Tube matching, any hints on how to do it?

Tube-specific discussions

Moderators: zaphod_phil, CurtissRobin, colossal

Post Reply
cloudhopper
Occasional poster
Occasional poster
Posts: 165
Joined: Mon 04/26/04 2:00 am
Location: wichita, KS

Tube matching, any hints on how to do it?

Post by cloudhopper »

I don't have a mutual conductance tube tester or a tester that measures current at a given plate voltage/bias voltage. I just have a lowly emissions tester. (a cute little Sencor TC109 MightyMite, it does do a great leakage test though) I have been contemplating building a simple SE bench amp with adjustable fixed bias to use to measure tube current by using a precision 10 ohm resistor at the cathode to measure voltage across and determine current. I also recently bought a tektronix 453 so I have the ability to measure actual signal gain now. Does anyone have any suggestions on how to best construct a bench amp to test our dear EL84's so we can make sure one of our pair doesn't redplate while the other is loafing along? I believe current matching is more important than mutual conductance matching in our 18W amps but I'm interested in other views about that as well. I have used the precision resistor trick before as a permenant fixture in some builds but would rather not have to add it in to every build if I can make a bench amp or tool to check with.

Thanks for the help,
Cloudhopper
0 x

lbet
Occasional poster
Occasional poster
Posts: 295
Joined: Wed 09/29/04 2:00 am
Location: Brampton

Post by lbet »

I have a fairly ok tube tester that test for mu. However it does not test for plate current so I put the tubes in a kendrick 2410 and measure the plate draw in amp.
0 x

User avatar
zaphod_phil
Builder, Admin
Builder, Admin
Posts: 15208
Joined: Wed 03/19/03 2:00 am
Location: YYZ

Post by zaphod_phil »

Actually in guitar amps, close matching of power tubes isn't that important at all. The great vintage amps that everyone loves and tries to clone never used matched tubes. In the factory, they just pulled the next two (or four) tubes out of the box and stuck them in the amps. And a small amount of mismatch helps to make for a sweeter tone, with more even harmonics in it. However, badly mismatched tubes will be lesseffctive at cancelling out hum, so you will hear more hum, and in extreme cases you may even get one of the pair red-plating. So for guitar amp use, the easiest way to match the tubes is simply to plug them in and find pairs that give you the least hum.
0 x
Nature abhors a clean tube amp

User avatar
Newt
Occasional poster
Occasional poster
Posts: 251
Joined: Thu 01/29/04 2:00 am
Location: South Carolina, USA

Post by Newt »

If you add in a couple of 1-ohm resistors as Rich Stevens did in his "13-watter",

http://richardstevens.home.mindspring.c ... schem2.jpg

you can easily check the current draws of each tube at idle to see if they are grossly mismatched. This won't tell you how they perform in operation, but it's something.
0 x

cloudhopper
Occasional poster
Occasional poster
Posts: 165
Joined: Mon 04/26/04 2:00 am
Location: wichita, KS

Post by cloudhopper »

I prefer precision 10 ohm over 1 ohm because it puts the voltage into a more easily measurable range, just have to remember to move the decimal place over one. It just would be nice to match the tubes up before sticking a pair in our hot little 18 watters. Zaphod, maybe you're right, maybe I'm putting too much importance into the current matching. Might be more important if using a 125 ohm cathode resistor though.
Ibet, can you give me an education? I thought the Kendrick 2410 was a 6L6 amp.

Thanks,
Cloudhopper
0 x

Coco
Frequent poster
Frequent poster
Posts: 777
Joined: Sun 05/25/03 2:00 am
Location: Toronto, Canada
Contact:

Post by Coco »

Here's how to build a cheap tube matching device that uses 2 test jacks, 1 female socket, 1 male plug, 1 ohm 1/2 watt resistor some cable and a box. It uses an existing amp to measure the cathode current. For this you'll need a digital meter to measure low voltage as they will be in the 0.02-0.05 range. This is described in more detail in Weber's Tube Amp Talk.

This is somewhat like lbets concept because it emulates testing tubes in place.

Install the socket and two test jacks in the box. Except for the cathode pins, connect all the pins of the female socket to the male plug with a equal lengths of wire to the matching pin. i.e. 1 to 1, 2 to 2 etc.

Test that the 1 ohm resistor is exactly 1 ohm then insert it between the test jacks and connect the cathode wires to it. Solder it in place.

Now, remove the output tubes of the amp, plug in the test device into one of the output tube sockets. Turn on the amp (power-standby) then let the amp settle down and measure the voltage across the jacks (i.e. the 1 ohm resistor). The low voltage e.g. 0.035 V, will be equal to the current draw (0.035 mA) by virtue of the 1 ohm value and Ohm's law.

Record the value, put the amp into standby and test each tube to find the ones with closest matching current draw.

Hope this helps out with your build!

Cheers,

PS Don't get too carried away with exact matching of tubes.
0 x
coco

cloudhopper
Occasional poster
Occasional poster
Posts: 165
Joined: Mon 04/26/04 2:00 am
Location: wichita, KS

Post by cloudhopper »

Thanks Coco, any suggestions on where to find the 9 pin male plug? I guess I could always solder some pins into a female plug to make one. When you check current, do you only check at idle or do you apply a signal as well?
0 x

Coco
Frequent poster
Frequent poster
Posts: 777
Joined: Sun 05/25/03 2:00 am
Location: Toronto, Canada
Contact:

Post by Coco »

Can't say where to get the 9 pin male socket. Making one is a good idea. Measure current drain at idle.
0 x
coco

Post Reply