Another SE amp design. - 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 14th June 2011, 12:28 AM   #1
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Default Another SE amp design.

While on vacation last week I was reading several articles on Mu stages/followers and came to the conclusion that a Mu stage with a Pentode on top might make a good driver for a sweep tube as a SE output.

I had recently read Karpov's article "SPECTRA" on the output spectrum of several Soviet Union tubes and decided to try the 6P41S. It looks like it should be nice in Triode mode, and might be worth listening to in Pentode mode with a little plate to previous grid feedback.

I decided to try direct coupling the output of the Mu stage, which requires level shifting, or a minus supply for the lower triode. Using a little sand for feedback to control the minus supply and thus the drive to the output tube grid looks promising in simulations.

An alternative method would be to put a series resistor in the B- supply and use current sink control to adjust the B_ to the lower tube.

The circuit needs some tweaking as the Mu stage currently is only running 1mA of current and more will be required if A2 is to be achieved.

Comments and recommendations welcome.
Attached Images
File Type: jpg ThunderBolt_1.JPG (82.3 KB, 1604 views)

Last edited by TheGimp; 14th June 2011 at 12:39 AM.
  Reply With Quote
Old 14th June 2011, 09:41 AM   #2
diyAudio Member
 
the_manta's Avatar
 
Join Date: Nov 2006
Location: Munich, Bavaria
Hi Steven,
Yes, Pentodes make good CCS and thus perform well as upper tube in µ-followers.
Indeed a nice circuit but there are some issues to discuss.
1mA is a bit low. Yes it's a pentode power stage and miller capacity shouldn't be a problem, but I've listened to many amps and I alway prefered a stronger driver. By the way, the 6Н1П likes higher currents and you can clearly hear that.
Of course, low currents are nice in µ-stages cause that means high upper-tube-cathode-resistors @ fewer voltage drop and that leads to better constant current sources. But the more this stage has to drive, the more its concept fails.
So I suggest you not to use the EF86 as upper stage. You could use it triode connected as lower stage. But for the upper stage I would take a typical wide-band Pentode with 5mA to 10mA current and higher gm.
Next thing is your feedback. You're using parallel derived and parallel applied feedback, effectively from the Anode of the power tube back to its grid. But the problem is, you are applying it on the Anode of the driver tube, that tube of which the current should be constant ! See the Problem ? That resistor (R7) destroys your CCS (a little bit, since it's rather large, but remember that the power tube can swing to twice the Ub)
Also bear in mind that this kind of feedback reduces not only the output resistance but also input resistance.
Another Feedback option is the cathode of the first tube. So you have the good old parallel derived, series applied feedback. (reducing output and increasing input resistance)
Finally I hope you know that you're dealing with sweep tubes. I tested them often times and in SE single ended the didn't sound impressive.....
regards
  Reply With Quote
Old 14th June 2011, 11:39 AM   #3
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
The mu-follower gives low output impedance but not particularly good drive capability - especially with an EF86 running at a mA! You have destroyed the whole point of the circuit (high impedance load for the lower anode, so low distortion) by sending feedback to the lower anode. If you mix bits from different good circuits together you need to be aware that the whole can be significantly less than the sum of its parts.

You would need a complete redesign to drive the output into A2.
  Reply With Quote
Old 14th June 2011, 01:08 PM   #4
SVStube is offline SVStube  Ukraine
diyAudio Member
 
Join Date: Jan 2011
Location: Vinnytsja
Cathode potential output tube you need to raise higher than the cathode potential EF86. Otherwise, should be put interstage capacitor, and do the bias for the output tube.
  Reply With Quote
Old 14th June 2011, 07:22 PM   #5
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
The ss circuit takes care of keeping output tube grid1 bias negative. I don't know if this is called a servo or not. It simply adjusts the lower triode cathode supply to move the plate of the triode and thus the drive to the next grid.

Since the upper tube is configured as a current source this can be done with a simple regulator using the grid voltage as feedback.

This may require a delay in the B+/B- supplies to insure proper bias of the output tube during power up.

Using a Mu stage to drive a dc coupled SE output is what I was interested in experimenting with.

A2 operation was more of an afterthought once I got the basic circuit designed, it is not required but I am interesting to play with. What I have not been able to find out is how much current is needed for positive g1 drive.
  Reply With Quote
Old 28th August 2011, 09:15 PM   #6
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
The breadboard worked fine, so I made a PCB for further testing. First PCB in 30 years and it worked on the first pass, although there are pinholes in the traces due to the poor mask quality I got printing on overhead viewer film with a HP Deskjet printer.

I used some 10 year old GC positive sensitized board with red devil lye for developer and Easy Off oven spray to remove the remaining mask after etching.

DipTrace was the easiest layout program I tried so I'm going with it.

Quick and easy.
Attached Images
File Type: jpg PWB_1.jpg (196.4 KB, 1278 views)
  Reply With Quote
Old 10th November 2011, 01:35 PM   #7
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
I finally got around to bread-boarding the entire amp.

I set the bias with the 6P41S out of the socket. It was necessary to change a couple of resistors to get the bias where I wanted it. Once bias was set to -30V I powered it down and installed the 6P41s, and powered it up. Vg 6P41S surges to 106 for several seconds, but no current spike was seen. Vg 6P41s rapidly drops to proper bias around -30V, so I suspect the thermal delay of the driver cathodes is much lower than the thermal delay of the 6P41S allowing bias to be achieved before conduction takes place in the 6P41S. A delay circuit of some sort may be a good idea and will be investigated later.

The 6P41s went Red-Plate after warming up. With B+ at 270V in UL, Vg2 was way out of spec. I changed the screen resistor from 470R to 15K with a 1uF bypass cap and idle current dropped from 250mA to 125mA. Adjusting Vg1 brought it down to 95mA.

Looks like I need to go back to the thread about reducing the screen voltage for UL operation with various techniques to optimize the design as the 15K resistor bypassed with 1uF rolls off my LF response.

As a note, the 6P41S takes a lot of drive voltage and a lot of bias (-30V) compared to a 6L6 at -20V!

After getting the amp running I played around with UL and Triode modes, but didn’t try Pentode mode yet.

The granddaughter and I listened to Lee Roy Parnell “Back to the Well” and Bonnie Raitt “Green Light”, and decided we like UL better although Triode sounded nice and may be better suited for some music. It sounded to me like it was going burst oscillation on peaks so I need to investigate that.

The 6P41S will take almost 125mA at 240V before going red plate. That is 30W and way beyond the 12W data sheet rating!!!. I may have to try a PP amp with them once I finish this SE amp. I suspect a 50W AB1 PP amp is possible with a pair of these tubes, although forced air cooling may be required. AB2 looks interesting as well. The tubes are HOT at 30W!

I bumped something with my elbow and the amp shut down, so I now have some debugging to do.
  Reply With Quote
Old 11th November 2011, 03:27 AM   #8
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Back up and running.

Amp had a 250KHz oscillation causing peaks to sound raspy.

FFTs and frequency sweeps are looking good.

SE amps require well regulated supplies. Including negative bias supplies!

DUH!
  Reply With Quote
Old 11th November 2011, 05:17 PM   #9
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
When I ran some frequency response plots last night I got a strange result when I tried a .068uF KY42y-2, per the attached plot.

The plot jumps up and down over 20dB with jaged edges.

I tried a second KY42y-2 with the same results. When I put the KY40y-9 back in it plotted with a smooth curve.

I cant' upload the plot right now for some reason. My browser crashes wehn I try. I'll try again later from home.

Last edited by TheGimp; 11th November 2011 at 05:22 PM.
  Reply With Quote
Old 11th November 2011, 05:57 PM   #10
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Here is the plot, it is obvious which trace is different.
Attached Images
File Type: jpg StrangeCap.JPG (70.2 KB, 840 views)
  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
SE 300B Integrated Amp: Help choose circuit design xenithon Tubes / Valves 18 15th April 2011 05:33 AM
Design Help (load line) on a SE Mono Block Amp build myDogFred Tubes / Valves 3 17th January 2009 08:44 AM
Would anyone help me to design a SE linestage/headphone amp? Thatch_Ear Headphone Systems 2 14th January 2007 02:42 PM
SIMPLE SE amp design wanted DrewP Solid State 3 25th August 2003 04:59 AM
Humming problem with 5687? SE amp design. G Tubes / Valves 7 21st February 2003 12:59 AM


New To Site? Need Help?

All times are GMT. The time now is 01:47 AM.


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