Ken's Tube Tester
After designing and constructing a couple tube/valve curve traces that used a pulse testing method I decided it was often misleading and made tubes look better than they actually are the cathode performing better than it would under sustained high power, grid current is often acceptable until the tube runs at significant power for many minutes.
Here is the URL to one of my previous efforts. https://sites.google.com/view/kens-schematics-2/tube-tester-and-curve-tracer
Progress 9 March 2021, still working on code, going to put 6BM8 (triode pentode) and 6AS7 in the automatic test list.
Requirements.
1/ Employ linear power supply's capable of running tubes at up to 30 watts continuous anode dissipation.
2/ Use relay switching to the tube sockets, over 10 million combinations !
3/ Be easy to operate. This is a tube tester not a curve tracer, the idea is to see if the tubes are any good.
4/ Have a full old fashioned manual mode.
5/ Have a fully automatic mode. Select the tube type and push the enter button. When used in the auto mode the only controls that do anything are the mode knob, the 10 turn tube select control and the keypad. The bias voltage has an extended range in the auto mode.
6/ Have a small inbuilt printer. (not fitted at the moment)
7/ Display voltages, currents, transconductance, amplification factor, plate resistance and have a satisfactory gas test.
8/ Be capable of testing the current leakage between any two elements inside the tube using the relays.
9/ Double as a bench power supply. up to 4A when <7VDC and 3A when less than 13.VDC, 150mA when < 160VDC , 300VDC at lower current, bias 0 to -60VDC. The bias PCB with very small modification could do +bias with grid current to neg 150V.
10/ Not require an external computer. I'm using a dsPIC
11/ Use a commonly available parts.
12//Be capable of some sort of life expectancy prediction. Maybe based on the change in transconductance at a reduced heater voltage TBD.
I would like to plug in a tube select type, press start and walk away, on return see a print out of results at maybe 5 minute intervals while running at a reasonable power level.
This is the "relay cube" prior to installation, I designed it to switch the tube socket pins. Software reduces all the power supply voltages (except the bias that is supplied via a 10k resistor) to <1 volt prior to switching. Plug in chips TPIC6C595 were used for easy repair, 45 relays were used they are rated at 10A 250VAC as I'm switching at zero power, operation up to 353VDC should be fine, the maximum current will be <=4A. I wrote a function to suit and simply bit bash the port, a single 10 wire ribbon cable supplies the data and power to the relay cube.
I decided to use pots for the manual mode pin select as one can see the selection by looking at the chicken head knob position, the code has been written such that a click occurs when the pot is moved to another position. A small cheap ebay purchased 16 to one MUX board was mounted on the front panel so all 15 pots could be connected to the custom designed micro PCB using a single 10 wire ribbon cable, leaving some spare wires for expansion. The dsPic board I designed can also be seen in this image.
I found the up down buttons used to select the valve type a pain so I fitted a 10 turn pot with a vernier scale, 10 tubes per turn so 100 tube types on a single knob, it emulates the old paper roll quite well, simply select and press enter, it is easy to use and will mechanically remember the last tube selected. I've only about 10 tubes entered so far, still messing with software.
The manual mode is dangerous as one can easily blow up a valve, automatic mode is better as pin select and voltages are automatic, I destroyed a rather weak 6V6 with a wrong anode connection!
The orange wire used on the tube sockets is oxygen free, solid copper, silver plated, teflon coated ! BTW I have pads on the micro PCB for a couple EEPROMS that I have not as yet fitted.
Testing a 807 tube at 21.1 watts anode dissipation, passed gas test, 6.15mA/V looks like a good tube to me. I fitted a small quiet low powered fan just to move some air slowly through the device. The device is only warm after running for > hour under load.
The front panel decal is a sticky back vinyl printed by Officeworks, it was only intended for prototyping, I intending to redo it with modifications and spray it with lacquer prior to fitting. BTW this is the same material used to print signage that is stuck to the floor for people to walk on, the print job cost<$10.
email: ken.w.kranz@gmail.com
