Tech help needed. Different readings between direct plate, cathode & bias probes.

[jasonguitar]

Measuring voltage drop accross the primary & dividing by DCR plainly does not seem to work. I have never heard this method described for reading plate current before.

From Aiken Amps website: Biasing
"The plate current can also be measured by first measuring the resistance across each side of the output transformer primary (it will usually be different on each side) with the power off. Make a note of the resistance on each side, and then, with the amplifier on, measure the DC voltage drop across each side of the output transformer. Divide this number by the previously measured resistance, and you end up with the plate current for the tubes on that side. Again, if there is more than one tube on each side, you must divide the total current by the number of tubes. This method is extremely accurate, and much safer than the shunt current measurement method, because a slip of the probe won't short anything out due to the high resistance of the voltage measurement setting on the meter compared to the very low resistance of the current measurement setting. You can also make a safer measurement by clipping the negative side of the voltmeter on ground, and measuring the center-tap voltage of the output transformer and the voltage at the plate of each output tube. Subtract the plate voltage from the center-tap voltage and you have the voltage drop across each side, and can then use this to calculate the current in each tube, again dividing by the number of tubes on each side."

[MWJB]

I have just tried it too, different results between the 2 methods, don't trust the method Aiken describes...your 6V6s are not dissipating 20W each.

Does your amp draw its B+ supply & screen supply from the same node? If not (nodes are separated by a dropping resistor & each has their own filter cap), then shunt the OT. Use meter leads terminating in clips or wire grabbers, red lead goes to B+, black to pin 3 that you are measuring...connect up & disconnect whilst unplugged from the wall, DO NOT short pin 3 to pin 2, DO NOT try measuring voltage with the meter still set to "<200mA dc"

[WholeToneMusic]

The transformer shunt method can be very dangerous if your not completely careful as MWJB describes however it's the most accurate way. One cool thing about it is you don't have to worry about resistance tolerances being exact as with other methods such as bias probes or installing 1 ohm resistors cathode to ground in a fixed bias amp.

[jasonguitar]

The transformer shunt method can be very dangerous if your not completely careful as MWJB describes however it's the most accurate way. One cool thing about it is you don't have to worry about resistance tolerances being exact as with other methods such as bias probes or installing 1 ohm resistors cathode to ground in a fixed bias amp.

There are such drastically differing opinions regarding the accuracy of these bias methods! I just read this post from a gearpage "senior member" in a related thread:

"The shunt method is dangerous and inaccurate and should not be used or recommended, at all.
Use the transformer-voltage-drop method instead.

You've already measured the resistance of each half of the primary - now all you need to do is measure the voltage drop across it, and divide the voltage by the resistance to find the current.

This is far safer because the meter is set to voltage and so has a high resistance, so if anything goes wrong it's much less likely to short anything.
It's also far more accurate because you're not relying on the resistance of the meter to be effectively zero, which you already know it isn't - but which is critical for the shunt method to work. And because by not running the full primary current through the meter, you're less likely to cause instability which will render the result meaningless. In fact, the voltage-drop method is the most accurate, because it measures the true plate current without including the screen current, as well.

It's still fairly dangerous in the sense that the points you're measuring are at the full B+ voltage (even though the voltage difference is only a few volts at most), so be careful anyhow. "

What the hey?

[MWJB]

Did you shunt the transformer? With the high DCR of your OT winding it will be the MOST accurate method, there shouldn't be any screen current at all in the result. But check that the B+ & screen supply are not derived from the same power supply node.

You may still see a slight variation between the shunt method and the cathode resistor method, they never tally exactly.

Being a senior member on a forum just means that you post a lot, it's not a qualification. He's probably just repeating what you read at Aiken's page. Yes, the shunt method is dangerous, but you're poking around the highest dc voltage in ther already...how much worse can you make it? Use clips on your meter leads & triple check tangibility of connections and for any risk of shorts before powering up the amp...you'll know if you have shorted the plate to the heaters...there will be fire, short the plate to ground and there will be huge sparks. So do it right...or just trust the cathode resistor method...at the voltages you have & the plate dissipation a handful of mA here or there isn't the end of the world. If you want to eliminate any margin for error, temporarily remove thew cathode bypass cap to be sure that the cap hasn't failed & is skewing the cathode voltage.

[jasonguitar]

Did you shunt the transformer? With the high DCR of your OT winding it will be the MOST accurate method, there shouldn't be any screen current at all in the result. But check that the B+ & screen supply are not derived from the same power supply node.

You may still see a slight variation between the shunt method and the cathode resistor method, they never tally exactly.

Being a senior member on a forum just means that you post a lot, it's not a qualification. He's probably just repeating what you read at Aiken's page. Yes, the shunt method is dangerous, but you're poking around the highest dc voltage in ther already...how much worse can you make it? Use clips on your meter leads & triple check tangibility of connections and for any risk of shorts before powering up the amp...you'll know if you have shorted the plate to the heaters...there will be fire, short the plate to ground and there will be huge sparks. So do it right...or just trust the cathode resistor method...at the voltages you have & the plate dissipation a handful of mA here or there isn't the end of the world. If you want to eliminate any margin for error, temporarily remove thew cathode bypass cap to be sure that the cap hasn't failed & is skewing the cathode voltage.

I hear ya. I've used the shunt method in the past. Not a prob. Just got accustomed to using probes and direct cathode current measurement which, until this amp, has always given me the same readings. What about the input impedance of the meter? Mine is 10M standard. As long as its significantly higher than the OT impedance there will be no question as to the accuracy, correct?

Won't be at the bench probably till Mon morning but I'll check it then or sooner if I get a minute. Cathode cap is a brand new sprague which I tested to be sure. I also did the measurement previously with the cap removed and got the same result.

[MWJB]

"What about the input impedance of the meter? Mine is 10M standard. As long as its significantly higher than the OT impedance there will be no question as to the accuracy, correct?" That's when reading voltage, so the meter doesn't affect the reading, when reading the current, the meter should show less resistance than the OT winding, so that all current is shunted through the meter.

[jasonguitar]

"What about the input impedance of the meter? Mine is 10M standard. As long as its significantly higher than the OT impedance there will be no question as to the accuracy, correct?" That's when reading voltage, so the meter doesn't affect the reading, when reading the current, the meter should show less resistance than the OT winding, so that all current is shunted through the meter.

Understood. Thanks.

[g1]

If you can, could you try these things:
1) current meter in series with cathode resistor
2) current meter in series between plate and OT
3) shunt method current readings
4)shunt method reading with plate disconnected from OT for side being measured.
I will be trying these methods later this week as I was having problems verifying bias methods myself and would like to get to the bottom of this.
My problem was that all methods were in agreement except the shunt method. Placed 1 ohm resistors between cathodes & ground. Measured voltage agreed with current meter in series with cathode (screen currents were accounted for). Measured OT resistance (around 150R per side), calculated current from voltage/resistance, agreed with cathode resistor & cathode current methods.
**Only shunt method readings disagreed with all other methods (showed lower current).
My conclusion was that my meter does not have a low enough resistance in current ranges to completely "shunt" the OT. This is why I want to try it with side of OT disconnected and then again in series between plate and OT.
I would recommend anyone using shunt method to check that their shunt measurements agree with series current readings.
Will try to post results later this week.

[MWJB]

I really don't recommend measuring current with loose OT wires hanging around, it's asking for trouble. Both probes are way above ground potential. Methods that involve calculating/reading from cathode current/voltage are fine but not always perfectly accurate, they normally involve some assumption as to resistor values & screen current, however you shouldn't really be running so hard that the screen current is enough to push things over the edge. With a 150ohm primary DCR the shunt method is likely to be most accurate.

What would I do? Add 1 ohm (exactly) 2W resistors between the 6V6 pin 8s and the 100ohm cathode resistor, measure dc voltage drop accross these resistors (not from pin 8 to ground).

For a cathode biased amp running 6V6 in the low 300's, I wouldn't lose too much sleep about knowing "exactly" what the current is, if the amp sounds OK & the highest measurement is not much over 40mA, you are fine.

[jasonguitar]

So the shunt method current reading is giving me pretty much the same result as the measuring the voltage drop across the primary method. I've got 70.2 ma on the Left and 68.5ma on the Right. My probes read 74.2mv Left side and 78.2mv Right side so although the reading is higher, its not completely out of the ball park. I don't understand why the voltage drop across the cathode resistor measurement gives such a different result. 6.66v/100.2 ohms = 66ma/2 tubes 33ma per side.

If I plug in the data I have to the DIY Guitarist Excel Power Tube Plate Dissipation Calculator I get the following results:

1 CATHODE CURRENT
a. Cathode Voltage: 6.66
b. Cathode Resistor (Ohms): 100
Cathode Current: 0.067
Conversion to Milliamps: 67

2 SCREEN CURRENT
a. B+ Supply Voltage to
Screen Grid Resistor: 270
b. Voltage at Power
Tube Screen Grid: 252.5
c. Voltage Drop Across
Resistor: 17.5
d. Screen Grid Resistor
Value: (Ohms) 1000
Screen Grid Current: 0.0175
Conversion to Milliamps: 17.5

3 PLATE CURRENT
a. Cathode Current
(value from above): 0.067
b. Subtract Screen Current
(value from above): 0.0175
Plate Current: 0.049
Conversion to Milliamps: 49

4 PLATE DISSIPATION
a. Plate Voltage 293.5
b. Plate Voltage minus
Cathode Voltage: 286.84
c. Multiply by Plate
Current: 0.049

Plate Dissipation (Watts): 14.083844

[g1]

It would be good to verify the cathode reading by inserting the resistors as suggested by MWJB. Still it would be nice to know why you are getting a reading roughly 3 times higher at the plates. If it were a small difference it would be easy to dismiss it, but 300%? As was stated above, they can't be dissipating 20W each or you would be seeing some very red plates.
I don't think anyone here wants to throw out the shunt method as inaccurate. It's as if the OT is feeding something else along with the plates.

P.S. Remember having very illogical results one time due to a weak battery in my meter, just a thought.

[jasonguitar]

It would be good to verify the cathode reading by inserting the resistors as suggested by MWJB. Still it would be nice to know why you are getting a reading roughly 3 times higher at the plates. If it were a small difference it would be easy to dismiss it, but 300%? As was stated above, they can't be dissipating 20W each or you would be seeing some very red plates.
I don't think anyone here wants to throw out the shunt method as inaccurate. It's as if the OT is feeding something else along with the plates.

P.S. Remember having very illogical results one time due to a weak battery in my meter, just a thought.

Yeah, it seems to me that the cathode current measurement is probably correct based on the fact that the amp sounds good and has been run for about 20 to 30 mins a couple times now and played at full blast with no adverse affects, glowing plates, noise, hum etc. It can't be running as hot as the shunt and bias probes seem to indicate.

My batteries are fresh and a second meter displayed the same results within a couple ma.

Hard to determine whether I have any resistors that are EXACTLY 1 ohm. Even though I have several 1ohm 1% resistors, I get slightly unstable readings measuring 1 ohm on both meters I have currently.

[MWJB]

I'm still wary of the OT too, did we ever get round to determinining a turns ratio? Weren't you getting 470VAC from end to end of the primary, with 6.3VAC (measured) injected at the secondary? 470/6.3= 75, 75*75= 5625...even with a 2ohm load that's a primary Z of 11.25K...seems very unlikely.

[MWJB]

Going back through the posts, it seems that your plate voltage rose when you fitted a smaller cathode resistor, this is not right, it should have dropped.