• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

ideas for tube burn in / bias match station

DIY tube matcher / ideas for tube burn in / bias match station

I have a couple of bogen MO200 chasis and want to put together a tube burn in station that will let me measure and grade Ip based on bias current. In my mind, for my use, the ability to measure transconductance is not needed since I have a tube tester for that. In my senerio, I have 75 or so 6550 tubes and I need 48 with the lowest Ip draw. So I'm thinking of putting together something that will both put the tubes in a real wolrd environment and let me select the tubes individually to measure Bias. I'm thinking a basic tube power am circut with bias resistors and meter controlled by a rotatry switch . I have 6l6's, el34s, KT88's, KT90's as well as the 6550's. I have one schematic for a ppower amp that has the ability to use all the tubes I indicate so I am thinking of starting with that. Main goals is

Plate Voltages 325, 400 VDC (or higher if it makes sense)
Bias Voltages -60, -48, -36, -24, -14 VDC
Test Frequency - not sure about this one...normally when we bias tubes it's with no input signal...do we think a test frequency is needed?

So basically, does anyone know of any projects with schematics of something simular to this. No sense in re-inventing the wheel if someone has already done this.
 
Last edited:
Hello Speakerfritz

For matching Tubes maybe this Schematic is usefull as a starting point:

Röhrenprüfgerät für Gegentakt-Röhren

Only problem is to find an amperemeter with positiv and negativ range and zero at center position.

Best regards
Michael

Interesting idea to match tubes all over their working range. The ammeter may indeed be a problem, but wouldn't it be possible to use a low value (10R?) resistor instead with a DVM across it? DVM can show both + and - without a problem.
 

Attachments

  • Gegentakt-RoePruef.jpg
    Gegentakt-RoePruef.jpg
    19.7 KB · Views: 365
hmm, how many and what make tubes where you planning on? ahh i see, power tubes.

you could do it the classical way, variac-insulation transformer 230 vac in series whit a 50-100watt ish lightbulb and all grids on anode, whit a resistor in series whit them to limit the current. t


this jig en order to burn in just needs to keep the cathode activity up, you can do it in a fancy way whit stabilized voltages and the Whole deal. but in the end where talking about a burn in rack. Not a tester


testers on another hand. you just need some power supplies, a bounch of sockets and a lot of meters to measure all the voltages/currents and from that you can test everything an engineer needs to know.

i made my own tester, its not fully finished yet. but it can already do a lot more than a hickock will ever do.

best advice: search for the sussex valve tester. and overbuilt it for power tubes. its not a burn in setup the sussex will overheat . rarely if any testers/matchers are made to burn in tubes, its entirely different


Btw, have another good look at your circuit. if you burn those tubes in whit multiples at a time. and one of them develops a short G1-K your gonna lose every tube in that rack, so its good to take some precautions.

V4lve.
 
Last edited:
v4lve lover , I have about 100 6550 tubes. need to sort thru them and pick 48 that are close in Ip. I have two VTL MB-1250 amps that use 24 6550 tubes each. I'm thinking of using a Bogen MO-200 as the donor for the chasis and power supply. The MO-200 has 8 octal sockets and would just need a make over. I do have a real good calibrated transconductance tube tester and I usally test the tubes and include a bias probe on the tester to grade the Ip at the same time. But if I setup a MO-200 as a tube matcher...I can grade 8 tubes in one shot. I was also thinking about just taking one of my other amps, and jumping pin 5 all together so that the bias voltage was exactly the same....and then use a variac to bring the amp up to a safe bias consumption....mark the line voltage and repeat the process. I have an amp that could do 10 tubes that way. But the matching circut is so simple. might as well go that route.
 
How accurate is matching the bias condition in triode mode vs pentode mode?

If one matches the tubes in triode mode (as the proposed circuit shows) does it guarantee they will match in pentode mode?

If not, then an adjustable screen bias circuit needs to be added to allow one to select the pentode bias conditions to match the tubes.
 
Its hardly the same. Beam tubes tend to deviate alot more.

It has to do whit the effects all the different dimensions of the grids have on one another. plate distance ect, and because all these parameters are never the same. the tube cannot measure the same.

By the way. 100 tubes is not gonna cut it. depending on vintage/brand , your going to have atleast 10-20 that fall out of spec by a wide margin and some Dead/gassy tubes

But as far as matching goes. make yourself a nice test panel whit adjustable supplies for anode,screen and negative grid voltage

get some mosfet regulated supplies for the HV 0-400v will suffice for most applications
important parameters for your application are monitoring the negative grid voltage. plate current, and to some degree screen current.

and to some degree making sure the heater voltage is spot on 6.3V as. ive personally noticed tubes will increase cathode current by a LOT if the tubes are overheated. this messes whit your results , and that's one of the reasons two tube testers rarely measure the same.

Really, a tube tester/matcher is pretty simple. Get yourself some of those digital panel meters witch are about 10usd each on ebay. incorporate a gas test, shorts test and a noise test and your all set.

V4lve
 
A very basic setup must have a few things:

Heater supply: In this case 6.3V at 1,5A DC and stabilized is prefered as it helps maintain accuracy, could be lineair Or switching LM338-LT3803

The Lt3803 is ideal as it is Low dropout you could get a 12 volt transformer and do 6.3 and 12.6 volt tubes whit it. a buck converter may be simpler though

Anode/screen supply: 0-400v there is a very good circuit whit a mosfet source follower. it has schort circuit protection
somewhat like this: http://www.roehrenseite.de/projekte/pruefgeraet/bilder/netzteil_ua_ug2.jpg if you recalculate the zener string for a bit over 400 volts and take a higher rated mosfet

full article Andy`s Röhrenseite Röhrenprüfgerät (german)

bias supply. 0 to -60 volt will be required. I have a tried circuit that is short circuit proof and provides 0 to 100 volt whit an option to switch polarity.

Meters to monitor all the voltages

here's a picture of a compatible setup i made, its working fine. but I'm still thinking of possible improvements. it is still missing the lettering, and the rest of the enclosure.
An externally hosted image should be here but it was not working when we last tested it.


specs:

Anode and G2 voltage: 0-320volts limited to 140 and 45mA
grid +- 100 volts 10mA
heater: 1.25 to 24 volts 5A (Lm338) whit 3 taps on the transformer to limit dissipation.
all Pots are 10 turn, and after this picture was made, i added a Dialer to the grid coarse and fine controls,

The 7 knobs are rotary switches, they set all the connections between the tube and the supplies.

the right knob operates a shorts test between all elements, when a short under 2mohm is detected the neon lights up, in the final setting (10) the relays for the high voltage are engaged

it has a lot more gizmo's i'm not going to bother you whit all of them, But of course i just wanted to show of my new toy:D
 
I have a couple of bogen MO200 chasis and want to put together a tube burn in station that will let me measure and grade Ip based on bias current. In my mind, for my use, the ability to measure transconductance is not needed since I have a tube tester for that. In my senerio, I have 75 or so 6550 tubes and I need 48 with the lowest Ip draw. So I'm thinking of putting together something that will both put the tubes in a real wolrd environment and let me select the tubes individually to measure Bias. I'm thinking a basic tube power am circut with bias resistors and meter controlled by a rotatry switch . I have 6l6's, el34s, KT88's, KT90's as well as the 6550's. I have one schematic for a ppower amp that has the ability to use all the tubes I indicate so I am thinking of starting with that. Main goals is

Plate Voltages 325, 400 VDC (or higher if it makes sense)
Bias Voltages -60, -48, -36, -24, -14 VDC
Test Frequency - not sure about this one...normally when we bias tubes it's with no input signal...do we think a test frequency is needed?

So basically, does anyone know of any projects with schematics of something simular to this. No sense in re-inventing the wheel if someone has already done this.


Here is something I came up with it works really well so far,just doing some tweaking now. It comes with a bias control and meter circuit, I’m thinking of using panel mount volt and amp meters if it goes into a case. At the moment I’m using about 495VDC for B+. I’m modifying it where there is just one control(trim pot) for both the reference voltage and meter calibration. Will post more pics later
 

Attachments

  • AF000B65-9CF9-42E7-9B48-ED7E3E6CB81A.jpg
    AF000B65-9CF9-42E7-9B48-ED7E3E6CB81A.jpg
    852.2 KB · Views: 72