Test Equipment Anonymous (TEA) group therapy thread

[BU508A]

* BU508A is showing the Dwagon the Caterpillar S61         

[Robert763]

Is that a selenium rectifier on the left? If so, must go. 

Edit....Yep....get rid of it. 100 ohm, 2 watt resistor in series with the sillycon diode.

Is 100Ω, 2W a generic or specific value?

This is not the plate voltage (which is dealt with by a hollow-state rectumfryer) but instead a low-voltage circuit that goes off the heater supply to supply drive to Ohms measurement. Relevant circuit below.

That's generic for a typical B+ supply. Selenium rectifiers drop 5 or more volts, and even more as the age, as opposed to a silicon which drops .6 volts. The circuit you have there is much different.

You may have enough adjustment to maintain the -4.2 volts. But if not you may have to increase the 33 ohm resistor a bit.

I hadn't realised that selenium rectifiers drop that much. Given that, I'd think that it was probably better to cook up an active solution rather than simply using a resistor, I'd think you want something closer to a constant voltage drop. So what about replacing the resistor with a V_be multiplier, or an amplified zener, or even just a whacking great power zener?

Jeez... if you're going to go that far off the purist-ideology trail, seems to me might as well plop a LM1117-A with a couple 0603 SMD resistors across the pins in there; just hide it all under the carcass of that Se rectifier.

mnem
 

Well apart from the fact that the output is half wave rectified AC not DC....
It's only used on th x1 Ohms range so just leave it be.
If you are really worried than stick a 1N400x in there. Measure the Vf of the selenium with with the unit on and if it's much more than a volt (I doubt it, this is a low voltage rectified not an HT stack) stick a resistor of Vf / 0.1A (load current) in series.
Adjust per manual.

[med6753]

Is that a selenium rectifier on the left? If so, must go. 

Edit....Yep....get rid of it. 100 ohm, 2 watt resistor in series with the sillycon diode.

Is 100Ω, 2W a generic or specific value?

This is not the plate voltage (which is dealt with by a hollow-state rectumfryer) but instead a low-voltage circuit that goes off the heater supply to supply drive to Ohms measurement. Relevant circuit below.

That's generic for a typical B+ supply. Selenium rectifiers drop 5 or more volts, and even more as the age, as opposed to a silicon which drops .6 volts. The circuit you have there is much different.

You may have enough adjustment to maintain the -4.2 volts. But if not you may have to increase the 33 ohm resistor a bit.

I hadn't realised that selenium rectifiers drop that much. Given that, I'd think that it was probably better to cook up an active solution rather than simply using a resistor, I'd think you want something closer to a constant voltage drop. So what about replacing the resistor with a V_be multiplier, or an amplified zener, or even just a whacking great power zener?

Jeez... if you're going to go that far off the purist-ideology trail, seems to me might as well plop a LM1117-A with a couple 0603 SMD resistors across the pins in there; just hide it all under the carcass of that Se rectifier.

mnem
 

Both of you need to go look at that schematic again. Up to CR-1 (The selenium rectifier) that's all AC voltage. Coming up with all these fancy schemes are gonna smoke on you unless you RECTIFY the AC voltage first. Take your choice. Half wave. Full wave. But do THAT first. 

[mnementh]

FFS, med... rectifying the voltage first is a given here; even inferred in my suggestion to "hide it all under the carcass of that Se rectifier"...  we're talking aboot how to make that 1.2V clean and stable without the huge voltage drop of that smokebomb. 

mnem
 

[mnementh]

and even if you have taken care of about anything that can go wrong Murphy will eff you from behind.

happened to me just yesterday.

Some highly skilled staff migrated from Exchange old-version to a newer version but did not take over the old policies including those for automated deletion (and the exceptions (!)).

Now guess what happened. Not allowed to discuss more details.

What an effing shitshow.

How many kilometers of backup tape are we talking here? And how many cubicle-monkey-hours of lost work in the intervening days between restore point and time when you had to drop the sword?

Just remember... like bd says: "My house is built out of other people's shitshows."  

mnem
*says a prayer for Saskia's already worn-thin sanity*

[McBryce]

Is that a selenium rectifier on the left? If so, must go. 

Edit....Yep....get rid of it. 100 ohm, 2 watt resistor in series with the sillycon diode.

Is 100Ω, 2W a generic or specific value?

This is not the plate voltage (which is dealt with by a hollow-state rectumfryer) but instead a low-voltage circuit that goes off the heater supply to supply drive to Ohms measurement. Relevant circuit below.

That's generic for a typical B+ supply. Selenium rectifiers drop 5 or more volts, and even more as the age, as opposed to a silicon which drops .6 volts. The circuit you have there is much different.

You may have enough adjustment to maintain the -4.2 volts. But if not you may have to increase the 33 ohm resistor a bit.

I hadn't realised that selenium rectifiers drop that much. Given that, I'd think that it was probably better to cook up an active solution rather than simply using a resistor, I'd think you want something closer to a constant voltage drop. So what about replacing the resistor with a V_be multiplier, or an amplified zener, or even just a whacking great power zener?

Jeez... if you're going to go that far off the purist-ideology trail, seems to me might as well plop a LM1117-A with a couple 0603 SMD resistors across the pins in there; just hide it all under the carcass of that Se rectifier.

mnem
 

Both of you need to go look at that schematic again. Up to CR-1 (The selenium rectifier) that's all AC voltage. Coming up with all these fancy schemes are gonna smoke on you unless you RECTIFY the AC voltage first. Take your choice. Half wave. Full wave. But do THAT first. 

Let's see what ElectroBoom would suggest...



McBryce.

[med6753]

Let's see what ElectroBoom would suggest...



McBryce.

One of these days Darwin is gonna snag his ass. Can't be soon enough. 

[Mr. Scram]

One of these days Darwin is gonna snag his ass. Can't be soon enough. 

You know he is fully qualified, right?

[bd139]

Probabilistic outcomes affect everyone taking risks however small.

Fine example: Eric Engstrong, of DirectX fame, apparently died from complications after dropping a monitor on his foot. He didn't wake up that morning with anything risky in mind.

In short, at average, professional life expectancy is somewhere between idiots and people not involved in doing it at all

[Mr. Scram]

Probabilistic outcomes affect everyone taking risks however small.

Fine example: Eric Engstrong, of DirectX fame, apparently died from complications after dropping a monitor on his foot. He didn't wake up that morning with anything risky in mind.

In short, at average, professional life expectancy is somewhere between idiots and people not involved in doing it at all

People not involved are at home and the home statistically isn't that safe at all.

[med6753]

One of these days Darwin is gonna snag his ass. Can't be soon enough. 

You know he is fully qualified, right?

(The only PF song where you hear Nick Mason's voice)

[Robert763]

FFS, med... rectifying the voltage first is a given here; even inferred in my suggestion to "hide it all under the carcass of that Se rectifier"...  we're talking aboot how to make that 1.2V clean and stable without the huge voltage drop of that smokebomb. 

mnem
 

FFS mnem, the LM1117 does not do work with neagative voltages either    Nor does it go down to the required 1.2V. I was being nice not pointing those out. The impedance of the circuit is also important as it is shorted to ground when the ohms probes are shorted.
Why can't just just admit when you are wrong 

[bd139]

People not involved are at home and the home statistically isn't that safe at all.

The people locked in my basement are perfectly safe! I just wish they'd stop banging so loudly.

[Mr. Scram]

(The only PF song where you hear Nick Mason's voice)

That song reminds me of certain people up to their elbows in high voltage electronics daily. 

[med6753]

(The only PF song where you hear Nick Mason's voice)

That song reminds me of certain people up to their elbows in high voltage electronics daily. 

Yes, I finally deciphered what my Type 535A has been whispering......

"One of these days I'm going to cut you into little pieces"   

[Mr. Scram]

Yes, I finally deciphered what my Type 535A has been whispering......

"One of these days I'm going to cut you into little pieces"   

Now we know what that fan is for!

[McBryce]

Let's see what ElectroBoom would suggest...

McBryce.

One of these days Darwin is gonna snag his ass. Can't be soon enough. 

I doubt that anything he does is really dangerous. It's all thought through and more show than anything else.

McBryce.

[med6753]

The weather people are in their glory with snow fall predictions. Now up to 18+ inches (45 cm). "Biggest storm in 5 years". Even NYC may see a foot of the white stuff.

But they have a habit of predicting duds and we get 2 inches. 

[tautech]

The weather people are in their glory with snow fall predictions. Now up to 18+ inches (45 cm). "Biggest storm in 5 years". Even NYC may see a foot of the white stuff.

But they have a habit of predicting duds and we get 2 inches. 

Only looks like a little tropical storm 
http://metvuw.co.nz/forecast/forecast1.php?type=rain&region=natlantic&tim=012

[med6753]

The weather people are in their glory with snow fall predictions. Now up to 18+ inches (45 cm). "Biggest storm in 5 years". Even NYC may see a foot of the white stuff.

But they have a habit of predicting duds and we get 2 inches. 

Only looks like a little tropical storm 
http://metvuw.co.nz/forecast/forecast1.php?type=rain&region=natlantic&tim=012

You are looking in the wrong spot. The storm is currently over the Western Rockies and heading East. Once it gets near the Atlantic shore it is predicted to intensify and then do it's dump. 

[med6753]

Oculus...trying to send you a PM but your in box is full.

[tautech]

Just looks like another day in US paradise:
http://metvuw.co.nz/forecast/forecast.php?type=rain&region=usa&noofdays=10

[Cerebus]

Jeez... if you're going to go that far off the purist-ideology trail, seems to me might as well plop a LM1117-A with a couple 0603 SMD resistors across the pins in there; just hide it all under the carcass of that Se rectifier.

mnem
 

Why use a whole IC where a single transistor would do? Either case is going to need a diode to protect from reverse bias and a couple of resistors. One might do it if one didn't know how to design a V_be multiplier or an amplified zener, but using a whole IC to do one transistor's work is inelegant.

[tautech]

3 days ago blowing tits here then stinking hot muggy 27C yet a nice 14C overnight and for the next few days some reasonable early summer weather.
http://metvuw.com/forecast/forecast.php?type=rain&region=swp&noofdays=10

One more trailer load required today to fill the firewood shed and 30mins on the tractor driven docking saw then all done for another year and done before it gets too furking hot ! 

[Cerebus]

Both of you need to go look at that schematic again. Up to CR-1 (The selenium rectifier) that's all AC voltage. Coming up with all these fancy schemes are gonna smoke on you unless you RECTIFY the AC voltage first. Take your choice. Half wave. Full wave. But do THAT first. 

I, for one, am fully cognisant that it's AC. I'm suggesting a rectifier circuit that mimics the voltage drop of the Se rectifier. Thusly:



On the left, just a selenium diode.

In the middle, the classic resistor plus Si diode, that will have highly variable voltage drop depending on load.

On the right a V_be multiplier with resistors in the ratio 1:5 (ideally 1:5.7) produces a fairly constant 5V drop across the whole assembly. The reserve biased diode is to ensure that leakage current from the main rectifying diode doesn't reverse bias the transistor beyond breakdown.

Apologies for the crude diagram, but...

Edited to add: I was, as the diagram clearly shows, working off Med's original 5V, if as others appear to be saying it's only 1.2V in this case then it's all moot - an simple extra silicon diode would do the trick.