New 6V6 Build - Suggestions Needed - Page 6 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 21st October 2012, 07:14 PM   #51
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
Quote:
Originally Posted by gokite View Post
Do you have two spare 100k Rs? If so, gator clip them each parallel to R14 and R9 and do your test again. This should increase both your Plate voltage swing limit as well as the available total swing.
I'll bet that would increase the possible swing through the cathodyne. But I have a question for you, out of curiosity. What would be the effect of a 50k plate and cathode load on that 6SL7? I mean, 50k ohms is appreciably less than the rp of a 6SL7. Gain would be reduced (so less negative feedback), gm would be less, rp would be higher (worsening the problem). I'm afraid it would be too nonlinear.

Then again, maybe I'm about to learn something?
  Reply With Quote
Old 21st October 2012, 09:53 PM   #52
SandyG is offline SandyG  United States
diyAudio Member
 
SandyG's Avatar
 
Join Date: Jul 2011
Location: Los Angeles
Quote:
Originally Posted by rongon View Post
Just out of curiosity, what is the voltage at pin 8 of the 6V6's?
The Bias on the 6v6's at pin 8 are 21 Volts for 5U4 and 22 Volts for the GZ34. I have those listed under 'Bias' in the measurments. The current Bias calculated to be a touch on the high side. The cathode resistor on is a 270 ohm and I recall it calculated out to be 37ma at 21Volts.

GoKite - I do have a couple, but just as easy to swap them out since it's all on the workbench. I was going to try a pair of 68K for R14 and R9 would that be a good test?

Sandy
  Reply With Quote
Old 21st October 2012, 11:16 PM   #53
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
Quote:
The Bias on the 6v6's at pin 8 are 21 Volts for 5U4 and 22 Volts for the GZ34. I have those listed under 'Bias' in the measurments. The current Bias calculated to be a touch on the high side. The cathode resistor on is a 270 ohm and I recall it calculated out to be 37ma at 21Volts.
The grid bias on the 6V6 defines the limit of the voltage it can accept at its grid before overload. In other words, you need the driver stages to give you 21V peak (about 15Vrms) AC audio signal at the 6V6 output tube grids. That means your input tube needs to deliver that from the voltage it gets at its grid.

So if you use a 6SL7 with -0.5V grid bias, and supposing you get a good 50x gain from that 6SL7, that's 50*0.4 = 20V peak at the 6V6 grids. Not quite enough.

Remember that today's sources are supposed to deliver up to 2Vrms (2.83V peak) audio signal. A CD player is supposed to output 2Vrms at 0dBFS (full scale). You could design your amplifier to accept up to 2V at its input. Now remember that whatever negative feedback you apply will make the whole amp less sensitive -- adding negative feedback will make it so that it will take a higher voltage from the source to drive the amp to full power. 6dB negative feedback means a 2x drop in voltage sensitivity. In other words, if it took 2V to drive your amp to full power before negative feedback was added (open loop), it will take double that -- 4V -- to drive the amp after 6dB negative feedback has been added.

If you use a 6SN7 with -2.0V bias, and you figure your 6SN7 gets 14x gain, that's (being conservative here) 1.7*14 = 23.8V peak (about 17Vrms). That gets you there with just a little room to spare.

The resulting amp run without negative feedback would clip with 1.5V peak (about 1Vrms) at its input. If you apply 6dB NFB to the amp, its sensitivity will be halved, so it will now take 2Vrms to drive it to clipping. Conveniently, that is the full possible output from a CD player.

It looks to me that the simplest fix would be to abandon the 6SL7 and go with a 6SN7, and apply no more than 6dB NFB to the amp from the OPT secondary to the input tube cathode.

OK, I'll stop nagging now.

--

Last edited by rongon; 21st October 2012 at 11:19 PM.
  Reply With Quote
Old 22nd October 2012, 01:12 AM   #54
diyAudio Member
 
Join Date: Oct 2011
Quote:
Originally Posted by rongon View Post
I'll bet that would increase the possible swing through the cathodyne. But I have a question for you, out of curiosity. What would be the effect of a 50k plate and cathode load on that 6SL7? I mean, 50k ohms is appreciably less than the rp of a 6SL7. Gain would be reduced (so less negative feedback), gm would be less, rp would be higher (worsening the problem). I'm afraid it would be too nonlinear.

Then again, maybe I'm about to learn something?
My suggestion was simply to provide a quick test for the OP to verify the issue as the PI. I believe i posted earlier i've never designed a cathodyne PI. In fact, i come from the guitar amp world and have an interest in building a home system, so i have been reading and watching diyaudio to pick up on the stuff you guys are doing, and why, so i can design my own hifi system.

Not looking at a set of curves, yes gain should be less, but the swing and headroom should increase, sending a higher-amplitude clean signal to the powertube(s) which would verify his/her issue in minutes. Again, just a quick test that could be done without even heating up the iron.
  Reply With Quote
Old 22nd October 2012, 06:42 PM   #55
SandyG is offline SandyG  United States
diyAudio Member
 
SandyG's Avatar
 
Join Date: Jul 2011
Location: Los Angeles
I'll give a few things a try, definitely will try out changing the PI's resistors just to see what happens. Easy and might have some different result.

I can try the 6SN7 route as well, I have plenty of parts for that, and it's not that complex of a change.

Hopefully I can catch a break in the next day or so to try a few things.

Can't thank everyone enough for all the tip, tech and suggestions.

Sandy
  Reply With Quote
Old 23rd October 2012, 08:07 AM   #56
SandyG is offline SandyG  United States
diyAudio Member
 
SandyG's Avatar
 
Join Date: Jul 2011
Location: Los Angeles
Had a sliver of time. Some oddly interesting findings!

I ran the PI's resistors R9 and R14 to about 75k down from the 103k's that were installed. Clipping resulted at about the same 8.3VRMS. This is with the 5U4. The initial test with 103k seems slightly better. Didn't have time to capture all the data points.

Then got the iron out and pulled out R7 and changed it to a 1.2k
R9 and R14 to 33k
R8 to 47K
And put in a fresh nos GE 6SN7

Code:
Voltages -
HV1  : 343
HV2  : 335
HV3  : 317
Bias  : 22.9 (measured bias current with adapter and right at 37ma with 300ohm)
V1PA  : 134
V1PB  : 200
R8    : 182
R14   : 134
R9    : 135
R7    : 4.1
Clipping again starts at about 8.7VRMS with the 6SN7!

From the other days test with the GZ34 and the 6SL7 ran right about the same. I'm watching the signal on the scope in the input and output and driving with a good signal source. Also looks like the 6SN7 requires about double the input for the same output. (Didn't mess at all with the FB loop).

The questions that I have could the 8K output transformer be a problem here, I had seen 6V6 output transformers run from 5K to the 8K range. The GE 6V6GT data sheet shows 8K Plate to Plate for PP AB1 amp with 285 Plate voltage. Might be a stretch, but figured I would toss it out since these are running HV1 (B+) right close to 350vdc. Hope that rating was what is to be used for the output transformer (Using Edcor GXPP15-8-8K)

That's my quick testing. Would have loved to see better or worse results but coming up similar, well leaves more room for exploring

Sandy

Last edited by SandyG; 23rd October 2012 at 08:24 AM.
  Reply With Quote
Old 23rd October 2012, 08:13 PM   #57
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
Quote:
Also looks like the 6SN7 requires about double the input for the same output. (Didn't mess at all with the FB loop).
Yes, to be expected. Gain of 6SN7 is max of about 20X. Gain of 6SL7 is max of about 70X.


Quote:
got the iron out and pulled out R7 and changed it to a 1.2k
R9 and R14 to 33k
R8 to 47K
And put in a fresh nos GE 6SN7

Code:
Voltages -
HV1 : 343
HV2 : 335
HV3 : 317
Bias : 22.9 (measured bias current with adapter and right at 37ma with 300ohm)
V1PA : 134
V1PB : 200
R8 : 182
R14 : 134
R9 : 135
R7 : 4.1
Clipping again starts at about 8.7VRMS with the 6SN7!

OK, with the 6SL7 in there, you get 8.3Vrms out from the amp. That equals 8.6 watts rms output.

With the 6SN7 in there, you get 8.7Vrms out from the amp. That equals 9.46 watts rms output. You increased the power out from the amp by 10%. I'd say that's progress.

Also, why did you change R8 to 47k? I think 39k would work just as well, and would allow you to bias the cathodyne in a better spot. (You can put 270k in parallel with the 47k resistor to get 40k.)

Also, You put 33k for the cathodyne plate and cathode resistors (R9 and R14), instead of 22k.

I don't understand your voltage chart. You say that for R8 you got 182V. Was that measured at the 6SN7 plate, or "above" the 47k resistor (in other words, you measured Ebb)?

Since the 6SN7 input amp and cathodyne are DC coupled, the difference between the cathodyne cathode and grid MUST be -2V or thereabouts. It CANNOT be below -1V or so. (Remember that the first stage plate is directly connected to the second stage grid.) Therefore, a couple of vitally crucial measurements are:

- the voltage at the voltage amp's plate and
- the voltage at the cathodyne cathode.

You listed the B+ to the cathodyne as +335V. The voltages for a healthy cathodyne should be:

B+ = +335V
plate = +224V
cathode = +112V

That means the voltage at the voltage amp's plate MUST be about +110V.

The voltage at the cathodyne's cathode should be at least about +2V above the voltage amp's plate voltage.

Does your version measure in that ballpark?
--

The attached image shows what voltages go where. The next step would be to decide on how much current should go through each triode. I settled on about 5mA, because that allows me to put the same current through both the 1st and 2nd stages. That's how I arrived at R8 = 39k and R9, R14 = 22k. You can change that, but ONLY if the cathodyne stays biased correctly.
Attached Images
File Type: jpg cathodyne_driver_voltages.jpg (111.2 KB, 135 views)

Last edited by rongon; 23rd October 2012 at 08:43 PM. Reason: Oh, I had goofed it all up!!
  Reply With Quote
Old 23rd October 2012, 11:06 PM   #58
diyAudio Member
 
Join Date: Oct 2011
I don't understand your voltage chart. You say that for R8 you got 182V. Was that measured at the 6SN7 plate, or "above" the 47k resistor (in other words, you measured Ebb)?

his HV is 317 so that has to be across the resistor (DMM leads on each side). his voltages confused me as well at first.
317 - 182 = 129 at input plate? 3.8mA
335 - 135 = 200V at PI plate? 4.1mA

Are these voltages DMMd with the 6SL7 or 6SN7?

6SN7 0-6.7mA with bias at 4mA is about -4V which R7 confirms. Gain is about (90-155)/4=16.25 (tungSol GT datasheet graph)

For the PI, graphically it is biased at 4mA which again puts the grid close to -4V which in this case is about 1V off center...looking at the DMM readings the input plate is DC coupled to the grid at 129V and the voltage on the cathode is 134V (R14) so it is -5V which checks out.

However, if he sends in a 2Vpp signal, the gain from input is 16 so that results in a 32Vpp input to the cathodyne which will run out of headroom quickly. The best thing is to increase the voltage to the cathodyne B+ which requires a new PT. trying to squeeze the most out of it by lowering the gain of the input stage? Seems to me the cathodyne can handle about 12Vpp before unwanted distortion? Or a voltage divider between stages? or feedback between the plate and grid of the input stage? or, or, or...is there a cap for C6 in the build?
  Reply With Quote
Old 23rd October 2012, 11:40 PM   #59
SandyG is offline SandyG  United States
diyAudio Member
 
SandyG's Avatar
 
Join Date: Jul 2011
Location: Los Angeles
Lots to reply to!

The Voltages listed are with the GZ34 and 6SN7. The result of 8.7VRMS is lower then the original GZ34 and 6SL7 which clipped right around 9VRMS. So lost some ground vs. the best results with the GZ34 in the original configuration.

Voltages for R7,R8,R9, R14 are measured across the resistor not to ground.

Rongon - On the values for the resistors (47k/33k), I pulled the ones from a different post by mistake, was a late night. I'm going to try the values you suggested and see what it does and gather new voltages.

Gokite - C6 is not installed. Left open as an optional component.

Part of the problem with the amp could be the name, it's called the EZ10, but I should have started with the NOTSOEZ-8.6

Going to try to hit the correct components tonight if I'm able to operate a dangerous soldering iron late at night!

Sandy
  Reply With Quote
Old 24th October 2012, 12:33 AM   #60
diyAudio Member
 
Join Date: Oct 2011
Not trying to derail as I believe this is pertinent to understanding your circuit. What is a typical hifi input voltage p-p? I read somewhere on this forum 2V pp or 2Vrms was typical of a CD player. Is this true? Sorry, new to the hifi world...
  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
Build guitar amp with 2 6L6 or 4 6V6? TubeTweaker Instruments and Amps 11 15th August 2012 01:01 PM
Advice needed on new 6L6 ro 6V6 SE or SET build GloBug Tubes / Valves 39 23rd November 2011 05:22 PM
A little help with 6v6 screen needed. lazzer408 Tubes / Valves 15 28th February 2011 08:32 PM
At Last....First Build Complete...6V6/6SL7 cbj591 Tubes / Valves 9 31st July 2010 02:51 AM
6v6 music machine build help! tubo Tubes / Valves 26 5th November 2009 05:34 PM


New To Site? Need Help?

All times are GMT. The time now is 12:29 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