Limiting the grid current of PowerDrive - Page 2 - diyAudio
Go Back   Home > Forums > Commercial Sector > Vendor Forums > Tubelab

Tubelab Discussion and support of Tubelab products, prototypes and experiments

Reply
 
Thread Tools Search this Thread
Old 18th April 2010, 07:55 PM   #11
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Folks,

The issue with my 6J5, 6BX7, 300B amp is not OPT saturation as I initially thought. Monitoring the o'scope a bit more carefully reveals that it's more like blocking distortion. Following a pulse (say a stomp on a bass drum), the output voltage will almost "lock" to one voltage and slowly sag for about 250 ms at which point it catches up with the program material again.

I'm wondering if I'm driving the 6BX7 into grid current, thus, causing the coupling cap between the first and second stage to lose charge and cause blocking distortion. That would make perfect sense and what I see on the o'scope is roughly the same as what I've seen with a single common cathode stage driving the 300B without the source follower.

I don't recall what I used for R28. I might have used 680 ohm - I'm 99.9 % sure I did. Increasing that would cause the 6BX7 to get into grid current at a higher input voltage, which would cause the output stage to clip before the driver stage -- the whole point of having that second driver stage. I might give that a whirl.

George - thanks for reducing my worries about grid current in the 300B. The amp is intended for playing music. Sine waves get boring after a while...

I'm finding it rather hard to improve on the Tubelab SE, by the way. It seems everything I do has either lower gain, thus, requiring more input voltage than my preamp is designed to deliver or sounds "muffled" (2-stage driver) or "compressed" (mu-follower). I like the high-mu triode - 300B sound.

~Tom
Attached Images
File Type: png 6J5_6BX7_300B_Amp.PNG (30.5 KB, 243 views)

Last edited by tomchr; 18th April 2010 at 07:59 PM.
  Reply With Quote
Old 18th April 2010, 09:29 PM   #12
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
OK, so I may have to eat my words. <HOMER_SIMPSON_DROOL=MAX>Mmmm.... Words...</HOMER>

I am currently running with the circuit in attached schematic. Note a few minor tweaks (coupling cap between 1st, 2nd stage and current of 6BX7).

This sounds pretty darn good, actually.

The only difference between my current setup and my previous setup is the "layout" (aka rat's nest - DeathTrap2000(TM)) and possibly the value of the coupling cap between the 1st and 2nd stage. I don't recall if I used 1uF or 100nF before. But at least with this circuit, I'm not able to make the amp "lock" the output voltage -- except when the source follower bottoms out, of course.

/me like...

I'll listen to this for a while, then switch back to 5842-300B and see which I like better.

~Tom
Attached Images
File Type: png 6J5_6BX7_300B_Amp.PNG (30.1 KB, 219 views)
  Reply With Quote
Old 27th April 2010, 01:34 PM   #13
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Tom,
Have you tried to raise the value of R9? With the FET you should be able to raise the impedance and take a load off of the driver tube to maximize the gain and linearity while at the same time extend the bandwidth with a smaller C4.


Since you are in "breadboard stage", have you considered a ccs under the fet? I want to build a test bed like this to see the implication in distortion and performance. I really don't know if it is good or bad, just curious.
__________________
Living Life Doing the Waltz in 4/4 meter.

Last edited by SGregory; 27th April 2010 at 01:40 PM.
  Reply With Quote
Old 27th April 2010, 09:06 PM   #14
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Quote:
Originally Posted by SGregory View Post
Have you tried to raise the value of R9? With the FET you should be able to raise the impedance and take a load off of the driver tube to maximize the gain and linearity while at the same time extend the bandwidth with a smaller C4.
I have. And it seems there's an optimum value for R9 around 100~220 kOhm. Higher values actually cause more distortion. I'm not sure why...

Quote:
Originally Posted by SGregory View Post
Since you are in "breadboard stage", have you considered a ccs under the fet?
Someone else mentioned this a while back and I do plan to test it out. I'll report my results when done.

Right now I'm working on getting a high voltage regulator (LM317-based) to work. I figured that rather than using the darlington configuration used in the original circuit, I would use a MOSFET. But that causes trouble with instability. I need to order a few parts before I can get any further.

~Tom
  Reply With Quote
Old 27th April 2010, 10:28 PM   #15
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Interesting on your measurements of R9. I'd be curious to understand why. Have to do some more reading I guess.

Along the same lines as what you are doing I was/am playing with a t7/FET/kt88. I use a CCS under the FET in this amp and have only diddled with a resistor. So this thread is very interesting. I was asked to build a 300b SET and I want to use a derivative of the power drive in that as well, almost identical to what you are doing.
__________________
Living Life Doing the Waltz in 4/4 meter.

Last edited by SGregory; 27th April 2010 at 10:34 PM.
  Reply With Quote
Old 2nd May 2010, 12:40 AM   #16
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Attached schematic shows my test bench. I have performed comparisons between the 5842 and 6J5 in place of TU5. All measurements were done with an 8-ohm load at 1 W RMS output power.

The 5842 shows slightly higher THD than the 6J5 (0.1 percentage point higher at 1 kHz) - no surprises here. Comparing the source follower with resistive load vs CCS load reveals a marginal improvement in THD with CCS load. For resistive loading of the source follower, I replace I3 with a 22 kOhm resistor.

In terms of sound quality, the 5842 sounds harsher to my ears than the 6J5. I'm not sure if it's purely the difference in THD that makes for the difference - I kinda doubt that personally. But I don't know for sure. As far as the sound quality with resistive loading of the MOS source follower vs CCS load, the difference is fairly minimal. I think I currently favor the more open sound of the CCS load, but I'm comparing my experience now vs my listening experience late last night. So take this with a grain of salt. In terms of complexity and cost, the resistor wins hands down.

I should add that I'm now running off of a regulated HV supply. It's based on that ancient app note from National (LB-47 to be exact) and consists of an LM317 with an emitter follower in front to keep the voltage across the LM317 within spec. See page 361 in Morgan Jones, 3rd ed. or download the app note from National.com directly.

~Tom
Attached Images
File Type: png 300B_6J5_SourceFollowerCCS_Schematic.png (18.3 KB, 170 views)
File Type: png THD_SourceFollower_Comparison.png (33.6 KB, 165 views)

Last edited by tomchr; 2nd May 2010 at 12:43 AM.
  Reply With Quote
Old 2nd May 2010, 02:06 AM   #17
rknize is offline rknize  United States
diyAudio Member
 
rknize's Avatar
 
Join Date: Feb 2009
Location: Chicagoland
Send a message via AIM to rknize Send a message via Yahoo to rknize
Otherwise known as the "Miada" regulator (the fellow who wrote the app note).
  Reply With Quote
Old 2nd May 2010, 10:06 PM   #18
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
So I've done a little comparative listening now. Adding a CCS to the source follower rather than the resistor does cause an audible improvement. I wouldn't call it huge, earth shattering, but it's definitely worth taking IMO. The highs open up and seem to become more precise and the sound seems to decompress. I can't comment on any sound stage improvements as I'm listening in 2-channel mono (i.e. L+R).

In my final design I'll plan to have the CCS in there but probably leave room for the resistor so I can experiment once I have two channels running.

~Tom
  Reply With Quote
Old 3rd May 2010, 03:51 AM   #19
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Tom,
Your listening observations matchs mine with the ccs. I find that resistive loading is for lack of better word boring. With my set-up, the sound stage looses depth, in addition to the detail and clarity you reported.

Just guessing, but I would think the difference would show up in a FFT of a square wave.

I did some rudementary testing this weekend on the impedance of the gate bias resistor. In the case I mentioned above I found that the higher impedance actually improved my THD. In my set-up 500k was 0.07%, 750k was 0.05%, and 990k was 0.03% measured at 1kHz and 1Watt. Again it is a different output tube and amp.


Have you tried a higher bias current. I have mine set at 10mA per some suggestions. I havn't played with the value however.
__________________
Living Life Doing the Waltz in 4/4 meter.

Last edited by SGregory; 3rd May 2010 at 03:55 AM.
  Reply With Quote
Old 3rd May 2010, 10:07 PM   #20
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Quote:
Originally Posted by SGregory View Post
In the case I mentioned above I found that the higher impedance actually improved my THD. In my set-up 500k was 0.07%, 750k was 0.05%, and 990k was 0.03% measured at 1kHz and 1Watt. Again it is a different output tube and amp.
Where are you measuring 0.03 % THD? Is that the THD of just the source follower? /me confused. A schematic would help...

Typical THD numbers for my amp are about 0.6 % THD+N for 1 kHz, 1 W, 8 ohm.

Quote:
Originally Posted by SGregory View Post
Have you tried a higher bias current.
I have not.

~Tom
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Help with current limiting device Rob11966 Power Supplies 2 23rd November 2009 06:47 PM
LM338 Current Limiting imix500 Parts 48 22nd October 2009 11:21 AM
Current limiting behavior?? stoc005 Solid State 10 10th February 2009 02:53 PM
PowerDrive Circuit for 805 (+ve grid bias) snoopyma Tubes / Valves 7 17th August 2006 04:18 AM
Load Current limiting k1jroth Solid State 2 20th November 2004 06:20 AM


New To Site? Need Help?

All times are GMT. The time now is 11:39 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