortunately I don't have any gear with that foot, but if you want a more accurate replica, get your vernier calipers out and I can make one while you measure for me.

[Vince]

I might order some of those chips you mentioned to see what I can do.
However I suspect that the actual rise time of this chip might be heavily influenced by the board layout and practical implementation overall.
So I would have no idea what rise time I am actually getting out of a complete, finished product, no ? Or am I being too pessimistic ?!....

Are there "reference" / proven designs and board layouts out there for this chip ?

[...] I put the chip and an LVDS oscillator on a 2 layer PCB without the fancy controlled impedance stuff.  It only needs a -3.3 V power supply. Measuring on a 7.1 GHz spectrum analyzer showed that its knee frequency is likely faster than what the spectrum analyzer can measure.  It did not feel right to measure an oscilloscope, and then buy a sampling oscilloscope just to measure what measures it.  I did not optimize the board because the first version was already good enough.

 I can upload the design if you want it.

Well yes why not ! Please do 

[Vince]

News from my dead HP 130C recently purchased for 20 Euros.

Seller forgot to include the obsolete HP power cord with the scope, even though he does have it.
He said he would send it to me in an envelope. 2 weeks no news, so I sent him a message yesterday.

Got a reply a minute ago :

"I regret selling it for so little, and I had to pay XX for the polystyrene and spend XX time. Would you consider adding an extra 10 Euros ??
As for the cable, I will send it to you but  by mail it would cost 5 Euros so I won't do that ".

I thought  I had seen it all on Leboncoin.fr, but I was wrong... I had not yet seen THAT ! 

Price and shipping were agreed, I paid... simple. But no, I should feel guilty and send him more money ?
And the cable, HE forgot to put it with the scope, not me ! Now he refuses to own up to it ?
So basically he is happy looking like a scum bag just to earn 10+5 Euros ?
Is that how much he prices his honour at ? That's pretty cheap !

I am flabbergasted....

Just as a matter of principle, I told him to fuck off.

I need an HP cable now... 

[timeandfrequency]

[...]
I need an HP cable now...

Is it for this TE ? https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/msg5752273/#msg5752273
Can you provide a picture or a link showing the male plug ?