when checking bias with a meter in series...

[daz]

....with the plate, is the current you read the correxct current or do you have to account for the screens. And if so, how? Add 5 ma, subtract? In the weber cathode bias current calculator it says "In these calculations, 5% of the cathode current is assumed to be screen current". I'm not sure whether they are saying that it's already included in the results or not. Anyways, when i checked the current with my meter in series with the plate, i got 5 less ma than the calculator. I calculated my amp's current with the sag resistor switched in and got 54 ma, but my meter showed 48.8. I then checked it with the sag resistor switched out and got 63 ma with weber's calc, got 56.8 with my meter. Both readings about 5 ma less than weber showed. So which is the correct number...what weber's calc showed or what my meter did? heres all the exact numbers...

WITH SAG RESISTOR SWITCHED IN
------------------------------
plate current: 48.8 ma
plate voltage: 346 VDC
cathode voltage: 25.5 VDC

weber calc plate current:54 ma




WITH SAG RESISTOR SWITCHED OUT
------------------------------
plate current: 56.8 ma
plate voltage: 399 VDC
cathode voltage: 29.7 VDC

weber calc plate current: 63 ma

From weber's tables i deduced that without the sag resistor switch in, the current, at least as weber shows it is right at 100%. My meter shows what would probably be around 85-90% w/o checking it exactly. I've been told it's fine to run cathode biased amps at 100% at idle, then again some people say it's nuts. I have no idea what to believ, tho my best guess is it's ok since i've used it like this for a long time with no ill effects and the tubes still sound new.

[SatchmoeddieII]

First question. Did you buy the bias meter, or build one? The makers of one of the bias meters have some base numbers. I ran an old fixed bias Twen at idle, and then full on with the 1000 Hz sine wave at around 500mv which is kind of hot, but pedals can put 3X that much voltage in. When I did all my calculations I got something like 18 watts and with signal 27. I have found that the plate current as measured with the meter in series is very close to what actually comes all the wat to K. I can try putting a bias pro/king/whatever on one of my tube testers, and compare its reading to that of the tester. The actual readings from the Simpson watt meter, the bias pro, and the plate voltage were really pretty much right on the money with the 1000Hz signal. How accurate the Simpson wattmeter was is a mystery, as I have never calibrated it, but the 260 it runs into is right on.

[daz]

It's just a basic a voltage meter and i soldered the leads from plate pin and OT lead and set to read ma.

[Wakculloch]

From your numbers, and i'm assuming your using 470R cathode resistors, it would appear that the weber calculation does include the screen current, and that this current is 5.2 and 6.2mA respectively.

[Robert M. Martinelli]

daz,
if you put your current meter in series with the plates, you're measuring the plate current alone, and that's why you're about 5 mAmps off from the Weber calculator, which takes the screen current into account instead. Should you put your current meter in series with the cathode, you'd read the TOTAL current flowing through that tube ( which is the sum of plate and screen current ), and that's exactly the current that tube is drawing from the PS ( and your readings would match ).

Cheers

Bob

[Enzo]

What's a Weber calculator? Is that what you are calling his Bias Rite? Or is it something else? Oh, I found it.


Assuming use of the Bias Rite, it even calls itself a cathode current monitor. COnsider what you are measuring. When you break into the plate lead, all you CAN measure is plate current. When you break into the cathode lead, you measure the current through the cathode. The cathode has no idea where the current goes or comes from, all it knows is what is flowing right there. The current flow from the cathode usually goes to the plate mostly and some to the screen grid.

The bias rite simply measures the cathode current, nothing more nor less. When they say that they assume 5% of the cathode current is screen current, then assume that whatever reading you get, 5% of it was from the screen.

But the calculator tells you what it does. It asks for the plate voltage, then it asks for CATHODE current. Try your example. I entered 369v in the plate volts box, and 63ma in the cathode current box, then I clicked "Calculate power." It filled in the power box with 22 watts, but note it also automatically reduced your 63ma to 60. SO If you want to know what your plate current is from a cathode current reading, you can subtract about 5% from it. If you use his calculator, just enter the numbers it asks for and the result comes out.

His 5% estimate of 3ma is not exactly your 5ma reading, but it is close enough for amp work. If you are concerned over a couple percentage points, it is easy enough to take the readings and use a pocket calculator. That is what I did below.

So whatever you measure at the plate will also be flowing through the cathode. But the screen current also flows through the cathode. SO the plate reading will be plate current. The cathode reading will be plate current and screen current added together.

I can;t get the quote thing to work now, grr...

You said:

WITH SAG RESISTOR SWITCHED OUT
------------------------------
plate current: 56.8 ma
plate voltage: 399 VDC
cathode voltage: 29.7 VDC

weber calc plate current: 63 ma

I assume you mean the Weber said 63ma of CATHODE current? In any case, you have 399v on the plate and 30v on the cathode, leaving 369v across the tube. 369v x 56.8ma = 20.9 watts. What tubes are you running? 21 watts is about right for a 6L6.

You said:

WITH SAG RESISTOR SWITCHED IN
------------------------------
plate current: 48.8 ma
plate voltage: 346 VDC
cathode voltage: 25.5 VDC

weber calc plate current:54 ma


346v on the plate less 26v on the cathode means 320v across the tube. And with 50ma plate current 320v x 49ma = 16 watts.

[SatchmoeddieII]

A bias probe is fairly easy to make, but the companies that make some of them have these large diameter tube bases that a socket fits into flush with the bottom. They work well with some amps, but not others with the Fender type base clips. All they are is a wire leading from the cathode socket to a 1 ohm resistor, and back to the tube base. You hook up a digital volt meter (nothing too fancy is needed, and use the 100 milivolts scale. With Ohm's law being what it is and the resistance is 1 than mV=mA. The same can be done interrupting the cathode, and using a mA meter. I bought a "Dual bias tool, and a plate volt tool. The dual has a DPDT on off on switch with either one socket or the other under test. In some amps my home made probe was too tall to fit. It is a socket, and tube base that is stuck together and the probe plugs into the tube socket, and the tube into the probe socket. The plate volt meter has a tap from the plate pin 3 and cathode pin 8 on 6V6 6L6 KT66, 6CA7 etc. This is plugged into the digital volt meter and the voltage from plate to cathode is read. I have used it at idle, and with a .5 volt 1KHz signal, .05 volt 1KHz etc. I just use my old Fluke model 74 from my electrician days, or some other bench or hand held, but use either a digital, or Vacuum tube volt meter, and not a Simpson 260, as those only load to 20,000 ohms per volt, and will throw things off more with that measurement(s). I do use the Simpsons, or my Triplet when doing old stuff where a 20K ohm/ volt meter is specified. These are easy to make. As I said buy 3 tube sockets, and 3 tube bases for each type you want to test. Only 7027s, and 7591s are different from all the common stuff. For EL 84s and 7189s a pair of 9 pin socket savers and some PVC pipe will work. Anyway take pin 8 (THE CATHODE PIN) and run to a small box, and get a mini DPDT switch, and run the wires through a 1 ohm 1/2 watt resistor 1% tolerance, and put the leads to go to the tester on the center terminals, and run those to banana plugs. Use the DPDT switch to select the tube you want to test. Plug into the volt socket, and test DC millivolts for idle, or AC for a test with a signal. I am unsure how high a frequency a handheld DVM will go. I use a Fluke 8842 with 3 freq ranges up to low band RF. It is in need of repair now, but a quick slap upside the chassis makes the display come back. My personal email is seb-ear-aaaa@cox.net. I would be glad to mail you a diagram, and schematic of how to make a bias tool. I paid $98 for mine, but I could not find those really large tube bases, and I needed the vertical clearance, and was getting really tired of making leads, and test fixtures, and not fixing stuff for paying customers. Now my only customer is the guy with the amp from he$$. I am at my wits end with this POS. It is a Sound City PA 120 to DR 103 mod. I am getting an overload everytime I hook up the B+. I took the output trans out, and still 8 amps. 4 is more in line. The claim by Aspen Pittman that "many of the SCs had partridge x-fmrs has people showing up, and not wanting to heat that theirs is an EV, and not a partridge. They also fall about 85 volts short on the B+ of a Hi Watt. I shall never take another mod job that someone else has started. NEVER EVER, but money can make me change my mind. Anyway I hope this helps some. Test across the cathode, and making one of those probes that uses your digital volt meter is not hard. Thanks! TTFN

[daz]

daz,
if you put your current meter in series with the plates, you're measuring the plate current alone, and that's why you're about 5 mAmps off from the Weber calculator, which takes the screen current into account instead. Should you put your current meter in series with the cathode, you'd read the TOTAL current flowing through that tube ( which is the sum of plate and screen current ), and that's exactly the current that tube is drawing from the PS ( and your readings would match

Thanks, thats what i needed to know. But a bit confusing because i was told once when i had 1 ohm resistors on the cathodes that if i measure across them to get plate current i need to add or subtract (can't recall which) screen current.

I assume you mean the Weber said 63ma of CATHODE current? In any case, you have 399v on the plate and 30v on the cathode, leaving 369v across the tube. 369v x 56.8ma = 20.9 watts. What tubes are you running? 21 watts is about right for a 6L6.

63ma is what the weber calculator showed for "Approximate plate current in each tube" when i entered the voltage drop across the cathode, the resistor value, (450R) and the plate to cathode voltage. I'm using el34's, and it showed what is 100% current according to thier scales and 25 watts. What am i missing?

SatchmoeddieII....thanks for the offer, but i'm fine with what i have. But as i mentioned to robert, when you measure across a 1 ohm resistor on the cathode, don't the results have to be compensated for by adding screen current? I was told that before.

[daz]

I put the meter in series with the cathode. It showed about 10 MA more, tho i'm using a different meter. What worries me is this. I read before that with cathode bias the amp's max current is at adle. But i strummed a guitar thru it and the current jumped as high as 124ma. WTF? At this point i have no idea wht to think because if thats the case even if i biased it at 70% at idle it would jump well past 100%. What am i missing here?

[SatchmoeddieII]

Not when you are on the cathode pin. I think someone else explained that. If you like send an address, and I will be more than happy to send instructions for how to make a set of these. I had to make a second set with narow tube bases, and tube sockets so it fits the Fender stuff with those base clips. I also was kind of in a pinch with making test fixtures for Hickock 539, and KS15750 tube testers for calibration. An 11153 ohm resistor that is + or - 10 ohms is a pain to make, but it did the trick. I also needed 6K ohms + - 10 ohms, and other values at either 1% or 10%, and a dummy rectifier. The regulator section was the only real problem on those testers. My Triplet 3444 only takes two fingers, and a thumb to calibrate. I must have gotten something right, as they regularly test within 2% of eachother now. The bias pro,king dual bias etc all use a 1 ohm resistor and the 100mV DC scale of a digital volt meter, and the 1 ohm is inline with the cathode pin #8 on the EL34, 6L6, 6V6, 5881, KT66, KT77, KT88, but the EL 84 uses 9 pin miniatures. For plate voltage a lead comes off pins 3 and 8 for the common octal tubes including 6550. For 7591 pin 3 & 5 is used for voltage reading, and 5 forcurrent, but the current reads on a mV scale, and is straight across converted to mA. EL-84 mA is on pin 3, and voltage between p & K is pins 6 & 3. 6973 is a 1 ohm across socket & pin for pin 7, and voltage is between pin9 & 7. Those are about all I use myself. I can no longer find 8714s, and convert those to 6550s. These are easy to make, and work quite well. I cut up socket savers for the El-84, and 6973 fixtures. I am a bit of a Valco freak, so Supro, Valco, and Gretsch amps that use the 6973 kind of attract my interest. I saw a 6V6 McKinney go for well over 800. I was bowing out of bidding, as by then I can aford the Mec. Mag. X-fmrs. or with the 6V6 I already have some nice sets from this and that kicking around. Which Weber book are you using for his conversions, and charts? I have the dark green paperback. Measuring the cathode amperage should get the stuff from the screens too. If you want some simple plans for the bias probes feel free. You have the email, I am already inundated with SPAM, and I lost internet service for a while, so I am way behind, and rather unhappy. My cable company has been screwing us for 23 years. We have put up with crappy pictures, and really poor service when digital came in, but they are the only game in town. 256K is a little too slow for me.

[daz]

Enzo....you were using the wrong calculator...well, either that or I was. But the one i used is under the one you used and i used it because my amp is cathode biased and the one i used says thats what it's for. I may be missing something, but it sure wouldn't be the first time.

[Robert M. Martinelli]

...What worries me is this. I read before that with cathode bias the amp's max current is at adle. But i strummed a guitar thru it and the current jumped as high as 124ma. WTF? At this point i have no idea wht to think because if thats the case even if i biased it at 70% at idle it would jump well past 100%. What am i missing here?

Who said that with cathode bias the max current is "at idle" ?

The current does not change significantly between idle and playing conditions only in a class A amp, because the tube(s) are always on, but in any AB1 class amp ( most of the amps out there in the real world ) the current does indeed swing from idle, or quiescent, to a maximum. The idle current, imposed by biasing the tube(s) is there to keep the cross-over distortion away, and it's normal for the current to swing from a minimum to a maximum in any AB1 amp.

Tubes can go over their max PD provided that the AVERAGE power the plate(s) dissipate is within their limits ( they can exceed 100% during the "on" condition because they're biased under their 100% at idle )

Hope I've managed to be clear enough

Cheers

Bob

[daz]

The current does not change significantly between idle and playing conditions only in a class A amp, because the tube(s) are always on, but in any AB1 class amp ( most of the amps out there in the real world ) the current does indeed swing from idle, or quiescent, to a maximum. The idle current, imposed by biasing the tube(s) is there to keep the cross-over distortion away, and it's normal for the current to swing from a minimum to a maximum in any AB1 amp

So it's ok that i was readung as high as 126ma when playing ? (that was with master way up)

[Enzo]

63ma is what the weber calculator showed for "Approximate plate current in each tube" when i entered the voltage drop across the cathode, the resistor value, (450R) and the plate to cathode voltage. I'm using el34's, and it showed what is 100% current according to thier scales and 25 watts. What am i missing?

Oh. I'm not doing very well here, am I?

I didn;t realize you had EL34s.

Your cathode resistor measures 450 ohms then? Or is it a 450 ohm resistor that we have not yet measured? Always check the value, because it affects your calculations. A 10% 450 ohm resistor can measure between 405 ohms and 495 ohms and be within spec.

You can measure cathode current by adding in a 1 0hm resistor and reading across it, like the bias probe adaptors do. Or you can measure voltage across the cathode resistor and calculate from Ohm's Law. It doesn;t matter, both ways work.

You are right, I was not looking far enough down the page, but the results should be similar. Looking at his calculator, enter 29.7v across 450 ohms into Ohm's Law and you get 66ma. His calculator shows 63ma for the plate. Clearly then his result has subtracted the 5% already.

When we "bias" an amp, we are setting idle current. As soon as you start playing, the amp is no longer idling. Current readings you get during play have nothing to do with setting the bias. The tube dissipation from idle is not related to power output.

[daz]

Thanks Enzo. Then let me ask you this...63ma is from my calculations 100%. Is there no amp aside from a true class A that should be at 100% at idle? I always seem to see people saying that with cathode bias you can set bias to 100% at idle. But if thats not the case, what should i do...bias it the same as fixed, 70%? By the way, i have seperate cathode resistors per side as you apparently realized, and they are precision so they are a true 450R each.