UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Homebrew Equipment (https://www.vintage-radio.net/forum/forumdisplay.php?f=99)
-   -   The "Sussex" Homebrew Valve Tester. (https://www.vintage-radio.net/forum/showthread.php?t=48853)

timohaveri 8th Mar 2011 10:38 am

Re: The "Sussex" Homebrew Valve Tester.
 
Quote:

Originally Posted by Brian R Pateman (Post 414000)
As these are used in a non-load breaking application they will probably be OK.

Thanks for the confirmation, Brian.

Quote:

Originally Posted by swordholder (Post 414362)
Although I havent tried them, I found these on the net

I wish I'd have found them earlier :-) They seem to be quite cheap comparing to the prices in Ebay.

I have another question. I too would like to include a diode / rectifier testing functionality into my version of the Sussex. The circuit similar to AVOs method of testing that was described here earlier seems to be the way to go. But does it really work the same way? I've been studying the AVO circuit and it basically just has a fixed resistance (700 ohms in total) in series with the valve (I assume that the capacitor can be discarded because we are using DC). In the lowest 3 ranges there is also an additional resistance added in series (18.6k/1mA, 3.6k/5mA and 600ohm/15mA). I've been doing some calculations and if I'm correct it seems that this arrangement lets the valve being overrun in some cases. On the other hand the current limiter / constant current circuit in the Sussex can apply up to full test voltage (60, 100, 175 or 225 volts depending on the range) across the valve enabling the maximum current to pass even if the valve is low in emission. For example a normal EZ80 will pass 40mA (per anode) if the voltage across it is 30V (according to the graph in the datasheet). Surely you can squeeze out that same 40mA from a weak valve by just raising the voltage? If the only goodness indicator is the anode current meter then all the good and some weak valves will show exactly the same reading which is the value of the current limit in that range. Or am I completely wrong about this? Perhaps a working solution could be achieved by combining the AVOs resistor arrangement and the current limiter? Would this still give meaningful results just with a current meter? Would it be difficult to incorporate a good/bad scale into the Sussex? It could be made for example from a multicolour LED bar graph display but how to control it? Anode current/voltage ratio? I assume it varies between valve types so it can't be easily used?

I hope you can make some sense of my ramblings :-/

Timo

Pamphonica 8th Mar 2011 10:59 am

Re: The "Sussex" Homebrew Valve Tester.
 
I looked at using these small thumbwheels as I had some "in stock", although I then found some larger ones. But they should be fine - I tested the insulation strength on mine with a modern megger and they showed no leakage at all at 1000V between adjacent sections. As long as they are not switched while carring HT or LT all should be well. But don't let them get greasy or wet!

FRANK.C 8th Mar 2011 1:58 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Hi Timo
In post #266 I posted a circuit for a diode function that I am using on my Sussex. It is current limited which varies depending on range selected and the voltage across the diode varies depending on range. It is up and running, but I haven't got a low emission diode to verify that it will show it up as low emission.
It is based loosely on the AVO circuit and on the higher ranges it will overload the HT transformer but as the test will be only for a brief period it is able for that short term overload.

Frank

timohaveri 8th Mar 2011 3:22 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Hi Frank. It would be very helpful if you could test how that circuit works since you have already built it. I'm by no means an expert so I might be fussing over nothing. Bytheway, does the AVO show just the good/bad reading on the scale or does it show the actual current value also (when testing rectifiers)?

Here's an excerpt from a web page that I found recently (http://www.jacmusic.com/Tube-testers/):
Quote:

The 80 rectifier to my experience should have a forward voltage of 25..40 Volts at 60mA to be good. Where 25V rarely happens, and above 45 the "bad" range begins.
It's not about AVO but I think this illustrates the point that I was making. Adding an anode voltage meter would show directly the forward voltage drop making it easier to evaluate the valve.

Timo

PS. I just found there are more types of those thumbwheel switches available that swordholder mentioned. The type KM2 is the smallest of them (dimensions are mentioned in the description). The bigger types might be more legible. They are also sold in Ebay for slightly lower prices.

http://www.sourcingmap.com/products_...type/all/bar/1

FRANK.C 8th Mar 2011 7:30 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Hi Timo
Referring to the diagram I posted. S1d sets the Anode voltage to the diode under test and this will not change unless the current goes over the current limit which is set by S1e. The Anode current will change depending on the emission and can be read the same way as a triode or pentode.

I don't know much about the AVO's the only ones I have seen were in photos.

Frank

Dekatron 9th Mar 2011 1:44 am

Re: The "Sussex" Homebrew Valve Tester.
 
The AVO Valve Testers (AVO Mk I, II, III, IV, CT160, CT160A and VCM163) have a RED/WHITE/GREEN scale of "Goodness" for the valve rectifier tests. In the CT160 manual on page 7 you can read that this scale is made so that the GREEN zone is from 130% to 70% and the WHITE zone from 70% to 50% and the RED zone from 50% to 0%.

Then on page 8 and 9 AVO says that the design is made in such a way that the circuit will let a valve rectifier pass a current that is equal to the maximum load current for the valve. The circuit is then scaled so that a valve passing this current deflects the meter to the center of the GREEN scale.

Since we know how the scale is made up from these coloured zones we could in fact read the current on the scale if there had been any lines to show it, but now we will have to make an educated guess from what we know from the percentages. Full scale deflection in fact corresponds to 135% of the current flowing through the valve at each range, with some small variation for each range except for the VCM163 which has larger deviations on some ranges.

I am currently writing an article explaining all of the maths and why the components are choosen as they are inside an AVO valve tester but it is probably a few days (if not a few weeks) before I can finish it. Then you will be able to read more about it here on the forum.

FRANK.C 9th Mar 2011 11:00 am

Re: The "Sussex" Homebrew Valve Tester.
 
Hi Martin
Thanks for explaining that, it will make it easier for me to interpret the readings I get and make them more meaning full.

Frank

neun elf 9th Mar 2011 12:23 pm

Re: The "Sussex" Homebrew Valve Tester.
 
5 Attachment(s)
Hi,

I started late, on my build but with help from various members I think I have caught up. This is my VT1 built into an RS service case (about £29) , at just over A4 footprint it makes avery compact and portable valve tester, and works a treat.

Brian R Pateman 9th Mar 2011 12:48 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Very nice job there, very professional presentation.

FRANK.C 9th Mar 2011 2:24 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Well done a neat and compact build.

Frank

neun elf 9th Mar 2011 3:58 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Thanks for the compliments guys, but I fear that the "works a treat" statement may have been slightly premature. On going through my stock of valves I am finding that the Gm figure is about twice what the AVO data would lead me to expect. Any clues as to why that should be? Thanks in advance.

FRANK.C 9th Mar 2011 6:50 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Hi
Firstly check that the output from the oscillator is 100 mV and the anode and screen voltages are what they are set to. Also check that the value of the resistor that Gm is measured across is correct.

Frank

neun elf 9th Mar 2011 8:28 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Sorry,

I was having a "senior moment". Forgot to engage brain before operating keyboard. The problem was with the oscillator output drifting all over the place. Extremely difficult to set precisely, I think I will replace the pot with a ten-turn one.

jimmc101 9th Mar 2011 8:43 pm

Re: The "Sussex" Homebrew Valve Tester.
 
It might be worth checking the amplitude stabilising circuit... D6, R19 & 20

Jim

neun elf 11th Mar 2011 9:39 am

Re: The "Sussex" Homebrew Valve Tester.
 
Now that my Valve tester is basically working, I have a few questions for the more knowledgeable on the forum, firstly, even on the 250v setting, my 230v mains produces heater voltages well above the +5% reccomended by the valve manufacturers (6.8v on 6.3 setting, 13.9 on 12.6v setting) is this critical and should I stick a resistor in the heater common line. Will it affect the readings?. Secondly, why use 47v and 15v Zeners to produce a 297v supply, when an 18v zener would produce a 300v supply, and does it matter? Lastly how does one interpret the readings, I often get near the expected gM figure for a valve, but the anode current may be higher or lower than given, what does that mean, or do I ignore it. Sorry if I am asking dumb questions, but maybe others have pondered these things too!! :dunce:

Ed_Dinning 11th Mar 2011 10:15 am

Re: The "Sussex" Homebrew Valve Tester.
 
Hi Gents, the heater voltage will drop on load (ie with a valve fitted), but should be within 10% of nominal when on load. There are also a wide range of heater currents in the valves tested from 0.1A to over 2A in common 6.3v valves.
To obtain a precise 6.3 volts in all cases from a simple transformer is not possible.

This is why mains voltage tappings are fitted to allow setting of the primary in 10v steps; also note that the mains supply will vary over the day in many locations so it should be monitored and re-adjusted (as per Mullard HSVT) in areas where this is a problem.

Ed

Denis G4DWC 11th Mar 2011 9:54 pm

Re: The "Sussex" Homebrew Valve Tester.
 
I am about to order some of the Digital meters for the Sussex and another couple of projects. On all the pictures I have seen there are only 3 digits showing although the Website claims 3 1/2 Digits. Are they 3 or 3 1/2? I really need 3 1/2 for the other Projects.

Thanks

Denis

timohaveri 12th Mar 2011 9:00 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Quote:

Originally Posted by Denis G4DWC (Post 415358)
I am about to order some of the Digital meters for the Sussex and another couple of projects. On all the pictures I have seen there are only 3 digits showing although the Website claims 3 1/2 Digits. Are they 3 or 3 1/2? I really need 3 1/2 for the other Projects.

The displays used in these meters are 3 1/2 digits but it depends on the measuring range whether the "1" is ever shown or not. For example a 500 VDC meter only uses the three full digits but a 199.9 mV meter uses all of them. The sellers all use basically the same picture showing the meter just powered up and displaying three zeroes without the decimal point so the pictures can be a bit misleading.

voltair 22nd Mar 2011 2:28 am

Re: The "Sussex" Homebrew Valve Tester.
 
Hi Mike,
Just found this thread so I am new to it, please bear with me.
I am just a few weeks away from retiring and was looking for something to keep the mind active. I have now found it. I think your Sussex VT1 is outstanding so I thought I would have ago at yours. I just hope I am up to it, is there now available a definitive circuit and parts list?

Peter

swordholder 22nd Mar 2011 5:06 pm

Re: The "Sussex" Homebrew Valve Tester.
 
Hello Peter,
If you go through the posts you will find a wealth of information. Post 341 gives a parts list.
Les has produed a super manual on his unit.
Martin has PCBs and Ed has a custom mains transformer ( I have used one in my prototype).
Various models are shown in the posts, and are really professional in appearance.
I'm sure you will find it a satisfying experience and you will have a good unit at the end of the day.

Regards

Mike


All times are GMT +1. The time now is 3:53 pm.

Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.