Test Equipment Anonymous (TEA) group therapy thread

[PA0PBZ]

I like the color. What brand?

https://www.alibaba.com/product-detail/LCD-Display-Low-Price-Digital-Multimeter_60304904663.html?spm=a2700.7724857.main07.115.424590dbNpjG0X

[neo]

No i don't have three... yet. It is definitely going to be one sacrificing itself for the good of the other at this point.

At this point it makes you wonder if it is all worth it and what the long term prognosis is for the working 8400A if you're successful in getting one working.

Let me put it this way, if i get it working that means that in theory i can keep one working. Worst case scenario i turn the one with the non butchered display into a clock.

[Specmaster]

Yes, nixie clocks are cool 

[neo]

Yes, nixie clocks are cool 

But really, i have quite alot of motivation to get one working. If the wallet hears that i not only failed to accomplish anything with this amount of money but ruined one and made the other into a clock, well then this would be in my future.

[Specmaster]

Not really, a clock would be in use 24/7 365 days a year and for more than the meter would have been 

[mnementh]

Sent from my PC using a good old fashioned keyboard and mouse, without WhackaWhine.

 

I think I have turned off the Tapatalk BS and this message is a test.

You just need to edit or delete the default signature to make it go away permanently; just like that stupid one in iOS.


mnem
*Sent by nuking an 8mm wrench in the microwave*

[neo]

Not really, a clock would be in use 24/7 365 days a year and for more than the meter would have been 

Is it bad i already know exactly how i would build it? I'd re purpose the sample rate control to turn the display on and off, to prolong life of tubes, and use the range and function buttons to set it.

[Specmaster]

Not really, a clock would be in use 24/7 365 days a year and for more than the meter would have been 

Is it bad i already know exactly how i would build it? I'd re purpose the sample rate control to turn the display on and off, to prolong life of tubes, and use the range and function buttons to set it.

Nope, thats thinking ahead

[neo]

Not really, a clock would be in use 24/7 365 days a year and for more than the meter would have been 

Is it bad i already know exactly how i would build it? I'd re purpose the sample rate control to turn the display on and off, to prolong life of tubes, and use the range and function buttons to set it.

Nope, thats thinking ahead

That's good, now to focus on one working as opposed to how pretty it would look as a clock.

[Specmaster]

Sent from my PC using a good old fashioned keyboard and mouse, without WhackaWhine.

 

I think I have turned off the Tapatalk BS and this message is a test.

You just need to edit or delete the default signature to make it go away permanently; just like that stupid one in iOS.


mnem
*Sent by nuking an 8mm wrench in the microwave*

Thats exactly what I did 

[bd139]

My new, well old, Fluke 8000A just blew up. Uses a really cheap and dumb zener+transistor regulator for the analogue rails. The analogue parts have a decent PSRR so it doesn't need anything more complicated. However it looks like the current source resistor went short, blew the arse out of the zener which went open circuit in the power supply and the output jumped to 22v instead of the expected 15v. This made the analogue ASIC rather unhappy and it went to silicon heaven. RIP.

I was very surprised to see that the digital side of things isn't all that nice in the power supply department as well. The thing has no regulator (!). It's just a couple of transformer taps, two diodes and a big electrolytic cap.

Sooooo. Now I need another 8000A.

This has actually got me thinking about building a good quality LED bench multimeter from scratch however. No ASICs, just logic and analogue goodness. Dual slope AD shouldn't be too hard to knock up.

[Specmaster]

Why reinvent the wheel, couldn't you simply build a superior power supply for another one to prevent a recurrence? 

[neo]

My new, well old, Fluke 8000A just blew up. Uses a really cheap and dumb zener+transistor regulator for the analogue rails. The analogue parts have a decent PSRR so it doesn't need anything more complicated. However it looks like the current source resistor went short, blew the arse out of the zener which went open circuit in the power supply and the output jumped to 22v instead of the expected 15v. This made the analogue ASIC rather unhappy and it went to silicon heaven. RIP.

I was very surprised to see that the digital side of things isn't all that nice in the power supply department as well. The thing has no regulator (!). It's just a couple of transformer taps, two diodes and a big electrolytic cap.

Sooooo. Now I need another 8000A.

This has actually got me thinking about building a good quality LED bench multimeter from scratch however. No ASICs, just logic and analogue goodness. Dual slope AD shouldn't be too hard to knock up.

Speaking of 8000A problems, mine only displays one digit. It used to do this when run solely off it's batteries, then go back to all digits dimmer when plugged in. Now even plugged in it does this. Any ideas? I ask because it seems you know this model rather well.

Sorry to hear your's blew up.

[bd139]

Why reinvent the wheel, couldn't you simply build a superior power supply for another one to prevent a recurrence? 

Because I like reinventing wheels

Actually I just brought it up on my E3630 and it's sort of working-ish. I think whatever happened might be external to the analogue IC. Might be the input amp.

My new, well old, Fluke 8000A just blew up. Uses a really cheap and dumb zener+transistor regulator for the analogue rails. The analogue parts have a decent PSRR so it doesn't need anything more complicated. However it looks like the current source resistor went short, blew the arse out of the zener which went open circuit in the power supply and the output jumped to 22v instead of the expected 15v. This made the analogue ASIC rather unhappy and it went to silicon heaven. RIP.

I was very surprised to see that the digital side of things isn't all that nice in the power supply department as well. The thing has no regulator (!). It's just a couple of transformer taps, two diodes and a big electrolytic cap.

Sooooo. Now I need another 8000A.

This has actually got me thinking about building a good quality LED bench multimeter from scratch however. No ASICs, just logic and analogue goodness. Dual slope AD shouldn't be too hard to knock up.

Speaking of 8000A problems, mine only displays one digit. It used to do this when run solely off it's batteries, then go back to all digits dimmer when plugged in. Now even plugged in it does this. Any ideas? I ask because it seems you know this model rather well.

Sorry to hear your's blew up.

Probably the batteries have had it. The batteries (NiCd) are used to regulate the voltage to the switching converter to drive the second half of the power supply.

If you have a triple rail supply, connect the 5v, +15v and -15v rails after the pass transistors on the secondary of the switching supply (without the mains plugged in) and see if it springs into life. If not, measure them and see if they are low. It wont power up properly if 5v or +15v are low.

[Specmaster]

What is it with these old Flukes then, I always thought they were the holy grail of meters and built like brick dunnies, seems the Americans weren't adverse to cutting a few corners either then. HP with their 3458's if their battery failed it wiped out all calibration data, so is there a safe bet among these precision meters?

[bd139]

They're not all that. Why do you think I bought a new Keysight meter

The main attraction at the time of production were they were less shit than the other brands, reasonably priced and had decent service and support.

Really the best multimeters out pinnacled with the HP427A analogue. It was all shit after that, even to this day.

Fluke plagues: shitty LCDs, stupid battery implementations, terrible quality switches, unobtanium ASICs, flexi-boards, bad quality reed relays, dangerous line power running across boards, weak input jacks etc. You win some, you lose a lot.

Got to be honest, the GW GDM-8341 I bought is rather well built:

[Specmaster]

Thats an analogue meter? So whats wrong with AVO's amd Simpsons, I thought that they were also rated meters?

[bd139]

Thats an analogue meter? So whats wrong with AVO's amd Simpsons, I thought that they were also rated meters?

The 427A was high impedance (10M) FET multimeter. Avos are very unreliable (it's difficult finding one that isn't wonky these days), have strange 15v batteries that don't exist. Plus you drop one and that's the end of it. Simpson were pretty good I will say, at least in the later revisions, but again still low impedance.

Edit: actually fuck it, I'm going to build an analogue meter

[Russ]

Its far quicker to use tapatalk on occassions..

   Why is Tapatalk easier, more convenient?

Russ

[Specmaster]

Thats an analogue meter? So whats wrong with AVO's amd Simpsons, I thought that they were also rated meters?

The 427A was high impedance (10M) FET multimeter. Avos are very unreliable (it's difficult finding one that isn't wonky these days), have strange 15v batteries that don't exist. Plus you drop one and that's the end of it. Simpson were pretty good I will say, at least in the later revisions, but again still low impedance.

Thats my world shattered, I was told in that AVO's were supposed to be the worlds best analogue meter (certainly of the 20,000opv type) Yes, I know that the 427A was FET with an input impedance of 10M, so does the Heathkit V-7AU and quite a few other VTVM's , so what makes the 427A the best then?

[Specmaster]

Yes it is, sometimes the full site can be very slow to open on mobile devices but Tapatalk is really fast.

[neo]

They're not all that. Why do you think I bought a new Keysight meter

The main attraction at the time of production were they were less shit than the other brands, reasonably priced and had decent service and support.

Really the best multimeters out pinnacled with the HP427A analogue. It was all shit after that, even to this day.

Fluke plagues: shitty LCDs, stupid battery implementations, terrible quality switches, unobtanium ASICs, flexi-boards, bad quality reed relays, dangerous line power running across boards, weak input jacks etc. You win some, you lose a lot.

To quote a TV show, life is an ever swirling toilet that just.....won't...... flush.  It is all shit, some looks better, some works better, some is just all around better. But regardless the entire thing is just graduations or variations life included in this description.

[Specmaster]

I think my Avo 8 is OK, although that said I have not yet tested the current ranges but the others certainly agree near enough with my DMM's, perhaps I'll test that tomorrow while I'm awaiting parts being delivered.

[bd139]

Current ranges are where the traditional analogue units win always.

Thats an analogue meter? So whats wrong with AVO's amd Simpsons, I thought that they were also rated meters?

The 427A was high impedance (10M) FET multimeter. Avos are very unreliable (it's difficult finding one that isn't wonky these days), have strange 15v batteries that don't exist. Plus you drop one and that's the end of it. Simpson were pretty good I will say, at least in the later revisions, but again still low impedance.

Thats my world shattered, I was told in that AVO's were supposed to be the worlds best analogue meter (certainly of the 20,000opv type) Yes, I know that the 427A was FET with an input impedance of 10M, so does the Heathkit V-7AU and quite a few other VTVM's , so what makes the 427A the best then?

427A: fully temperature compensated, battery or mains powered, can be dropped multiple times and go “meh”, no weird probes required like VTVMs, really easy maintenance and calibration, gold PCBs and pot casings meaning no corrosion, easy maintenance, nothing that will kill you inside, wonderful service manual, big clunky high quality switches, stupidly high quality movement, parallax mirror.

I had one for a bit which had been in storage for about 20 years and was covered in dusty shit. Plugged it in, turned it on. Was bang on.

They're not all that. Why do you think I bought a new Keysight meter

The main attraction at the time of production were they were less shit than the other brands, reasonably priced and had decent service and support.

Really the best multimeters out pinnacled with the HP427A analogue. It was all shit after that, even to this day.

Fluke plagues: shitty LCDs, stupid battery implementations, terrible quality switches, unobtanium ASICs, flexi-boards, bad quality reed relays, dangerous line power running across boards, weak input jacks etc. You win some, you lose a lot.

To quote a TV show, life is an ever swirling toilet that just.....won't...... flush.  It is all shit, some looks better, some works better, some is just all around better. But regardless the entire thing is just graduations or variations life included in this description.

That is exactly what it is. Well said. I prefer my turds rolled in glitter instead of diharrea though

[bd139]

Designed me an analogue voltmeter on paper already. Going to try it out.

Basically CA3140E based voltage controlled current source with an LPF and 10M divider network up front.

Also found a 100uA meter movement I had for a PA current meter.