The "Sussex" Homebrew Valve Tester.

[retailer]

Well done swordholder you stand tall amongst us, anyone that has scratch built something before will know the amount of effort that you have put into your project.
I built a valve tester from a circuit designed by Steve Bench about 6 yrs ago, some readers will know it as the RAT valve tester. I'm a hobbyist not a design engineer so it wasn't a hard decision to use someone else's idea. The RAT tester uses a variable constant current source at the cathode to control the anode current and the valve then sets it's own bias, it does work well. I think that the AVO 163 is able to use either variable constant current source or variable grid bias to test valves.
I had planned to wire the valve base setting switches so that the AVO data manual could be used, but decided that it wasn't entirely a good idea. I can't recall exactly what the problem was but it may have been peculiar to the RAT tester circuit only.
The 200mV AC panel meter that you've used looks very good, I didn't know of their existence 6 yrs ago and had to make my own. I'll look forward to reading the full HOW TO article when you get it done.

[Guitarist28]

Hi all,
This is such a brilliant piece of work that I would rather spend money building a practical and up to date item of equipment than invest in an old valve tester (and to learn from the building of it too).


Well done and lets hope that this is a highly successful project.

[MichaelR]

Originally Posted by Ed_Dinning

Hi Gents, if there is sufficient interest I can probably arrange to get transformers wound and vacuum impregnated. If I have an idea of the quantities I may get a decent price from my local winder.

I will have one ED

Happy New Year to you

Mike

[swordholder]

Ed and others,
If you are going to get a transformer wound, may I make a couple of suggestions.
Don't bother with a 250 - 0 - 250 winding, make it a single 250v one and use a bridge rectifier, I could redesign the PCB to include the 2 extra diodes if required.
Include the 30v winding for the LT supply.
Include the 3 x 6.3v windings for the DVMs
It may be worthwhile to have a multi-tapped heater winding to include 4v valves etc.
If you are testing directly heated valves, an extra switch will be required to connect one side of the heater to 0v. This will of course mean that the C/H LED will stay alight unless a second pole on the switch disconnects the LED.
I realise that the cost of a custom wound transformer depends on the number of windings, so it may not be financially viable to replace the LT and DVM transformers.
What do other members think?
Regards to all
Mike

[Mike Phelan]

I've only just found time to look at your design in detail and read this thread; I would have been very tempted to make this if I did not already have a CT160.

Well done, and certainly worth publication.

[AmadeusMozart]

I have just completing collecting all the parts for the RAT tube tester when someone pointed out to me this thread. The RAT tube tester is based on a CCS in the cathode lead and this results in lower g2 and anode voltage as measured to the cathode which I was going to change to using a negative grid voltage. The advantage of the CCS is that it is easier to adjust.

I would make two changes and this is based on observations I made while restoring some Mercury 1000 tube testers and feedback from someone else who has the RAT tester.

The first one is that I would use some ferrite beads at either tube tube sockets or at the switches to try to prevent parasitic oscillation. The second would be to use a setup for valve sockets like the Mercury 1000 / 2000 is using (circuit diagrams are to be found on the web). What is done in these valve testers is to have multiple sockets for the same base but which have different filament connections. Then hardwire the filament connections and do not connect the pins that have a filament connection to the switches. In that way a short of the filament across the switch is avoided (basically that position on the switch for that particular tubebase is n.c.).

There are only something like 4 different filament connections for the Octal and 4 for the Noval and 2 for the 7 pin. The Mercury 1000 has only 2 for the Noval. A worthwhiloe safety measure!

I am using 10 turn bourns pots for the voltage adjustments and am using a 5 A, 15 V variac (that I found on a local auction website) for the filament adjustment (have a 15 V winding on the transformer). This will allow me to drop the voltage on the filament and see the corresponding drop in transconductance which is an indication of how much life in the tube is left (see the RCA tube manual on this).

Anyway this is an elegant solution and I like the layout of the whole. Just wish I could get a similar cabinet locally, my metal working skills are not that wonderfull.

Thanks for publishing and congrats on making something so good looking, well done.

AM

[Guitarist28]

Originally Posted by Ed_Dinning

Hi Gents, if there is sufficient interest I can probably arrange to get transformers wound and vacuum impregnated. If I have an idea of the quantities I may get a decent price from my local winder.

I would like one too please Ed

Thanks

Rob

[Ed_Dinning]

Hi Gents, I'm talking to Mike to see if there are any other features he would like to see on the transformer and I'll get something costed out when this is complete.

Best regards, Ed

[swordholder]

Hello All,
Ed has asked me what windings to put on the mains transformer.
I have opted for a 250v 100ma secondary (use a bridge rectifier)
3 windings for the DVM (6v)
LT -ve winding 30v.
Withe regard to the heater voltage would any prospective builders reply WITHIN THE NEXT 7 DAYS their requirements.
Please bear in mind that only the most popular voltages can be catered for (ALSO REMEMBER THAT THIS HAS TO BE SWITCHED AND THAT NORMAL SWITCHES CANNOT SWITCH HIGH CURRENTS AND ARE ALSO LIMITED IN THE NUMBER OF WAYS)
Please also bear in mind that the cost of the transformer will undoubtably rise with each heater tap.

Regards
Mike

[MichaelR]

For me 6.3V is adequate.

Regards
Mike

[Guitarist28]

yes 6.3 volts is ok

Thanks

[AmadeusMozart]

I am not interested in a transformer but here are my thoughts on the subject:

Some of the power tubes have a plate dissipation in excess of 35 watts. It is always best to test a tube in a setting which is close to its intended use. Similarly some tubes are tested at 250V / 100mA anode current (e.g. 6L6-GC).

I found a toroid on eBay from a seller in Hong Kong which does 200 mA which is a bit overkill. When I was buying this I was aiming for a HT winding of 275 - 300 V @ 135 mA.

Hope this helps, AM.

[georgesgiralt]

Hi !
I thought that the transformer was the tough part to get;
Before subscribing for the transformer, I had the idea to check availability of the DVM. It seems the seller has none of them to sell. (actually the Ebay shop is empty) So I wonder if the project is still feasible at low cost

[swordholder]

Hi All,
The last thing I thought there would be trouble with was the module.
I did a search on E Bay for "Blue LCD Meters" and a company called Asia Engineer has them in stock.
The tester will supply 250v @100mA OK, the current is limited to about 110mA. The transformer shows no sign of stress as it is only used for a short period of time.
Thanks to others who have given their heater requirements, will hold those on file.
Regards
Mike

[GrimJosef]

Originally Posted by AmadeusMozart

It is always best to test a tube in a setting which is close to its intended use.

This is SO true, especially with the larger and more expensive valves.

Originally Posted by swordholder

The tester will supply 250v @100mA OK, the current is limited to about 110mA. The transformer shows no sign of stress as it is only used for a short period of time.

25W should be OK for most valves. But the reason for running at high power is so that the internal metalwork and, to some extent, the glass envelope can get properly hot. This takes at least a couple of minutes.

A useful feature, if it's available, is the ability to run at high power for many hours. This can be an effective way of rescuing gassy valves. I've had about a 50% success rate with vintage KT66s. Since each one goes from being worth <£10 as a dead display item to >£50 as a working valve it can very quickly pay for a high current transformer. However I do realise that this is not what more than 90% of the unit's users will be doing with it. So if it would seriously increase the price then I'd say it probably isn't justified.

Cheers,

GJ

[Ed_Dinning]

Hi Gents, let's have a good discussion on this one.
I had thought of a seperate heater transformer for those who wanted a full compliment of taps.
It would also be possible at say, a 20% cost increment to have a higher current rating that was continous for soak tests. These could also be done at reduced anode voltage if the GB were also reduced (as we only need to get the valve hot, so a lower voltage tap would work).
Initial thoughts were to use an unshrouded transformer with mounting decided by the purchaser, but I know many of you take pride in the appearance of their equipment so I will see if shrouds for upright or drop through mounting are available. (not available for many lam sizes these days).

Keep the ideas coming.

Ed

[Richard]

Hi everyone

Well done Mike on a great project.

Re the transformers, how about one of these for the HT
http://uk.rs-online.com/web/search/s...duct&R=5357783

and 1 or 2 of these for the 6v for the meter units
http://uk.rs-online.com/web/search/s...duct&R=3472464

Re the LT, for the people wanting a range, could this not be DC and variable using a LM317 or similar type device, say 0 - 40v with an additional 40v supply you can connect in series, monitored by another cheap meter unit?
http://uk.rs-online.com/web/search/s...duct&R=0805489

I don't know how the cost of these compares to a custom wound unit?

Richard

[ppppenguin]

Originally Posted by Richard

Re the LT, for the people wanting a range, could this not be DC and variable using a LM317 or similar type device, say 0 - 40v with an additional 40v supply you can connect in series, monitored by another cheap meter unit?....

This is going to need some hefty heatsinking to cover that voltage range at a high enough current to do some heaters. For example a 6080 with a 6.3V 2.5A heater would disspate over 80W in the regulator. I know this is an extreme example but it shows the potential problems. This is why multitap transformers are often seen as the best method. The alternative might be a switchmode regulator or pre-regulator.

[Sean Williams]

Or a variac on the primaries.......

[Top Cap]

I know it may cause some re-design, but I have a qty of 70 of these devices if they would be of any use in the project. I certainly would not be asking the £25 I have seen them going for on web sites, more like £2.50.
Les