Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Dave on October 22, 2010, 08:35:01 am
-
Hey fellas...... I'm confused.
I have always used the OT shunt method for biasing, but have decided to try the 1ohm cathode to ground method a lot of you guys use.
I tested the same amp under the same conditions today just to make sure that the numbers would come out right and they didn't.
The amp runs a quad of EL84 tubes fixed bias.
Shunt method ----- 43.8 mA side 1
----- 59.6 mA side 2
Cathode resistor method ------ 36 mV tube 1
------ 27.3 mV tube 2
------ 34.9 mV tube 3
------ 49.5 mV tube 4
Am I missing something?
Dave
-
I just noticed something wierd.
36 + 27.3 = 63.3 * .7 = 44.31 (43.8 w/ shunt method)
34.9 + 49.5 = 84.4 * .7 = 59.08 (59.6 w/ shunt method)
So, I am getting roughly 30 percent higher bias readings by using the 1ohm resistor method.
Why?
Dave
-
are you using a 1% one ohm resistor?If it's not exactly one ohm the readings will be off. 1.4 ohms is not close enough.
And the shunt method is fraught with problems.The meter you use may have reading issues with that much current going through it or not all the current is being shunted through the meter.
Try this other method:measure ohms from pin 3 to the OT centre tap with the amp off and caps drained.Then fire the amp up and measure voltage between the same two points.Then divide the voltage by the ohms to get your MV/MA results.
e.g. say 1.7 v / 49 ohms = .0347 or 34MA.If the 1 ohm method is closer to this method then it's right.If it's closer to the shunt method then you don't have exactly one ohm resistors.
-
Not sure about it but while using shunt method, will it read screen current as well or just plate current? I know 1 ohm bias sensing resistors read both, how about O.T. shunt method?
Hope this helps.
Best Regards,
Rzenc
-
Not sure about it but while using shunt method, will it read screen current as well or just plate current? I know 1 ohm bias sensing resistors read both, how about O.T. shunt method?
Hope this helps.
Best Regards,
Rzenc
I was about to say that the shunt method and the OT method Physconoodler describes both measure plate current only while the cathode "test resistor" includes screen grid current as well.
Also, the accuracy of any testing depends on how good your instrument is. My DMM isn't all that great when it comes to small MV readings - I trust my analog FET meter more frankly. I had been thinking about using a 10 ohm resistor on the cathode for future builds because of accuracy concerns and it would make any test lead resistance irrelevant. However, two of Hoffman's bias checkers makes biasing a breeze.
HTH
Chip
-
"However, two of Hoffman's bias checkers makes biasing a breeze."
Agreed! With as much money, time, and effort as we spend making amps, not having one of these doesn't make a whole lot of sense. :shocked:
-
I had been thinking about using a 10 ohm resistor on the cathode for future builds because of accuracy concerns and it would make any test lead resistance irrelevant.
All half-decent DMMs allow you to zero the readings with shorted leads - Fluke standard leads show 0R36 - a very large value when compared to 1R, and zero after pressing the magick button.
-
e.g. say 1.7 v / 49 ohms = .0347 or 34MA.If the 1 ohm method is closer to this method then it's right.If it's closer to the shunt method then you don't have exactly one ohm resistors.
OK, I did this test. The results came out within 2% of the shunt method on the original meter. The problem is that I am using 1% tolerance 1 ohm resistors and I have measured them with 3 different meters just to make sure.
Also, I have now done both the original tests again but this time I used 3 different meters from three different manufacturers.
All three meters showed results within 1% reading mV on the resistor method.
But....... They had wildly different readings for the shunt method.
51 - 72 Meter # 1 shunt method
47.5 - 67 Meter # 2 Shunt method
42.5 - 59 Meter # 3 shunt method
I am seeing here that the shunt method doesn't seem to work so well. (Been doing it for years. Oops!)
Now, though, I wonder if I should take into consideration that the resistor method figures in screen current and if I should adjust to compensate.
Dave
-
I guess my question is......Does it really matter? An amp needs to be biased to where it plays and sounds best. So long as the tubes do not red plate, does it matter if the tubes are 45Ma and 70% or 38Ma and 59%. All the fretting over bias numbers has always seemed like a non-issue to me......
Of course, I could be wrong.......
-
So long as the tubes do not red plate, does it matter if the tubes are 45Ma and 70% or 38Ma and 59%.
Of course.
:wink:
-
Hello,
you should consider that the measurement via the 1R resistor not only has the anodecurrent flowing through it but also the screen-grids current.
Thatīs why your readings differ.
- Swen