How to set bias on X100b

[brad44]

Would like to check bias and replace tubes.

Service manual says bias is adjusted to 40vdc from pin 3 of any 7868.

So that's measuring between pin 3 and chassis ground?
And then the adjustment pot is turned to 40VDC and that sets all 4 output tubes at once?

I know there is a mod to be able bias each individual but that's beyond me right now... I just need to check and maybe try some new tubes.

Please advise.

 

[dcgillespie]

Brad -- Your understanding is correct. Because the one pot adjusts all four output tubes at once, they will all need to be well matched so that they all share the current load through the output stages equally.

Good luck with your X-100B!

Dave

 

[brad44]

Thanks. Im surprised to see that the bias was spot on 40 for all 4 tubes... which are 50+ year old slyvanias that surely wouldn't test new.

 

[dcgillespie]

Because the cathode terminal of all four output tubes are connected together in the stock design, you will in fact get the same reading at each tube -- but that does not mean that each of the four tubes is drawing the same amount of current. To determine that, you would need to install 10Ω current sampling resistors at the cathode terminal of each output tube, and then measure the voltage produced across each resistor. Then you would be able to determine if the tubes are all drawing the same amount of current or not.

Dave

 

[brad44]

Ahhhha I knew it wouldn't be that easy.

So what does the basic measurement tell me?

I hoped that it would tell me my old old tubes are still working properly and holding proper bias. Or alternatively, if I install a matched quad, the measurement would tell me they are biased properly.

The resistor is placed between which pins on each tube socket?

Not a very user friendly thing I guess if the consumer had to modify the circuit to assure its running correctly.

 

[Tom Bavis]

No, you were supposed to buy a matched set of output tubes.

 

[brad44]

Tom- Just so I understand.... your implying that it would be standard for the average consumer to replace all 4 with a matched quad if only one tube would fail?

 

[larryderouin]

Because each tube is hand made, there will be slight to gross differences in operational plot's on each tube. If all are set @ 40V you could have all 4 with different bias current levels, 4 the same, 3 the same and 1 different, 2 at one level and 2 different, ad infinatum. Get hold of 4 10 ohm 1/4watt resistors. Disconnect the wire at pin 3 on a 7868. Connect a 10ohm resistor to the pin and then the wire you just removed. Repeat 3 more times. Then fire up the 100-B, and take your meter set to 20vdc and put a probe on each side of the resistor. You'll be measuring the voltage across the resistor, then converting to milliamps.

Say for Example you got .110v Move the decimal 1 place to the left and it's now ,011a because you divided Volts by resistance. .011a = 11ma. This is about 1/3 of what a normally operating 7868 should be at. to get wattage you multiply the Plate voltage by the ma but use the .0xxa reading to get the correct wattage. The 7868 has a max plate dissipation of 19W. You ideally want your wattage to be between 70% and 80%. 70% is about 13.5 and 80% is about 15.2 right off the top of my pointy little head(math always gives me a headache.)

Anyway, get the following voltages
Pin 3 on all tubes across the resistors.
Pin 9 on one or two tubes (they should all be within a volt of each other). (In excess of 350VDC)
Pin 1 on one or two tubes. " " " " " " " "" " " " " " " " " " " " " (In excess of 350VDC)

Write these voltages down and post them. We can then analyze what you've got.

 

[brad44]

Thanks Larry, that's the info I was looking for. But that might be on hold for now as I am chasing some static and buzz.

Problem 1:

I noticed some static in the right channel. I cleaned all the pots and the sockets and it is still there. Tapping on the right phase inverter tube I can recreate the static.

Problem 2:

Likely related. I am getting low level hum/buzz in both channels with the tape monitor switch in the "off" postion. Lowering the volume to 0 stops the buzz. And switching the monitor switch to "on" also stops the buzz.

Both are intermittent problems, since I switched the 2 phase inverter tubes from side to side and both problems disappear??!! It seems like in this position the tubes are maybe held a little tighter??? Could either or both problems just be a failing tube?

Its a frustrating old thing that will act up again. But it sure sounds nice right now playing Revolver thru a pair of Dynaco A25's.

 

[larryderouin]

clean the P.I. sockets with DeOxit, and then tighten the leaves in the socket with a dental pick. TEST. If Ok you fixed it. If not, Swap the 2 12ax7 tubes all the way in the Front center (low/high tone amp & Voltage amp tubes) with the Phase inverter tubes. Or if you have a pair of spare tubes, swap them in.

 

[brad44]

I got some extra money which I might splurge on 4 matched tubes for this and adding the 10-Ohm resistors like you described above. Im still trying to wrap my head around the milliamp conversion though, By doing the math, it will tell me if the tubes are balanced.

But when it comes to setting the bias, you would still measure pin 3 to ground (not across the resistor) and set the voltage with the bias pot as usual?

Having the 4 matched tubes means that the voltage stated in the manual would be acceptable.

 

[gadget73]

yep, volts and ohms tells you amps. The value of the resistor can be almost anything but it makes the math really easy if you make it 10 ohms. 0.350 volts across 10 ohms is 0.035 amps, or 35 milliamps. The 10 ohm makes it so all you have to do is shuffle around the decimal point.

If you want the real math, its amps = volts / ohms. Using that you can calculate any of the 3 if you know the other 2. Ohm's Law is the name for the magic juju. Its actually dead simple to work out, though I will admit to cheating and referring to a picture all the time to tell me what to do.

I have a big version of that as my desktop wallpaper. I use it pretty often.

 

[brad44]

To clarify though..... Even with the resistor in place, ultimately you set the bia the same way.... measuring pin 3 to chasis, correct?

The addition of the resistor is to see if the tubes are matched and confirm that the bias voltage is evenly distributed.

 

[gadget73]

you measure across the resistor, not to chassis.

 

[mibag6]

If you get the resistors, I would like to suggest that you either order extras and hand match them so that they are all very close tolerance or get very close tolerance resistors like1%.
Also, I don't know how critical the line voltage is in regards to the bias voltage so I'll ask if this needs to be done with a variac at 117V or not?

 

[brad44]

In fact, I have 1% 10ohm resistors on hand, but I don't want to install them until I have grasped this concept which I have not.

Turning the bias pot will change the voltage across the resistor.

So using the .110 example above, this is where I am at in it understanding:

Say for Example you got .110v Move the decimal 1 place to the left and it's now ,011a because you divided Volts by resistance. .011a = 11ma. This is about 1/3 of what a normally operating 7868 should be at.

ok^

to get wattage you multiply the Plate voltage by the ma but use the .0xxa reading to get the correct wattage.

Where do I measure the plate voltage? Which pin to ground? (larry mentions pin 9 and 1 and I am confused here).

The 7868 has a max plate dissipation of 19W. You ideally want your wattage to be between 70% and 80%. 70% is about 13.5 and 80% is about 15.2

Once I get the plate voltage and multiply by .011 then I will get the wattage that needs to be within 70%/80% of 19W. It will be different for all 4 tubes so I want an average not in excess of 80 percent. This is achieved by the bias pot. I would be changing voltage across the resistor.

And then I would be properly biased (easier said than done).

 

[gadget73]

Pin 9 is the plate.
Pin 3 is the cathode.
Pin 1 is the screen.
Knowing what each of these voltages is will tell more about what is going on overall.

 

[mibag6]

I'm not sure what you are trying to accomplish here. Are you trying to determine if you need to replace the 7868s? If you are just trying to get the amp biased you were good at post 1. There are a couple of threads in this forum restoring the X100 B with lots of info on mods and upgrades which may have drawings and or pictures to give a clearer idea of what is being proposed by the other posters here. Just a thought.

 

[brad44]

I'm not sure what you are trying to accomplish here

Bias a new matched quad of 7868 and how to do that with the 10ohm resistors in place.
Also read the 10ohm resistor is protection in case a tube failed and is a standard modification for most tube amps.

 

[mibag6]

Set it referencing pin 3 to ground. Just measure 1, set your bias and be done.