Test Equipment Anonymous (TEA) group therapy thread

[BU508A]

There are no colour codes on my scopes BNC inputs, only the fancy modern stuff has such a luxury ! 

'
Wouldn't they all have to be green?

You lost me... non-native here. For my defense even native Terra Operative didn't get it right away, so that makes me feel a little less stupid.

Now will someone please explain it to me....

CRTs of (Tek) scopes are mostly green (green traces) ...

 

<--- not a native speaker (english)

[Robert763]

Just incase anyone has not used a modern 'scope, the colour code on the input inputs and controls matches the couour of the trace on the screen.

 “and for those of you who are watching in black and white, the pink is next to the green.” Ted Lowe

[Vince]

There are no colour codes on my scopes BNC inputs, only the fancy modern stuff has such a luxury ! 

'
Wouldn't they all have to be green?

You lost me... non-native here. For my defense even native Terra Operative didn't get it right away, so that makes me feel a little less stupid.

Now will someone please explain it to me....

CRTs of (Tek) scopes are mostly green (green traces) ...

 

<--- not a native speaker (english)

Ohhhhh.... OK OK !!!  Thanks BU ! 

You are right, it was not a language barrier then.... it's just that I was not the same wavelength altogether ! 

W were talking about thye INput jacks... but the CRT is an OUTput device !
Brain was 100% trying to make snes of the green in the context of the input connectors.

So I was trying hard to come up with an explanation, and the best I could figure out was :

That Tek type 502A I got a week ago in my big haul, it's a bit rough on the outside and in the bottom left corner, the UHF connector and grounding post next to it, are completely covered in green corrosion !
so I though maybe he caught on that, and it was a way of implying I bu old scopes that are so crusty that their UHF connectors are all green with corrosion ! 

It was a tad far stretched hence why I asked for clarification ! 

[Vince]

Just in case anyone has not used a modern 'scope, the colour code on the input inputs and controls matches the colour of the trace on the screen.

 “and for those of you who are watching in black and white, the pink is next to the green.” Ted Lowe

Oh yes I remember we discussed this subject not too long ago, and it was a slightly heated discussion ! 

[tautech]

Just incase anyone has not used a modern 'scope, the colour code on the input inputs and controls matches the couour of the trace on the screen.

 “and for those of you who are watching in black and white, the pink is next to the green.” Ted Lowe

Except for those with a User colour palette so you can have the trace any colour you like.



Then Henry Ford proclaimed you could have his Model T in any colour you liked as long as it was black. 

[factory]

The converter boards for my 200kHz HP 5090A 5090B off air receivers arrived during the week,
[...]
hopefully Vince will be happy as it's not blue. 

David

I like you enough that I feel I owe you the truth.

I didn't even know violet existed ! .. but now that I am confronted with it, I have to make it clear : I hate it even much more than blue ! 

Main problems with blue is that it looks completely anachronistic in last century's electronics, which is what I deal with.
Also, blue reminds me of blue LEDs and backlight that I hate with a passion words can't even begin to describe.

After 20 years of blue, I see that it's now progressively coming of fashion now replaced with a new fashion, for white LED.
White looks much less ugly / shocking / damaging / stressful to my eyes, however it has one huge problem, not esthetics but functional... I find that every time I am confronted with consumer stuff that use while LED, like my work laptop and it's base station.... my brain simply does not register white. I simply don't notice if it's on or not. the fact that they make the white very dim and covered with a diffused doesn't help.
It seems like they engineered it 100% to look nice (to the marketing guys at least), but forgot it's main goal : be NOTICED and convey information to the user !

I hate blue LEDs too and cold white LEDs even more.

The other choice of colours for the PCB were green (seen too many like this at work), blue (don't like either), red (not sure), yellow (again not sure) and white or black (don't like as they make the traces hard to see).

Might try red or yellow next time. Would prefer clear, but this can only be done with no solder mask and I want the solder mask. 

You got some interesting routing on that board...


I'll be making a PCB order soon myself, gonna get blue for mine to match the original PCB's in the thing.

Can you tell "I don't know what I'm doing"  (to quote LMNC on youtube), the lack of ground fill was done on purpose, as some of the old HP stuff is done exactly like that (example below from the 5090B).



The via for the test point was added to have the track on the back, as I will be drilling the plated holes to fit the Oxley test pins I have. Couldn't figure out how to create a footprint for a drilled pad on one side only (with non-plated hole) and have it connected to tracks.

Forgot to add another 0.1uF cap for the 12V regulator too.

Kind of wish I had made the board a little larger (which seems to double the price) and used the 9mm/10mm length resistors, as I have loads of those.

David

[Vince]

ROCHAR Niksea DVM restoration





Finally resuming work on this cute little puppy, round #3  I think it is now.
On the agenda now is a defective Neek-c board. I thought it would be a walk in the park, relaxing, rewarding, my ass... this DVM is nothing but trouble from top to bottom   

If you want to provide moral support or psychiatric help, feel free to pay a visit :

https://www.eevblog.com/forum/repair/nixie-dvm-restoration-rochar-a-1335/msg4522280/#msg4522280

[TERRA Operative]

You got some interesting routing on that board...


I'll be making a PCB order soon myself, gonna get blue for mine to match the original PCB's in the thing.

Can you tell "I don't know what I'm doing"  (to quote LMNC on youtube), the lack of ground fill was done on purpose, as some of the old HP stuff is done exactly like that (example below from the 5090B).



The via for the test point was added to have the track on the back, as I will be drilling the plated holes to fit the Oxley test pins I have. Couldn't figure out how to create a footprint for a drilled pad on one side only (with non-plated hole) and have it connected to tracks.

Forgot to add another 0.1uF cap for the 12V regulator too.

Kind of wish I had made the board a little larger (which seems to double the price) and used the 9mm/10mm length resistors, as I have loads of those.

David

Eh, it's nothing a bit of a tweak to the traces can't fix for the next revision, if you can be bothered.. If it works, it works. I have the opposite problem where I'll sit for faaaar too long tweaking the traces bit by bit until they are 'just so', like the worlds shittiest and slowest autorouter.



Also, I too hate blue LED's, but the Japanese seem to love them. They even have them as factory options for the front and rear of their cars. It's so shit and so 2000's, but I guess Japan can be a bit behind with adoption of newer tech, or maybe it's just nationalistic pride.

[DH7DN]

Greetings comrades! 

A short teardown of a Soviet-era Wilkinson-type Power Divider/Power Splitter. There are three N-type connectors, one being the input (top, see pictures) and two outputs (bottom).

The front plate shows the designator: Зб-11 (in cyrillic script, easily mistaken as 36-11) or ZB-11 (in latin) and was presumeably manufactured in week month 02 of 1989 [thanks to Zoli for this correction].

All components of the power divider are sandwiched together and held together with screws. The plates, gaskets and traces are made of/covered with silver in order to improve conductivity/shielding and to reduce losses (please correct me if I'm wrong). The resistor between two traces is 100 Ohm and seems to be pressed onto the traces by the face plate. Its value corresponds to 2*Z_0 = 100 Ohms, so it's matched for Z_0 = 50 Ohm RF applications.

[Zoli]

...
The front plate shows the designator: Зб-11 (in cyrillic script) or ZB-11 (in latin) and was presumeably manufactured in week 02 of 1989.
...

Slight correction: in the soviet era, the electronics were dated as month.year, so the manufacturing period is February 1989.

[Peter_O]

Then Henry Ford proclaimed you could have his Model T in any colour you liked as long as it was black. 

That might be a solution, as any LED can simulate a black LED. This is a hidden function most people don't know.   

[McBryce]

Then Henry Ford proclaimed you could have his Model T in any colour you liked as long as it was black. 

That might be a solution, as any LED can simulate a black LED. This is a hidden function most people don't know.   

It probably requires something really complicated though, like a PWM signal with a duty cycle of 0%

McBryce.

[Robert763]

I've loads of Black LEDs


aroumd 900 nm 

[tautech]

Then Henry Ford proclaimed you could have his Model T in any colour you liked as long as it was black. 

That might be a solution, as any LED can simulate a black LED. This is a hidden function most people don't know.   

It will show just fine with an inverse screenshot print. 

[dl6lr]

That might be a solution, as any LED can simulate a black LED. This is a hidden function most people don't know.   

That's the reason I use LEDs for illuminating my darkroom 

[tggzzz]

Then Henry Ford proclaimed you could have his Model T in any colour you liked as long as it was black. 

That might be a solution, as any LED can simulate a black LED. This is a hidden function most people don't know.   

It was widely publicised in '78, in "The Hitchhiker's Guide to the Galaxy", viz:
Zaphod Beeblebrox: It’s the wild colour scheme that freaks me. I mean, when you try an’ operate one of these weird black controls which are labelled in black on a black background, a small black light lights up black to tell you you’ve done it. What is this? Some kind of intergalactic hyper-hearse?
https://clivebanks.co.uk/THHGTTG/THHGTTGradio5.htm

[themadhippy]

 

I too hate blue LED's, but the Japanese seem to love them. They even have them as factory options for the front and rear of their cars

As its japan  dont be surprised  if the blue turns out to be green

[TERRA Operative]

I too hate blue LED's, but the Japanese seem to love them. They even have them as factory options for the front and rear of their cars

As its japan  dont be surprised  if the blue turns out to be green

Ooooh, I have this conversation with my wife all the time.
They call the green light blue.. Both due to the relatively recent invention of the word for green (midori) and that the green light in a traffic light was originally a bluer shade of green that now.



Today I am a happy camper. I just successfully upgraded my TDS210 scope to a TDS220.
Just needed a couple GPIB commands and then a full calibration, so I used my PS5004 (TM5000) Precision Power Supply for the DC voltages needed, my SG503 Levelled Sine Wave Generator for the high frequencies and my HP 3325A Function Generator referenced to my 34461A 6.5 Digit Multimeter for the lowest frequencies that the SG503 couldn't get down to.

I got through the calibration on the third try (lol) and it's all looking good.
Was a slight pain as each time I messed up I had to return to the beginning and start all over again.....  Would have been nice if the Tek firmware guys made it possible to retry a cal step rather than have the scope nope out to square one, but whatever.


Anyway, I just upgraded my scope from 60MHz to 100MHz so I'm happy. 

[Vince]

Electronic Load

So really, I just don't know, I am baffled, defeated... I am running out of ideas here 

I just can't find any explanation other than defective op-amps... I just can't.

So I think I will just get a bunch of them, from several different manufacturer to try and avoid any bad batch problem...

I need to order some caps for my Vintage TE so I could order these opamaps at the same time.

Just checked Farnell.... oh no.... these LMV7701 cost are a no go sadly !!

5 Euros a pop, and not in stock in anyway, lead time 12 MONTHS ! 

Can't I replace that opamp with something else, more mundane, much, much cheaper, and actually in stock ?
Some jelly bean opamp ?  What are the special requirements that this opamp must meet that others can't  ?

From what I see... bandwidth is not an issue.
Supply voltage not either, 10V, surely lots of devices can work with 10V.
Input current ? Anything will be good enough for our trimmer driving it...
rail to rail ? Doesn't even need to be, given that the gate won't turn on before 2 or 3 volts anyway. And if the ouput can't reach 10V no big deal, the gate doesn't need that much to turn on the MOSFET anyway, according to the datasheet.

So I will just get... I don't know, I am not too clued... what opamp is trendy these days, and cheap, and avaiiable in 5 pin SOT23 package ?

I need to study op-amp land and write some notes so I can have an overview of the landscape of the most common op-amps, their main specs / reasons for using them.

Almost any RRIO(focus on input, down to 0) op-amp should do it; decent GBW(>2MHz), offset(<1mV) and supply voltage(>12V) should do it. A good choice would be the TLV271; double check the pin-out, since has reverse supply pin version.
Edit: I've chosed the LMP7701, because I've inherited(from lean transformation ) plenty of them.

Oy !!

Just received my Farnell order....  split in 3 packets, yet there is still one single item still missing so they will have to send another, fourth parfel just for that one ! 
Talk about ecology...

Anyway, so I replaced that opamp and hey presto it works ! 

So I was not stupid... I was indeed unlucky and both previous opamps were indeed faulty with that weird fault of sending the V+ rail internally to their inputs !
so I guess I should learn to trust myself a bit more maybe, and just believe the facts...

No problem anymore with the TLV271 opamps. The input is now driven properly by the trim pot, great 

However the load as a whole still has problems... I can't reach the full 5A current, can get only 3.8A or so.
Yes, I do have enough voltage to force 5A through that old FET despite it's high 1+ ohms Rdson.
Using 12V or so, the two outputs of the lap power supply in //.

A quick probing around later reveals that the current shunt amplifier is not amplifying enough.
I reduced it's amplification from x50 down to x25 by putting two 10K feddback resistors in  // rather than the single one on the original schematic, the other day. I checked the current shunt, still good, it drops the proper voltage. But the output of the amplifier is way less than it should be.
Maybe I need to replace that chip as well !   

Gimme a minute, stay tuned...




. Problem now appears to be the other opamp, the current shunt amplifier

[DC1MC]

I did a reasonable good deed for a TEA member, if it goes OK, if my karma cleans a bit I hope they will not find too much crap on my incoming MRT  , wish me luck !!!

Cheers,
DC1MC

Also I hope foe WW3 to not start after the shite in Poland, I didn't get my Geiger counter yet and by now all the good stuff is probably gone already.

[Vince]

Electronic Load


OK I replaced the other opamp , teh current shunfg amplifier.... guess what, I was right it was indeed bad too ! 
With the replacement TLV271 it now amplifies x25 as it should ! 

Wow, so I got really unlucky with those LMP7701....

BUT.. it's still not working !

.. but at least now I know the load circuitry is doing what it's supposed to be doing, now worries there.... problem is just my old, ancient MOSFET, it just won't spit out more than more than 3.5+ amp or something.  It wanders a bit between 3.5 and 4+ amps, depends on the mood and how hot it is I guess.

So in order to confirm this, I will disconnect that MOSFET and replace it with my hand made load with its x3 IRF 630 in  //, which we have already demonstrated can put out bags of current even under only 5Volts.

Stay tuned....

[factory]

Had a spare hour or two today, located the missing resistors for the HP 5090 198kHz converter board, also drilled & fitted the test pins. Hopefully the parts from Mouser will turn up this week and I can finish & test it.



Going to collect my lots of rusty/corroded* junk  from PP tomorrow.

*Hopefully not as bad as some items in this thread; https://www.eevblog.com/forum/testgear/pp-november-auction/

David

[Vince]

Electronic Load


OK I replaced the other opamp , the current shunt amplifier.... guess what, I was right it was indeed bad too ! 
With the replacement TLV271 it now amplifies x25 as it should ! 

Wow, so I got really unlucky with those LMP7701....

BUT.. it's still not working !

.. but at least now I know the load circuitry is doing what it's supposed to be doing, now worries there.... problem is just my old, ancient MOSFET, it just won't spit out more than more than 3.5+ amp or something.  It wanders a bit between 3.5 and 4+ amps, depends on the mood and how hot it is I guess.

So in order to confirm this, I will disconnect that MOSFET and replace it with my hand made load with its x3 IRF 630 in  //, which we have already demonstrated can put out bags of current even under only 5Volts.

Stay tuned....

OK replying to myself, did what I just said 5 minutes ago.

So desoldered the board from that old MOSFET / small heat sink, and wired it to my big ass Robert(tm) heat sink and its x3 more modern IRF 630.

As I hoped, it now works just fine !

I can get almost to 5A.  Got 4.65A IIRC.

However this time I know it's not because of tired MOSFETs because the opamp that drives the gate(s) puts out only 5.6V , not the full 10V supply, which means it's not even trying to turn the MOSTFETs on fully.. it's limited by what the other opamp, the current shunt amplifier, is telling it to do. The latter does its job still just fine, it's just clipping that's all, as it should.... I modified its gain to double it, which was supposed to give me ABOUT 5A... not spot on 5A...

So that's cool, I think this load adventure is now over. It works. If I want more current from it all I need is to further reduce the feedback resistor of the amplifier. Was 10K, I added another 10K in // to make that 5K. So I could make that 2.2K or something, if I wanted 10A. Now that I have received a bunch of 0603 resistors from 1K to 10K, E6 series (along with the E12 100R to 1K Zoli suggested for the gate), I can fine tune the gain of the amplifier to my specifications, to get the current I want.

So thanks Zoli ! 

It was a cool little adventure, learned a lot and now have an adjustable load that can take lots of current, a very useful tool added to the lab ! 

I learned :

- You can make a Robert(tm) quick and dirty load with only just a few bits and bobs lying around, for free, that works just fine !

- MOSTFETs can get tired and not spit out their rated current.... or maybe I did some ESD damage to it when handling it, I guess could be that as well....

- There exists modern opamps these days that work only at silly low supply voltages, like 3 or 5 volts !

- You can get very unlucky and have opamps do weird things like leak the ir supply rail to their inputs, then you replace it and it still does the same thing, yet it was indeed the two chips in a row that were bad... just bad luck.

- 0603 resistors and SOT25 packages are not that difficult to solder....and the SOT25 are actually more difficult than the resistors because they have even narrower terminals ! 
... yet after replacing a few of these opamps, somehow I am now already finding them quite easy and quick to solder, go figure ! I find the quickest and easiest method is to put a dab of solder on all the pads, then use the hot air station to solder it, surface tension automatically guides the chip perfectly into place. Then I finish the job by putting some liquid flux over the chip and quick touching up of every pin with the soldering iron, to make it all perfect and be 100% sure the joints are good / not dry.
Takes no time and gives nice looking joints.

- Did a stupid wiring mistake because it's late and I am tired, and I now know that my 12A 5V old Adtech PSU can put out 16A !!!

- That also showed me that my Metrix MX54 DMM  can measure and sustain 16A without blowing its fuse even though the meter range and the fuse, are only 10A.... good to know.


Thanks Zoli and Robert for this dummy load adventure, was a cool experience ! 

[TERRA Operative]

Moar upgrade hacks!

I got me a working Advantest R6142 DC volt/current generator for cheap (¥10,000) partly to have a working unit and partly to help me repair my faulty R6144.

So, after checking it powered up and gave the expected voltage outputs, I cracked it open and soldered the two cut wire links inside, one on the top logic board, and one on the bottom main board.

This is all it takes to convert an R6142 that tops out at 11.999V output to an R6144 that goes all the way up to 32V. 


So, time to calibrate it and take some measurements to hopefully narrow down the issues in my faulty unit....

[Specmaster]

Specmaster,
Back in September last year you mentioned that you had been able to get an improved scan of the DSA511/524 service manual.  I too have on of these adapters, actually a DSA524 and hope to get round to working on it.  Would you mind sharing the manual that you received?
Also, you mentioned you wanted some information on the front panel.  A while back I was able to contact a very helpful guy called Chris Wood who sent me some information which included an improved calibration routine and also a diagram that showed which pins on IC412 and IC416 you should see changing for each of the buttons and pots on the front panel. I could return the favour by sharing this if you think it would be useful.

Service manual emailed to you already as a compressed file because it is over 9MB in size and will require a free downloaded copy of WinRAR to decompress it.