DCAudioDIY.com
http://dcaudiodiy.com/phpBB3/

Current Sensing of Output Tubes
http://dcaudiodiy.com/phpBB3/viewtopic.php?f=5&t=1979
Page 1 of 1

Author:  FerdinandII [ March 18th, 2021, 10:53 pm ]
Post subject:  Current Sensing of Output Tubes

Walt has mentioned some advantages to tying the cathodes of output tubes directly to ground. I am interested if anyone has ideas on how to measure the bias current without a resistor between the cathode and ground. Something like a hall effect sensor, maybe? I want to see if there's another option besides putting a jumper in there and switching in a 10 ohm resistor for measurement duties.

Author:  ratbagp [ March 19th, 2021, 12:28 am ]
Post subject:  Re: Current Sensing of Output Tubes

Measure the voltage across the output transformer and divide by the resistance of the wiring in the transformer.

ray

Author:  David McGown [ March 19th, 2021, 7:43 am ]
Post subject:  Re: Current Sensing of Output Tubes

Ray,

That is difficult to work with tubes and UL connections, since some current is passing through the screen. The total current is the sum of the plate and screen current, and the only place you can measure that is at the cathode.

For triodes you can measure anywhere, at the plate, at the power input to the output transformer, at the cathode (of course).

Do not understand the reluctance of measuring across a fixed, known resistor at the cathode. A 10 ohm resistor is not going to affect the output tube performance, adding between 40 to 100 ohms to the overall plate resistance. You can measure the voltage drop (and therefore current) to high precision with a high input impedance meter that is parallel to the DUT. Many analog current meters are basically a volt meter measuring voltage drop across a reference resistor between the terminals. To me, using a resistor is the simplest method, and simple is better from a reliability and confidence standpoint. Also alot safer using meters or sensors that are not rated for high voltage measuring from a low voltage cathode position. On my GM70 amp with 1100V plate voltage, I have no way to safely measure current except at the cathode.

You can always use a precision 1 ohm resistor instead, and measure current in mV=mA. If your meter is sensitive enough, that may be the best if you are concerned with minimizing any power loss or added overally plate resistance by the cathode resistor.

David

Author:  SoundMods [ March 19th, 2021, 9:46 am ]
Post subject:  Re: Current Sensing of Output Tubes

FerdinandII wrote:
Walt has mentioned some advantages to tying the cathodes of output tubes directly to ground. I am interested if anyone has ideas on how to measure the bias current without a resistor between the cathode and ground. Something like a hall effect sensor, maybe? I want to see if there's another option besides putting a jumper in there and switching in a 10 ohm resistor for measurement duties.

A digital multi-meter in series with the ground and cathode set to milli-amps gets the job done. The easy way is using a bias meter. The meter sockets are between the power tubes and the amp and reads directly in cathode current. You can look at all four tubes in one sitting. I have one myself and it has served well when I had a push-pull amp and when I helped a friend out with his.

Here is but one example of many:
https://www.amazon.com/Vacuum-Tester-Am ... 0849&psc=1

Author:  David McGown [ March 19th, 2021, 10:08 am ]
Post subject:  Re: Current Sensing of Output Tubes

But it is so easy to have test points with permanently installed resistors on the cathode on each tube. You don't have to shutdown the amp, remove the power tubes and plug them into the test sockets. If you put a test point next to the tube socket on the top of the amp, connect the cathode pin to the test point, and install a 10 ohm (or 1 ohm) resistor to ground at the test point, then you can get individual bias readings with the amp in operation, to allow touching it up after it warms up, checking for bias while in operation. You don't even need a switch. You can use a switch if you have an installed meter on the amp, otherwise connect the (-) probe to the speaker ground terminal, and measure at each test point. Easy Peasy.

This all assumes one is designing and building an amp from scratch, and not trying to implement bias measurement on an existing amp and disturb the cosmetics.

Should like to point out that switching out the cathode connection for installation of a meter in-line does have risks, not the least of which is contact reliability and forgetting to switch back to normal connection.

David

Author:  SoundMods [ March 19th, 2021, 10:52 am ]
Post subject:  Re: Current Sensing of Output Tubes

David McGown wrote:
But it is so easy to have test points with permanently installed resistors on the cathode on each tube. You don't have to shutdown the amp, remove the power tubes and plug them into the test sockets. If you put a test point next to the tube socket on the top of the amp, connect the cathode pin to the test point, and install a 10 ohm (or 1 ohm) resistor to ground at the test point, then you can get individual bias readings with the amp in operation, to allow touching it up after it warms up, checking for bias while in operation. You don't even need a switch. You can use a switch if you have an installed meter on the amp, otherwise connect the (-) probe to the speaker ground terminal, and measure at each test point. Easy Peasy.

This all assumes one is designing and building an amp from scratch, and not trying to implement bias measurement on an existing amp and disturb the cosmetics.

Should like to point out that switching out the cathode connection for installation of a meter in-line does have risks, not the least of which is contact reliability and forgetting to switch back to normal connection.

David

The bias meter is temporary -- you set it up -- let the amp heat up for about a half-hour -- then go about setting your bias. What makes the bias meter a valuable addition to the tool box is that with all four tubes plugged in at once, you get to see how setting the bias of one tube affects the others. You can "diddle" with the settings until all four tubes are reading the same quiescent cathode current. You can even ear-test the results with the meter installed. That way you can find the "sweet spot" for the bias as I did. Once the bias is set this way you're good to go for the long term.

Author:  David McGown [ March 19th, 2021, 11:49 am ]
Post subject:  Re: Current Sensing of Output Tubes

Walt,

That does look handy, particularly for someone involved with repair/modification. I may actually pick one up to play with.

BTW, one thing I like to have is bias balance pot. I set overall bias level for a PP pair combined, then measure and adjust the balance (null) between the tubes at their respective test points. The overall bias level I only need to set infrequently, usually on a tube change, but sometime balance can drift a bit over time and nice to touch it up easily. But I build all my amps rather than working with commercial amps where the designer already has setup how to adjust bias. I am sure there are +/- with either approach.

Of course, this only works for single PP pairs, not parallel PP pairs. Gets to be a bigger problem overall when using 3 or 4 PP pairs on BIG tube amps.

David

Author:  SoundMods [ March 19th, 2021, 12:04 pm ]
Post subject:  Re: Current Sensing of Output Tubes

David McGown wrote:
Walt,

That does look handy, particularly for someone involved with repair/modification. I may actually pick one up to play with.

BTW, one thing I like to have is bias balance pot. I set overall bias level for a PP pair combined, then measure and adjust the balance (null) between the tubes at their respective test points. The overall bias level I only need to set infrequently, usually on a tube change, but sometime balance can drift a bit over time and nice to touch it up easily. But I build all my amps rather than working with commercial amps where the designer already has setup how to adjust bias. I am sure there are +/- with either approach.

Of course, this only works for single PP pairs, not parallel PP pairs. Gets to be a bigger problem overall when using 3 or 4 PP pairs on BIG tube amps.

David

With parallel PP you do one channel at a time. ;) If the tubes and passive parts are matched I expect you can look at the single PP pairs and presume the pair not observed are obediently following.

Author:  Stuart Polansky [ March 19th, 2021, 1:21 pm ]
Post subject:  Re: Current Sensing of Output Tubes

Or use a terminal strip with a 1 or 10 ohm resistor across the contacts. In normal use, simply add a shunt across the terminals, either pre-made, or make your own. Incredibly simple.

Page 1 of 1 All times are UTC - 5 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group
https://www.phpbb.com/