Test Equipment Anonymous (TEA) group therapy thread

[med6753]

Gee....I wish all fixes were this easy. 7A26 plug-in acting out on channel 1. Of course 20mV and .2V ranges the worst (always seems to be the case).

Opened it up. Pulled the attenuator blocks. Careful application of Deoxit Gold G5 on visible leaf switches with a cotton swab. Short sprays on the back side. Insert and test. Perfect. 

How do you get them out without damaging those tiny wires with a ferrite bead that seems to connect the attenuator board(s) with the amplifier?

I suspect my 7A26 needs some deoxit as well.

[edit] I think you mean the metal covers on the top side? That say "caution delicate board".

Yep, the 7A26 has those VERY DELICATE wires on that ferrite bead connecting the attenuator board to the main board. Luckily I didn't have to mess with that. I pulled the metal cover off the channel 1 switch assembly then I removed the 4 attenuator blocks. The used a cotton swab soaked with Deoxit Gold G5 on the exposed leaf contacts and attenuator board. Then I removed the rear cover on the plug-in and several quick shots of Deoxit down into the switch assembly where the rear contacts reside.

Got lucky. It worked. Will it stay that way? Time will tell. But I was prepared to pull the attenuator board if required and do that "brute force" polishing method I used (and documented) on the 7A24 plug-in. Which BTW still works fine despite the nay sayers who said I would ruin the assembly.         

[Kosmic]

Ok I am 100% sold on the Hameg component tester / curve tracer now.

My current project ground to a halt miserably for another debugging session. Suspected 2n2219 failure. So stuff it on the chinese "transistor tester". Yeah looks fine. Still won't oscillate. Hmm. Stuff the Hameg across the B-E junction. Oh that's not right - should be a diode but apparently has turned into a resistor when reverse biased!



This is quickly becoming a very useful purchase.

I could not find a Hameg with a component tester but I started reading on Curve Tracer.

Apparently it's quite easy to put one together ghetto style. Juste need a transformer and a resistor 



Zener Diode:



10uF Capacitor:



More info:

[med6753]

Bd139....too bad about the 177. But at least you know EXACTLY what's wrong with it so it will be an excellent parts mule candidate.

[bd139]

Ok I am 100% sold on the Hameg component tester / curve tracer now.

My current project ground to a halt miserably for another debugging session. Suspected 2n2219 failure. So stuff it on the chinese "transistor tester". Yeah looks fine. Still won't oscillate. Hmm. Stuff the Hameg across the B-E junction. Oh that's not right - should be a diode but apparently has turned into a resistor when reverse biased!



This is quickly becoming a very useful purchase.

I could not find a Hameg with a component tester but I started reading on Curve Tracer.

Apparently it's quite easy to put one together ghetto style. Juste need a transformer and a resistor 

More info:

Yeah that’s what the hameg one is basically. It has a separate transformer tap for it.

Bd139....too bad about the 177. But at least you know EXACTLY what's wrong with it so it will be an excellent parts mule candidate.

Indeed. Will lurk for another one.

[mnementh]

Yeah can’t moan for £3 at all. An afternoon of debugging fun is a pretty cheap exercise.  I’d be annoyed if I’d spent £10 on it. 

I looked on the Caddock site just out of dumb curiosity; it appears they've stepped into the 21st century and don't even acknowledge the existence of products in DIP package. 

Would be an interesting exercise to see if you can re-create the network (and how close to "right" you can get) using precision SMT resistors though. 

mnem
 

[Kosmic]

Also I finally found something the Adalm2000 is actually good at. The XY mode is not that bad. Better than my Lecroy anyway.

I'm thinking of designing a bcp that will extend the Adalm2000 for this application. Maybe use the signal generator instead of a transfo.

[0culus]

Gee....I wish all fixes were this easy. 7A26 plug-in acting out on channel 1. Of course 20mV and .2V ranges the worst (always seems to be the case).

Opened it up. Pulled the attenuator blocks. Careful application of Deoxit Gold G5 on visible leaf switches with a cotton swab. Short sprays on the back side. Insert and test. Perfect. 

How do you get them out without damaging those tiny wires with a ferrite bead that seems to connect the attenuator board(s) with the amplifier?

I suspect my 7A26 needs some deoxit as well.

[edit] I think you mean the metal covers on the top side? That say "caution delicate board".

Yep, the 7A26 has those VERY DELICATE wires on that ferrite bead connecting the attenuator board to the main board. Luckily I didn't have to mess with that. I pulled the metal cover off the channel 1 switch assembly then I removed the 4 attenuator blocks. The used a cotton swab soaked with Deoxit Gold G5 on the exposed leaf contacts and attenuator board. Then I removed the rear cover on the plug-in and several quick shots of Deoxit down into the switch assembly where the rear contacts reside.

Got lucky. It worked. Will it stay that way? Time will tell. But I was prepared to pull the attenuator board if required and do that "brute force" polishing method I used (and documented) on the 7A24 plug-in. Which BTW still works fine despite the nay sayers who said I would ruin the assembly.       

OK, cool. I haven't been able to find anything online about deoxit G5 being unsafe for the special PCB material (polyphenylene oxide). How do the attenuator blocks come out? I'm afraid of breaking something. 

[Specmaster]

EEEEEEEEEEK!!! OUT OF COFFFEEE!!!   

None to brew... no instant left... not even the cheap mexican shit.

mnem
*chewing on a used coffee filter*

Time to start drinking TEA then 

[Specmaster]

Ok I am 100% sold on the Hameg component tester / curve tracer now.

My current project ground to a halt miserably for another debugging session. Suspected 2n2219 failure. So stuff it on the chinese "transistor tester". Yeah looks fine. Still won't oscillate. Hmm. Stuff the Hameg across the B-E junction. Oh that's not right - should be a diode but apparently has turned into a resistor when reverse biased!



This is quickly becoming a very useful purchase.

I could not find a Hameg with a component tester but I started reading on Curve Tracer.

Apparently it's quite easy to put one together ghetto style. Juste need a transformer and a resistor 

More info:

Yeah that’s what the hameg one is basically. It has a separate transformer tap for it.

Bd139....too bad about the 177. But at least you know EXACTLY what's wrong with it so it will be an excellent parts mule candidate.

Indeed. Will lurk for another one.

Yep, my first decent scope had one of those on it but it is still no match for dedicated testers, for instance, it will test a resistor and will present you with a straight line at an angle and the angle will determine the resistance value, but you cannot read that value, but a DMM does that perfectly, yes it will also test a capacitor, but so will many DMMs or dedicated LCR meters etc and will also give a value for them, so do I miss not having a component tester on my current scopes, not in the slightest bit 

[med6753]

Gee....I wish all fixes were this easy. 7A26 plug-in acting out on channel 1. Of course 20mV and .2V ranges the worst (always seems to be the case).

Opened it up. Pulled the attenuator blocks. Careful application of Deoxit Gold G5 on visible leaf switches with a cotton swab. Short sprays on the back side. Insert and test. Perfect. 

How do you get them out without damaging those tiny wires with a ferrite bead that seems to connect the attenuator board(s) with the amplifier?

I suspect my 7A26 needs some deoxit as well.

[edit] I think you mean the metal covers on the top side? That say "caution delicate board".

Yep, the 7A26 has those VERY DELICATE wires on that ferrite bead connecting the attenuator board to the main board. Luckily I didn't have to mess with that. I pulled the metal cover off the channel 1 switch assembly then I removed the 4 attenuator blocks. The used a cotton swab soaked with Deoxit Gold G5 on the exposed leaf contacts and attenuator board. Then I removed the rear cover on the plug-in and several quick shots of Deoxit down into the switch assembly where the rear contacts reside.

Got lucky. It worked. Will it stay that way? Time will tell. But I was prepared to pull the attenuator board if required and do that "brute force" polishing method I used (and documented) on the 7A24 plug-in. Which BTW still works fine despite the nay sayers who said I would ruin the assembly.       

OK, cool. I haven't been able to find anything online about deoxit G5 being unsafe for the special PCB material (polyphenylene oxide). How do the attenuator blocks come out? I'm afraid of breaking something. 

Pull straight up (carefully) with a pair of duck bill or wide needle nose pliers. They simply plug onto the attenuator board with 6 pins. And don't mix them up. Plug into the same location they came out from. 

[bd139]

Yeah can’t moan for £3 at all. An afternoon of debugging fun is a pretty cheap exercise.  I’d be annoyed if I’d spent £10 on it. 

I looked on the Caddock site just out of dumb curiosity; it appears they've stepped into the 21st century and don't even acknowledge the existence of products in DIP package. 

Would be an interesting exercise to see if you can re-create the network (and how close to "right" you can get) using precision SMT resistors though. 

mnem
 

That is definitely possible but it is entering the "i can't be arsed" territory  .

Yep, my first decent scope had one of those on it but it is still no match for dedicated testers, for instance, it will test a resistor and will present you with a straight line at an angle and the angle will determine the resistance value, but you cannot read that value, but a DMM does that perfectly, yes it will also test a capacitor, but so will many DMMs or dedicated LCR meters etc and will also give a value for them, so do I miss not having a component tester on my current scopes, not in the slightest bit 

I think both are important. In the case mentioned above it told me when a transistor turned into a resistor because I abused it 


Anyway onto item two which is a lot better. Got the heathkit VTVM up. Cleaned it up a bit and cracked it open and there was a mildly crusty battery in it. It didn't destroy the battery holder for once which was good. Insides are not particularly well constructed. 1959 dated electrolytic and stripey selenium rectifier of doom have been evicted and replaced with a cap / diode / resistor combination to bring the HT total to 150V. Turned out 1.2K and 1N4007 was about right after some experimentation on the breadboard. New cap was radial - that's all I had. I did use PCB pins though . Gave the switches a quick spray and it zeroed straight away and takes a DC reading almost on the mark. Was slightly surprised there.

To do: new mains lead and earth it, internal 1M resistor (sod the nice probe kit for this one!), replace the large waxy cap in it (on order RS 10x for £0.20 inc delivery!), apply some turd polish.

Inside:



Outside:



So total expenditure on that is £12.20. Pretty good.

[med6753]

EEEEEEEEEEK!!! OUT OF COFFFEEE!!!   

None to brew... no instant left... not even the cheap mexican shit.

mnem
*chewing on a used coffee filter*

Time to start drinking TEA then 

In this house that would be worse than contacting the bubonic plague. 

[med6753]

Speaking of Caddock thin film networks.....

I did build up a discrete network of 0.1% resistors to replace a defective one on a Fluke 8010A and it worked great....and still working.

I simply hand selected the resistors and carefully soldered them in place.

[bd139]

Yeah looking at this one it's definitely a possibility. The main divider network is ok which is the difficult one and the second one is just a bunch of close tolerance standard values. I reckon I can knock something up on JLCPCB and use 0.1% or better SMD resistors.

[Specmaster]

beaten again  going for a TTi 1906 bench meter and a Yamaha 45w stereo amp and piped at the post again  , back to the hunt again.

[worsthorse]

Yeah can’t moan for £3 at all. An afternoon of debugging fun is a pretty cheap exercise.  I’d be annoyed if I’d spent £10 on it. 

I looked on the Caddock site just out of dumb curiosity; it appears they've stepped into the 21st century and don't even acknowledge the existence of products in DIP package. 

Would be an interesting exercise to see if you can re-create the network (and how close to "right" you can get) using precision SMT resistors though. 

mnem
 

I was actually wondering if you could turn the 177 into a kim-1. 



The KIM was the first single board computer I learned to program. In a fit of nostalgia, I spent a little time looking for one.  Apparently I do not have the financial means to support revisiting my wistful recollections. 

[worsthorse]

Anyway onto item two which is a lot better. Got the heathkit VTVM up. Cleaned it up a bit and cracked it open and there was a mildly crusty battery in it. It didn't destroy the battery holder for once which was good. Insides are not particularly well constructed. 1959 dated electrolytic and stripey selenium rectifier of doom have been evicted and replaced with a cap / diode / resistor combination to bring the HT total to 150V. Turned out 1.2K and 1N4007 was about right after some experimentation on the breadboard. New cap was radial - that's all I had. I did use PCB pins though . Gave the switches a quick spray and it zeroed straight away and takes a DC reading almost on the mark. Was slightly surprised there.

To do: new mains lead and earth it, internal 1M resistor (sod the nice probe kit for this one!), replace the large waxy cap in it (on order RS 10x for £0.20 inc delivery!), apply some turd polish.

So total expenditure on that is £12.20. Pretty good.

bd, do you want a pair of fetrons to replace those tubes? i am pretty sure that VTVM has the right tubes for it and i have a couple of them around here...

[med6753]

Anyway onto item two which is a lot better. Got the heathkit VTVM up. Cleaned it up a bit and cracked it open and there was a mildly crusty battery in it. It didn't destroy the battery holder for once which was good. Insides are not particularly well constructed. 1959 dated electrolytic and stripey selenium rectifier of doom have been evicted and replaced with a cap / diode / resistor combination to bring the HT total to 150V. Turned out 1.2K and 1N4007 was about right after some experimentation on the breadboard. New cap was radial - that's all I had. I did use PCB pins though . Gave the switches a quick spray and it zeroed straight away and takes a DC reading almost on the mark. Was slightly surprised there.

To do: new mains lead and earth it, internal 1M resistor (sod the nice probe kit for this one!), replace the large waxy cap in it (on order RS 10x for £0.20 inc delivery!), apply some turd polish.

So total expenditure on that is £12.20. Pretty good.

bd, do you want a pair of fetrons to replace those tubes? i am pretty sure that VTVM has the right tubes for it and i have a couple of them around here...

Starting with the Heath V-7A in the mid 1950's and right up until they quit selling them in the late 1980's they all used a 12AU7 and a 6AL5. So the answer to your question is "yes" 
 

[bd139]

bd, do you want a pair of fetrons to replace those tubes? i am pretty sure that VTVM has the right tubes for it and i have a couple of them around here...

Thanks for the offer but will pass on this one at least. The guts of this thing are too badly constructed for me to let it hang around so it's getting turd polished and chucked on ebay.

I was actually wondering if you could turn the 177 into a kim-1. 

The KIM was the first single board computer I learned to program. In a fit of nostalgia, I spent a little time looking for one.  Apparently I do not have the financial means to support revisiting my wistful recollections. 

Now we're talking. Proper computers! This 177 is 6800 based rather than 6502 of the KIM-1 however. We had our own version of the KIM-1 here: https://en.wikipedia.org/wiki/Acorn_System_1 . I actually cut my teeth on 6502 assembly on the BBC micro which is the descendant of that, then built a Z80 SBC. I stole all the Z80 books from my local library. They probably have bounty hunters out for me now 

[tggzzz]

Yeah looking at this one it's definitely a possibility. The main divider network is ok which is the difficult one and the second one is just a bunch of close tolerance standard values. I reckon I can knock something up on JLCPCB and use 0.1% or better SMD resistors.

The issues will be thermal tracking and any residual stresses in the resistors either after soldering or due to differential expansion when the temperature changes.

Thermal tracking can be minimised by hillbilly engineering. I don't know the significance of the the stresses, but through hole will be better. For an extreme example, I refer you to

[tggzzz]

The KIM was the first single board computer I learned to program. In a fit of nostalgia, I spent a little time looking for one.  Apparently I do not have the financial means to support revisiting my wistful recollections. 

My first was my own homebrew 6800, with 128 bytes of RAM. Definitely not single board, though.

My second was a Nascom-1.

My third, and first bought as a professional, was an Acorn System 1.

And that was all, thank god. Some poor sods on the next bench had to play with an intel SDK-86, bletch.

[Vgkid]

Yeah can’t moan for £3 at all. An afternoon of debugging fun is a pretty cheap exercise.  I’d be annoyed if I’d spent £10 on it. 

I looked on the Caddock site just out of dumb curiosity; it appears they've stepped into the 21st century and don't even acknowledge the existence of products in DIP package. 

Well, yeah. Caddock makes almost entirely TH products. All of their precision resistors/networks are sip radial mounts.

[worsthorse]

bd, do you want a pair of fetrons to replace those tubes? i am pretty sure that VTVM has the right tubes for it and i have a couple of them around here...

Thanks for the offer but will pass on this one at least. The guts of this thing are too badly constructed for me to let it hang around so it's getting turd polished and chucked on ebay.

okay on the fetrons. but i thought you gave up selling on ebay?   

Now we're talking. Proper computers! This 177 is 6800 based rather than 6502 of the KIM-1 however. We had our own version of the KIM-1 here: https://en.wikipedia.org/wiki/Acorn_System_1 . I actually cut my teeth on 6502 assembly on the BBC micro which is the descendant of that, then built a Z80 SBC. I stole all the Z80 books from my local library. They probably have bounty hunters out for me now 

I learned assembly language programming on the 6502 and the Z80. My first post-university job, I microcoded bit slice AMD processors. At least one of my summer jobs involved altairs, switch setting and paper tape readers.  Who needs a stinkin' compiler? 

EDIT: The KIM-1, at 245USD in 1976, would cost the equivalent of 1100USD today.  Wonder what the equivalent computing power per dollar would be?

[mnementh]

EEEEEEEEEEK!!! OUT OF COFFFEEE!!!   

None to brew... no instant left... not even the cheap mexican shit.

mnem
*chewing on a used coffee filter*

Time to start drinking TEA then 

In this house that would be worse than contacting the bubonic plague. 

Naaahhh... I just shambled about zombie-like until we made it to my son's Pokemon League meet, whereupon I downed a couple NineBucks Mocha Frappucinos. They're alright as long as you don't expect them to taste like coffee.     

Picked up a big can of Chock-Full-o-Nuts when we headed back, which my wife brewed while I heated up the bacon & eggs & biscuits "breakfast for dinner" she'd made while we were out. Which was, in fact, exactly what I was thinking of making for myself all the way home. 

mnem
I used up a lifetime's worth of luck finding her... *sigh*

[mnementh]

Yeah can’t moan for £3 at all. An afternoon of debugging fun is a pretty cheap exercise.  I’d be annoyed if I’d spent £10 on it. 

I looked on the Caddock site just out of dumb curiosity; it appears they've stepped into the 21st century and don't even acknowledge the existence of products in DIP package. 

Well, yeah. Caddock makes almost entirely TH products. All of their precision resistors/networks are sip radial mounts.

Ummm... okay...  Caddock made the 14-pin DIPS in question, they just don't anymore. 

mnem