Test Equipment Anonymous (TEA) group therapy thread

[tggzzz]

I suppose in this case the gain is that if you design it on the shittiest possible medium then it should work fine when I do the final build on FR4.

Ah, bless. Someone that hasn't become intimately acquainted with St Murphy.

With RF, I've seen designs frozen by capturing the component positions using a wax block, and telling the assemblers that the components go exactly there. No, I had nothing whatsoever to do with that!

Of course nothing like that ever occurs in software or semiconductor plants. Oh no, that couldn't possibly exist.

[BU508A]

Yeah have checked it already. They are good fortunately. I’ve got a bag of 1N270’s around somewhere which would work (poor man’s 1N34A  )

10 pieces for EUR 3.84 including shipping is not that expensive. Imho. 

https://www.ebay.de/itm/291935959284

[bd139]

I suppose in this case the gain is that if you design it on the shittiest possible medium then it should work fine when I do the final build on FR4.

Ah, bless. Someone that hasn't become intimately acquainted with St Murphy.

With RF, I've seen designs frozen by capturing the component positions using a wax block, and telling the assemblers that the components go exactly there. No, I had nothing whatsoever to do with that!

Of course nothing like that ever occurs in software or semiconductor plants. Oh no, that couldn't possibly exist.

Murphy and me have a different relationship than usual. I actively bait his arrival. For when he arrives and you send him back again, only then can the causal transition between broken and working be established. Call it a red green test if you like. I can’t sleep knowing what may happen. I can sleep knowing what has happened.

[tggzzz]

I suppose in this case the gain is that if you design it on the shittiest possible medium then it should work fine when I do the final build on FR4.

Ah, bless. Someone that hasn't become intimately acquainted with St Murphy.

With RF, I've seen designs frozen by capturing the component positions using a wax block, and telling the assemblers that the components go exactly there. No, I had nothing whatsoever to do with that!

Of course nothing like that ever occurs in software or semiconductor plants. Oh no, that couldn't possibly exist.

Murphy and me have a different relationship than usual. I actively bait his arrival. For when he arrives and you send him back again, only then can the causal transition between broken and working be established. Call it a red green test if you like. I can’t sleep knowing what may happen. I can sleep knowing what has happened.

Ah. The vaccination theory of development. Unproven benefits, in my experience.

[bd139]

Yeah have checked it already. They are good fortunately. I’ve got a bag of 1N270’s around somewhere which would work (poor man’s 1N34A  )

10 pieces for EUR 3.84 including shipping is not that expensive. Imho. 

https://www.ebay.de/itm/291935959284

Hmm. Buying 1N34A diodes is a bit of a shit show to be honest. Some sellers sell other cheap Ge diodes as equivalents. They don’t have the same steep curve as the proper ones and way too much capacitance. And then there’s the lot selling small Schottky diodes as 1N34A’s.  The popularity is defined by the ubiquity of the things in articles spanning back 50 years now I suppose. This has given them a holy status along with the 40673 dual gate MOSFETs and other devices. This in turn leads to supply being gobbled up by chancers trying to make a quick buck (like me  ) and prices increasing. Then attractive low cost options appear which really aren’t 1N34’s.

But 100% of the time you can throw a 1N270 in there. And those are cheap.

Another lore diode is the holy mixer diode, the HP 5082-2800. Which is exactly the same as an ST 1N5711 but commands a 20x price tag 

This is why I hate watching people spend on a specific part. On another forum I saw someone pay £30 for a basic TO3 pass transistor because it had a different number to 2n3055 written in it...

[bd139]

I suppose in this case the gain is that if you design it on the shittiest possible medium then it should work fine when I do the final build on FR4.

Ah, bless. Someone that hasn't become intimately acquainted with St Murphy.

With RF, I've seen designs frozen by capturing the component positions using a wax block, and telling the assemblers that the components go exactly there. No, I had nothing whatsoever to do with that!

Of course nothing like that ever occurs in software or semiconductor plants. Oh no, that couldn't possibly exist.

Murphy and me have a different relationship than usual. I actively bait his arrival. For when he arrives and you send him back again, only then can the causal transition between broken and working be established. Call it a red green test if you like. I can’t sleep knowing what may happen. I can sleep knowing what has happened.

Ah. The vaccination theory of development. Unproven benefits, in my experience.

Not quite. More keeping a tight failure modelling and analysis loop.

[med6753]

My 7586 nuvistors are floating around between different New Jersey locations after finally getting out of Wisconsin. USPS says delivery today but that's not happening. But they are FINALLY at the Newark Distribution and priors show I should have them tomorrow. But I would have had them a lot sooner if the seller hadn't sat on them for 2 days after generating the shipping label and delivery to USPS.

Yep, been spoiled by Amazon. 

[med6753]

Yeah have checked it already. They are good fortunately. I’ve got a bag of 1N270’s around somewhere which would work (poor man’s 1N34A  )

10 pieces for EUR 3.84 including shipping is not that expensive. Imho. 

https://www.ebay.de/itm/291935959284

Hmm. Buying 1N34A diodes is a bit of a shit show to be honest. Some sellers sell other cheap Ge diodes as equivalents. They don’t have the same steep curve as the proper ones and way too much capacitance. And then there’s the lot selling small Schottky diodes as 1N34A’s.  The popularity is defined by the ubiquity of the things in articles spanning back 50 years now I suppose. This has given them a holy status along with the 40673 dual gate MOSFETs and other devices. This in turn leads to supply being gobbled up by chancers trying to make a quick buck (like me  ) and prices increasing. Then attractive low cost options appear which really aren’t 1N34’s.

But 100% of the time you can throw a 1N270 in there. And those are cheap.

Another lore diode is the holy mixer diode, the HP 5082-2800. Which is exactly the same as an ST 1N5711 but commands a 20x price tag 

This is why I hate watching people spend on a specific part. On another forum I saw someone pay £30 for a basic TO3 pass transistor because it had a different number to 2n3055 written in it...

My stash is definitely 1N34A. Had them for at least 30 years. Just sitting. Lonely. 

[tggzzz]

On another forum I saw someone pay £30 for a basic TO3 pass transistor because it had a different number to 2n3055 written in it...

Apparently the newer ones have a sufficiently higher f_T that they can oscillate in circuits designed for the original parts.

[Mr. Scram]

Ah, bless. Someone that hasn't become intimately acquainted with St Murphy.

With RF, I've seen designs frozen by capturing the component positions using a wax block, and telling the assemblers that the components go exactly there. No, I had nothing whatsoever to do with that!

Of course nothing like that ever occurs in software or semiconductor plants. Oh no, that couldn't possibly exist.

I yelled "Copy Exactly!" before even seeing the link.

[tonyalbus]

i hope she never finds out about this machine...

[mnementh]

The 2N3904/6 (in array form) and the HFA3096 were exactly the two ends of the spectrum that I was thinking of. I was playing (in simulation only) with the HFAxxxx arrays for some active differential probe designs a while back - must get back to it at some point.

They are rather nice little ICs. However quite fragile as I found 

Well I won. The issue was decoupling (tggzzz will be laughing at this and the prototype layout) and missing zobel network which was causing it to squeg. After a little motorboating due to missing compensation capacitor I got the turd polished. Also found out even at way too loud a volume it'll quite happily drive an 8 ohm speaker with 3904/3906 output pair without getting dangerously warm so I swapped that in.

So schematic, a fairly standard discrete opamp. Feedback path / gain setting through R10/R9. C1 decreases bandwidth in the loop. C7 adjusts low end response. C6 adjusts high end response. Phase flips have no gain

   Prototype:   

(SNIP)

Actually sounds ok through that feeper. I will invest in this idea further. It's a shit load better than an LM386 anyway.

There's something wrong... I don't see the terminals for you to plug the key into your little practice tone generator here... 

How dooya suppose it stacks up compared to something readily available as a drop-in board like a PAM8610...? Class D, 0.1% THD @5W/Ch-8Ω w/12V applied ?

https://www.diodes.com/assets/Datasheets/PAM8610.pdf

   So fukkin' cheap it makes your teeth hurt: https://www.ebay.com/sch/i.html?_from=R40&_nkw=PAM8610&_sacat=0&_sop=15

mnem
 

[Cubdriver]

 

I was testing dummy loads I made up for work.  Did the first one on Friday; ran it for a few hours, everything peachy-keen.  Hooked #2 up today and cranked things up.  Working fine, resistors warming but not too much, all looks good.  Went to hit the head.  Returned a few minutes later to a cloud of smoke in the room and the wonderful olfactory treat of eau de Allen Bradley - one of the fecking pass transistors in the power supply apparently let go and barbecued the balance resistor.



I just got the manual from Dave at Artek, now the fun part begins - getting at the damned thing.  It seems to be related to a pilot bearing - the first part installed and then everything else is built up around it.   


This will be fun.  Wonder how many extra screws there will be when it goes back together. 


And I wonder how long the house will reek of barbecued resistor...

[0culus]

Geez, that is properly charred.

[Cubdriver]

Geez, that is properly charred.

Yeah, because it couldn't have let go during the several hours it was on while I was sitting two feet away from it - no, it had to wait till I was gone for a few minutes.   

 

-Pat

[mnementh]

Well, if you're gonna fukkit up, fukkit up good an' proper!!! 

mnem

[Cubdriver]

Well, if you're gonna fukkit up, fukkit up good an' proper!!! 

mnem

Oh, it fukked itself good and proper like - there was literally a cloud of smoke when I came back into the room!  I'm currently figuring out how to extract the pass transistors and their driver board from it - they're trapped under Option 001 that must first be removed.  Glad I got the manual because getting it apart is NOT intuitive and I would have taken a lot of stuff that didn't need to be removed off if I'd set off on this quest unguided.

-Pat

[mansaxel]

This is why I hate watching people spend on a specific part. On another forum I saw someone pay £30 for a basic TO3 pass transistor because it had a different number to 2n3055 written in it...

My Studer mixer has BC309b transistors, according to schematic. I don't. I only have BC550 and 560 in my stash. I reluctantly was prepared to substitute the suspect BC309b with a BC560, after reading up on it being possible (a bit more study than "it's the same symbol and the same casing shape" actually  ). My hesitation swiftly disappeared when I discovered a Siemens BC560 already there, and no, that was no bodge, that was the original from factory.

With some careful design (like building and making it work on a breadboard first  ) one can make many circuits pretty foolproof wrt component choice.  And the Studer 169 input module is the stuff of dreams for some recording audiophools (they perform all kinds of ritualistic mods to it.), so it apparently works well too. I can confirm, it is a good mixer, although a bit limited in features.

[med6753]

Geez, that is properly charred.

Yeah, because it couldn't have let go during the several hours it was on while I was sitting two feet away from it - no, it had to wait till I was gone for a few minutes.   

 

-Pat

For future reference, get a piss bucket. 

Seriously, this is exactly why I never leave burn-in's unattended. Although I do take piss breaks. Maybe I should get a piss bucket too. 

[Zucca]

On the other side I attended the Keysight webinars about FFT. Interesting stuff, I report here my question and how they answer it:

Q: Does the 10MHz input on the high end Scope improve the FFT performance?
A: Hello Zucca, Thank you for the question. I am not sure if this would provide any improvement, I will check with the factory and get back to you via email.

Further answer:

I assume you refer to the 10MHz common instrument clock, typically used to synchronise instruments and which often come from a high-precision source. I checked with the factory and they state that think this would have very little effect. You wouldn’t notice a few parts/million difference in frequency measurements. Johnnie Hancock will be presenting a virtual seminar in early October on “Understanding Oscilloscope Specifications”. One thing he plans to point out is that although the scope’s time-base accuracy specification is often touted, it is overrated. There are many other factors that go into determining timing accuracy that are much bigger than time-base accuracy. He states the same can be said for frequency measurements in the frequency domain.

If you are referring to the 20MHz filter used in the oscilloscope front-end (hardware bandwidth limit filter): then this will have an effect - if you are only interested in frequency components below 20 MHz, this will definitely help. It should also improve the scope’s noise floor.

[Zucca]

@Zucca

https://www.ebay.de/itm/Sony-TA-E86B-Vintage-Vorverstarker/174396217750?hash=item289ad31996:g:ObgAAOSwaU1fOQfA

fantastic preamp.
Immediate buy if in working condition.

You are playing the Eve role and I feel like Adam... 



I have to take a pause of meditations and decide what to do...
Thank you so much!

 

[Cubdriver]

Geez, that is properly charred.

Yeah, because it couldn't have let go during the several hours it was on while I was sitting two feet away from it - no, it had to wait till I was gone for a few minutes.   

 

-Pat

For future reference, get a piss bucket. 

Seriously, this is exactly why I never leave burn-in's unattended. Although I do take piss breaks. Maybe I should get a piss bucket too. 

I wasn't burning in the power supply, but letting the load heat up and monitoring it with a thermal camera.  I thought the power supply worked, as it always had prior.

And it sure as hell burned!  This is the BACK side of the board.  I think I'll change all six of the resistors, and will be spacing them off the board with lead kinks.

-Pat

[TorinoFermic]

Spara, I have found the root cause of poor readings in my HP 3478A. It was the Front/Rear Switch that has gone bust, only the rear works fine.

Shall I buy a new switch or a switch contact cleaner ?

[tonyalbus]

@cup  oops... that was an unwanted bbq....sorry man..looks terrorble...

[tonyalbus]

Spara, I have found the root cause of poor readings in my HP 3478A. It was the Front/Rear Switch that has gone bust, only the rear works fine.

Shall I buy a new switch or a switch contact cleaner ?

i used contact cleaner and its good sens after....
if it fais again i remove the switch or switch back to front...and only use front..