Test Equipment Anonymous (TEA) group therapy thread

[med6753]

I wish BD would get his Limey arse back on here. 

[nfmax]

Surely it can't be page 16034 already? 

Most recently used to find out how close to an aluminium chassis we could mount Coilcraft 2% tolerance air-cored SM inductors, without affecting the inductance value. Answer: surprisingly close!

[Neper]

A Plessey AVOMETER in Ghent, Belgium:

https://www.ebay.com/itm/303557366571

[Cerebus]

One day I'll buy a 34401A and will replace the LM399 with an ADR1399KHZ 
They are available at analog.com for 12.28 Dollars per piece (one has to order at least 18 pieces).

Or you can order a sample for free, as I did today. They only have themselves to blame, if the MOQ was smaller I've have bought some, but no way am I going to pay 18x$12.28 = $221.04 just to get one for evaluation and I'm too impatient to wait until next year sometime when they will become catalogue parts at Digikey, Mouser, or whoever (RS will probably get there in 2030).

[med6753]

Working on this mini 212 is like trying to stuff 5 pounds of shit into a 2 pound bag. Anyway, cleaned up the corrosion on the PSU/HV board. Replaced all the HV rectifiers and the ceramic load balancing capacitors. Results? Didn't fix anything. Trace is still bouncing around and going out of focus. But I do have a new problem. It won't power up on batteries alone, only if plugged into mains.  So something new to troubleshoot. What a PITA.   

Edit....fixed the battery problem. One of the battery connectors to the board was a pin off.

[Cerebus]

It is difficult to know whether popcorn noise is in the voltage source or meter's internal reference. A good way of eliminating the possibility of the meter's reference generating noise is to measure a "noiseless" voltage source such as a Weston standard cell. (I speculate that other types of chemical cell would also have little popcorn noise).

NIST have a paper on that - "Measurement of Voltage Noise in Chemical batteries" presented at the 1995 IEEE International Frequency Control Symposium.

The TL;DR is that voltage noise from a battery is almost entirely down to the Johnson noise of the cell's internal resistance. The paper is quite old, and the 'winner' that they pronounce from the calls they tested is an AA NiCd with a voltage noise of -194dBVrms/srqt(Hz) at 1 Hz (200 pV/sqrt(Hz)). Sadly they didn't test lead acid or NiFe, both cells with very low internal resistance, which one suspects might perform very well if the "Johnson noise of the cell's internal resistance" thesis holds up. Should that be the case, we probably all have a 12V super-ultra-low-noise voltage source to hand.

[Cerebus]

While there's the whiff of volt-nuttery in the air, I'll show what's on my screen today. Waveforms from some Verilog simulation from a WIP project:


(Click for full size to get a properly readable version)

The cognoscenti will be able to figure out my Dastardly Plan.

[Specmaster]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

[capt bullshot]

The cognoscenti will be able to figure out my Dastardly Plan.

You're trying to copycat the HP3458 ADC (and take over total world domination by selling your CE3459)?

[Cerebus]

The cognoscenti will be able to figure out my Dastardly Plan.

You're trying to copycat the HP3458 ADC (and take over total world domination by selling your CE3459)?

Something like that - not quite copycat but perhaps 'reimagine'. There's a lot more to producing something like the 3458A than just the ADC but I thought it would be fun to build a multi-slope ADC and play around with seeing how far you can go on cheap-arsing the parts and still get decent results. The core of the ADC is a couple of op amps, a comparator, a cheap small FPGA and some very cheap analogue switches - a £10-15 BOM if you're aiming at full 3458A capabilites, more like £7-8 if you're aiming at 34401 level performance. Op amps have come a long way since HP designed the 3458A and you can buy an op amp that's up to snuff for the integrator whereas they had to design a custom amp. Other things have changed too, they needed precise and stable slope setting resistors because they didn't have the compute power available to deal with calibrating them, with modern microcontrollers you can cope with uneven slope values and merely need stable resistors.

First pass in solder is either going to go to the very bottom end and use very ordinary precision SMD resistors to set the slope currents or possibly a statistical array á la Pickering built out of Vishay NOMCA resistor arrays.

[tautech]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

He graced us with his presence on Discord and TEA is just too much for him to fit in now BD140 is on the scene.
For now it's priorities.....teenage kids, new job and possible new partner....an hour on Discord is all we can expect for a bit.

[Specmaster]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

He graced us with his presence on Discord and TEA is just too much for him to fit in now BD140 is on the scene.
For now it's priorities.....teenage kids, new job and possible new partner....an hour on Discord is all we can expect for a bit.

Wow, did he say about BD140 then, I did know but was not aware that he had told anyone.

[tautech]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

He graced us with his presence on Discord and TEA is just too much for him to fit in now BD140 is on the scene.
For now it's priorities.....teenage kids, new job and possible new partner....an hour on Discord is all we can expect for a bit.

Wow, did he say about BD140 then, I did know but was not aware that he had told anyone.

Openly. Seems like an ongoing Sat night dinner date before Discord. You need honor us with your presence there again to catch up on the latest. 

[AVGresponding]

The cognoscenti will be able to figure out my Dastardly Plan.

You're trying to copycat the HP3458 ADC (and take over total world domination by selling your CE3459)?

Something like that - not quite copycat but perhaps 'reimagine'. There's a lot more to producing something like the 3458A than just the ADC but I thought it would be fun to build a multi-slope ADC and play around with seeing how far you can go on cheap-arsing the parts and still get decent results. The core of the ADC is a couple of op amps, a comparator, a cheap small FPGA and some very cheap analogue switches - a £10-15 BOM if you're aiming at full 3458A capabilites, more like £7-8 if you're aiming at 34401 level performance. Op amps have come a long way since HP designed the 3458A and you can buy an op amp that's up to snuff for the integrator whereas they had to design a custom amp. Other things have changed too, they needed precise and stable slope setting resistors because they didn't have the compute power available to deal with calibrating them, with modern microcontrollers you can cope with uneven slope values and merely need stable resistors.

First pass in solder is either going to go to the very bottom end and use very ordinary precision SMD resistors to set the slope currents or possibly a statistical array á la Pickering built out of Vishay NOMCA resistor arrays.

Hopefully when the microcontroller crashes, you'll get a nonsense value out so you can see that it needs resetting...

One day I'll buy a 34401A and will replace the LM399 with an ADR1399KHZ 
They are available at analog.com for 12.28 Dollars per piece (one has to order at least 18 pieces).

I wonder if the LM399 on the 34401a were hand pick ? if they were, replacing just by any regular ADR1399 might not be a great move.

The service manual makes no mention of whether it is hand picked or not, but given it's listed as an HPAK part rather than having both an HPAK and OEM number, I think it's a fair chance that it is, presumably after the aging process.

So you're probably right about it being a potentially ruinous move.

[AVGresponding]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

He graced us with his presence on Discord and TEA is just too much for him to fit in now BD140 is on the scene.
For now it's priorities.....teenage kids, new job and possible new partner....an hour on Discord is all we can expect for a bit.

Wow, did he say about BD140 then, I did know but was not aware that he had told anyone.

Openly. Seems like an ongoing Sat night dinner date before Discord. You need honor us with your presence there again to catch up on the latest. 

He's too busy repainting his entire house!

[Specmaster]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

He graced us with his presence on Discord and TEA is just too much for him to fit in now BD140 is on the scene.
For now it's priorities.....teenage kids, new job and possible new partner....an hour on Discord is all we can expect for a bit.

Wow, did he say about BD140 then, I did know but was not aware that he had told anyone.

Yeah, I keep trying to, but lately it seems that real life seems to get in the way, maybe this week I might get lucky.

[factory]

Possibly a nice restoration project if the seller takes a low-ball offer? https://www.ebay.de/itm/234276755883

McBryce.

Seriously doubt they will take any low-ball offers, they also seem to part out vintage TE to flog the glassware and cRapacitors separately.
This one looks suspiciously similar to those found in some Tek TE, for parts not working. 
https://www.ebay.de/itm/224682354681


And I'm fed up of seeing this murdered hp TE, month after month, even that has missing parts. 
https://www.ebay.de/itm/234245845670


David

[AVGresponding]

One day I'll buy a 34401A and will replace the LM399 with an ADR1399KHZ 
They are available at analog.com for 12.28 Dollars per piece (one has to order at least 18 pieces).

Now that you have that international VAT, wouldn't it be a shame to let it go to waste?
I'd guess if you post a thread in the Metrology section á la "I'll order 18 and look for 3 people to take 4,4,5 each in europe"
You might find a few kindred souls there... Please tell if you do, I'd take one 4pcs lot please 

I'd jump in for a lot of 4, too.

You can pencil me in for a couple as well, I'm always happy to buy fancy stuff to live forever in my parts organiser...

[Saskia]

packing ...

Organizing the move.
Hubby's next heart surgery is coming up  @End of November ...

[tggzzz]

And I'm fed up of seeing this murdered hp TE, month after month, even that has missing parts. 

Tee hee:

https://www2.ppauctions.com/lot/161599/lot-351

[factory]

One day I'll buy a 34401A and will replace the LM399 with an ADR1399KHZ 
They are available at analog.com for 12.28 Dollars per piece (one has to order at least 18 pieces).

Or you can order a sample for free, as I did today. They only have themselves to blame, if the MOQ was smaller I've have bought some, but no way am I going to pay 18x$12.28 = $221.04 just to get one for evaluation and I'm too impatient to wait until next year sometime when they will become catalogue parts at Digikey, Mouser, or whoever (RS will probably get there in 2030).

Do you have to part of a business to order samples, or need to buy other parts at the same time?

Sample Free of Charge Online

    Over 3,000 models available to sample free or charge on the web in over 50 countries around the world
    These products are shipped to directly to you – free of charge (including free shipping)
    The limitations include:
        Up to 2 pieces per product
        Up to 4 products per order
    For many products, you will have the option to buy additional if you reach your sample limit

David

[Cerebus]

The cognoscenti will be able to figure out my Dastardly Plan.

You're trying to copycat the HP3458 ADC (and take over total world domination by selling your CE3459)?

Something like that - not quite copycat but perhaps 'reimagine'. There's a lot more to producing something like the 3458A than just the ADC but I thought it would be fun to build a multi-slope ADC and play around with seeing how far you can go on cheap-arsing the parts and still get decent results. The core of the ADC is a couple of op amps, a comparator, a cheap small FPGA and some very cheap analogue switches - a £10-15 BOM if you're aiming at full 3458A capabilites, more like £7-8 if you're aiming at 34401 level performance. Op amps have come a long way since HP designed the 3458A and you can buy an op amp that's up to snuff for the integrator whereas they had to design a custom amp. Other things have changed too, they needed precise and stable slope setting resistors because they didn't have the compute power available to deal with calibrating them, with modern microcontrollers you can cope with uneven slope values and merely need stable resistors.

First pass in solder is either going to go to the very bottom end and use very ordinary precision SMD resistors to set the slope currents or possibly a statistical array á la Pickering built out of Vishay NOMCA resistor arrays.

Hopefully when the microcontroller crashes, you'll get a nonsense value out so you can see that it needs resetting...

I'm insulted! Microcontrollers I've programmed don't crash, the GPSDO has been running continuously for months and is only due to crash on the next leap second ('cause I haven't coded the bit that decodes the GPS datastream to get the date of the next leap second).

[Zucca]

3 - older antenna with VHF-only amplifier is drowning the IF stages so none of the higher frequencies, but breaking the core allows the DVT receiver to receive the un-amplified signals in the UHF band as if they were common-mode noise. This was a very common problem during the  early parts of the switchover from analog.

mnem
 

Holy shit the dwagon got the cake today.
Yes I have an older antenna but I have a VHF-UHF amp. Funny, if I take a DVBT2 decoder and hook it up in place of the internal TV decoder everything is okay. Not sure how to proceed. I leave the core disconnected for now and everybody is happy.  .

My best guess is the internal TV DVBT2 decoder has some special RF input stage... , or something is out of spec.
Can't say broken because a broken core do the trick.

I bend my knees to the RF Gods.

[med6753]

I wish BD would get his Limey arse back on here. 

I was thinking the same thing, its just so quiet without him, I must make contact with him and see just how he is doing.

He graced us with his presence on Discord and TEA is just too much for him to fit in now BD140 is on the scene.
For now it's priorities.....teenage kids, new job and possible new partner....an hour on Discord is all we can expect for a bit.

Wow, did he say about BD140 then, I did know but was not aware that he had told anyone.

Openly. Seems like an ongoing Sat night dinner date before Discord. You need honor us with your presence there again to catch up on the latest. 

Women are the devil. Another one bites the dust. 

[Saskia]

just solve the problem the pommijätkät way