diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Tubes / Valves (http://www.diyaudio.com/forums/tubes-valves/)
-   -   10M45S problem (http://www.diyaudio.com/forums/tubes-valves/229292-10m45s-problem.html)

scott17 3rd February 2013 10:36 PM

10M45S problem
 
My first attempt using the 10M45S as a CCS.
The driver is the pentode section of a 6GH8A running at 3.5mA using LED cathode bias with a 100R resistor to ground and gNFB applied at the junction of the two.

The circuit works fine with a resistor in the anode instead of the CCS. When I switch in the CCS in place of the anode resistor I'm getting some sort of an extreme low frequency feedback or something on the output of the amp.

Before I ran a source through the amp, I had dummy loads connected to the output. I checked the voltages and current with the CCS in place and all measured normal. Correct voltage drop, correct current. I am using a 1K stopper on the gate as suggested by tubelab.

Has anyone had this problem or know where I should start looking?

Thanks,

Scott

AJT 3rd February 2013 10:55 PM

motorboating? check your psu decoupling caps, earthing lay-outs....a scheme we can look at will help....

FWIW, mosfets are better used at currents of more than 10mA, cascode bjt's are better at currents less than 10mA.....this is what i gathered from postings here although i have yet to try it myself....

scott17 3rd February 2013 11:23 PM

Thanks Tony.

Not sure I would call it motorboating. It's more a very loud low frequency howl. The circuit does work fine with a resistor in the anode instead of the CCS.

I'll come up with a schematic eventually, as most of it is on scratch paper.

Scott

Miles Prower 4th February 2013 12:52 AM

Can't load a pentode with a CCS. Pentodes are also CCS's, and the one will fight the other, causing the oscillation you experienced. If you're gonna do that, you need to parallel a CCS with a plate resistor.

iko 4th February 2013 12:58 AM

Turn it into a gyrator, that should work, no?

cotdt 4th February 2013 12:59 AM

Have you tried a higher current?

scott17 4th February 2013 01:03 AM

Well thanks Miles. That would explain it. Does this also hold true if the pentode is wired as a triode? The pentode has a 510R resistor from the plate to the screen.

If I wanted to parallel a CCS with the plate resistor, how do I determine the current sharing between the two devices? Is this method discussed somewhere?

Thanks,

Scott

cotdt 4th February 2013 01:06 AM

If triode-strapped, it should work fine. I've tried it with a lot of different tubes.

scott17 4th February 2013 01:12 AM

Quote:

Originally Posted by cotdt (Post 3354892)
If triode-strapped, it should work fine. I've tried it with a lot of different tubes.

The 10M45 data sheet indicates that it will work down to 2mA. I think I will try a higher current and see what happens.

Thanks.

cotdt 4th February 2013 01:25 AM

Quote:

Originally Posted by scott17 (Post 3354898)
The 10M45 data sheet indicates that it will work down to 2mA. I think I will try a higher current and see what happens.

Thanks.

I've only used it down to 8mA myself, so I don't know how it works at lower currents. Hence my recommendation to try a higher current.


All times are GMT. The time now is 01:54 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2