807 Tube Triode Connected - 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 5th August 2008, 06:08 AM   #1
oshifis is online now oshifis  Hungary
diyAudio Member
 
Join Date: Mar 2004
Location: Budapest, Hungary
Default 807 Tube Triode Connected

I intend to use 807 beam tetrodes in a push-pull circuit triode connected with B+ 425V. The data sheet specifies Vg2 < 300V and Pg2 < 3W. I am afraid of reduced lifetime of the tubes when Vg2 is at the anode voltage 425V DC and peak AC voltage approaches 600V. I use 47 ohms resistors at g2. Do I need to worry about the Vg2 limitation? Could be reduced lifetime the consequence of elevating Vg2? If it is, how can I extend the lifetime of the tubes? Could it help increasing the g2 resistor, or would the triode characteristics be affected then? A series of Zener diodes could drop some 250V, but I am afraid it would destroy the good triode sound.

Thanks for any insight.
  Reply With Quote
Old 5th August 2008, 06:48 AM   #2
kmaier is offline kmaier  United States
diyAudio Member
 
kmaier's Avatar
 
Join Date: Jul 2007
Yes, I would worry about screen grid limitations and reduced life. I grew up them... my Dad's amp used them... +550V on the plates, +275 on the screens with fixed bias... great sounding amp but even then it could toast them over time.... lotsa blue then eventually red plates. We were able to buy NOS JAN surplus 807s back in the late 60's and 70's for about $1 each... so no big deal. Nowadays... not so cheap. You might want to consider some NOS 6L6GC in triode mode.... most of the older tube manuals show triode-connected plate curve charts for them.

Regards, KM
  Reply With Quote
Old 5th August 2008, 09:01 AM   #3
diyAudio Moderator Emeritus
 
ray_moth's Avatar
 
Join Date: Jan 2004
Location: Jakarta
Patrick Turner has some things to say on this topic at the rec.audio newsgroup.
  Reply With Quote
Old 5th August 2008, 09:35 AM   #4
oshifis is online now oshifis  Hungary
diyAudio Member
 
Join Date: Mar 2004
Location: Budapest, Hungary
Thanks for the info. I made up that Vg2 may be tied to the anode if B+ is less than 450V (triode or UL). I built and use the attached circuit. The problem is that the grid bias needs frequently re-adjusted and the tubes have to be frequently replaced. I replace them when setting 30-50 mA bias current is not possible any more within -30V ... -50V Vg1 range. Could it be because of rapid tube aging due to over-spec Vg2?
Attached Images
File Type: jpg 807_pp.jpg (77.4 KB, 1613 views)
  Reply With Quote
Old 5th August 2008, 10:22 AM   #5
kmaier is offline kmaier  United States
diyAudio Member
 
kmaier's Avatar
 
Join Date: Jul 2007
Yes, at 450 volts you are certainly stressing the screens and causing premature failure. There are some triode-connected specs located here:

http://tdsl.duncanamps.com/show.php?des=807

I suggest you more closely following the recommended ratings, and note that these are closer to maximum and should be toned down if you expect longer life. Good luck.

Regards, KM
  Reply With Quote
Old 6th August 2008, 01:15 AM   #6
diyAudio Member
 
Miles Prower's Avatar
 
Join Date: May 2005
Location: USA
Blog Entries: 6
If you're designing with 807s, you'll need the STC 807 Application Report. Includes information for triode operation. Operating at Vpp= 425V is a bit much for trioded operation.
__________________
There are no foxes in atheistholes
www.dolphin-hsl.com
  Reply With Quote
Old 28th November 2011, 09:22 AM   #7
diyAudio Member
 
Alain Poitras's Avatar
 
Join Date: Aug 2011
Location: Laurentides QC
Blog Entries: 1
Quote:
Originally Posted by oshifis View Post
Thanks for the info. I made up that Vg2 may be tied to the anode if B+ is less than 450V (triode or UL). I built and use the attached circuit. The problem is that the grid bias needs frequently re-adjusted and the tubes have to be frequently replaced. I replace them when setting 30-50 mA bias current is not possible any more within -30V ... -50V Vg1 range. Could it be because of rapid tube aging due to over-spec Vg2?
Hi oshifis,

You ask me about two weeks ago if I can simulate your circuit on my SIMetrix simulator and it is done. This is a very nice amplifier design and it is very difficult to improve it.

I am glad about the results because that's a very good way to verify my 6DJ8 and 6L6GC Spice models, like I tell you before, the 6L6GC and 807 have very close caracteristics and I get very close results to what you schematic said

It is a really good idea to use the 1N5312 and 1N5313 diodes current limiter for the phase splitter current mirror but I don't have their Spice model so I just use the incorporated SIMetrix current source, it does exactly the same thing .

Here's your original circuit and results :

Click the image to open in full size.

The first thing I notice is your 6DJ8 drivers are in great danger ...

Click the image to open in full size.

This is something you should take care of as soon as possible because I know the 6DJ8 are very expensives ... The phase splitter tubes operation point look very good to me so I draw a corrected version with a close operation point for the driver too and a safe Va under 130V ... Not 269,5V ...

Click the image to open in full size.

Like you can see, you just have to change the 33K cathode resistors for 18K (R12 and R13) and add two 4,7K 2W (or more) resistor (R23 and R24) and a 47uF filter capacitor (C3) to get a 268,8V supply for the driver tubes. There is just a little bit more distortion this way but nobody will notice any differences for sure ...

This is a "pure class A" amplifier and just like a "class AB1", there is no current pull by the power tubes grids, so those four drivers tubes are not really needed :

Click the image to open in full size.

Of course, the distortion is a little bit higher but it is so simple ... However, there is a simple way to improve it, since you already use a current mirror for the splitter cathodes, why not use adjustable ones for their plates, using a good "TO220 high voltage DN2540 power depletion MOSFET" as current limiter ...

It improve the distortion and the gain ... It is still very simple and cheap, a DN2540 cost about 2,50$ and last for life but each time you replace your four 6DJ8 drivers, it cost you over 100$, think about it ... A phase splitter like that just need a 200V supply or a little more if necessary so you can get a better filtered supply for this stage. I left a 48V "headroom" for a signal up to 34V RMS plus 15,8V for the DN2540 Vds, it is plenty enough, just provide a very small heatsink for each of them will help if they get too hot.

Click the image to open in full size.

About using thoses MOSFET with the "corrected" circuit version :

Click the image to open in full size.

Now, you get a "Deluxe" version with safe operation point for the drivers and a distortion as low as the "original version" ...

About your 807 "longevity problem", there is not much to do for that because it is a fixed bias amplifier and this is hard for any tubes, the best is using a cathode bias circuit to protect the tubes ...

It is possible to make a circuit half way between fixed bias and cathode bias just by inserting small value resistors in the cathode circuit and use a corresponding lower fixed bias. Maybe it will help the 807 to live longer and produce less distortion but there will probably be a small lost of power ... I will make a simulation test about that this week.

Alain.

Last edited by Alain Poitras; 28th November 2011 at 09:33 AM.
  Reply With Quote
Old 29th November 2011, 06:45 AM   #8
diyAudio Member
 
Alain Poitras's Avatar
 
Join Date: Aug 2011
Location: Laurentides QC
Blog Entries: 1
Hi oshifis,

I made some modifications to the version 6 and it did improve performances a lot ...

But first, I have to tell you something :

Since your drivers plate to cathode voltage is more than twice the 6DJ8 design center maximum, arcing between the plate and cathode can occur very often and since it is a cathode follower circuit, those arcs will produce some 415V positive peaks on the 807 grids and that is very bad for them ... Maybe, this is why they don't last very long ...

Yesterday, I tell you it is possible to make a circuit "partially fixed bias and partially cathode bias" and it work very well with just a little 5V cathode bias without too much power loss :

Click the image to open in full size.

I notice in my tests the lower frequency response was not very good, it was not because the transformer but just because your 220nF coupling capacitors ... This is much too low for the following grid leak 100k resistors ... So I change them for some 2,2uF capacitors and it make a very big difference, but this is a minimum, you can use some larger ones, up to 4,7uf is OK, over that is useless.

I also add a "ficticious" serial 60K resistance at the input to represent the "ideal" source output impedance for this circuit, it is important for a good the high frequency response. Notice the 85 ohms resistors at the transformer (R2 and R3) are also "ficticious", they represent the DCR of the transformer primary ... Those resistors are not really in the circuit, it is just for the Spice simulation ...

Like you can read on the schematic, the maximum power is now slightly lower at 11,336W but the distortion at 1Khz drop to 0,740275% ... This is really good ... The frequency response is now vitually "flat" from 2,5Hz to 20Khz, within 0,135dB ... This is fantastic ... But to get this bass response, you need big 2200uF bypass capacitors across the 100 ohms cathode bias resistors ...

In a push-pull circuit, the maximum power is limited by the maximum voltage swing on the grids before the 0V clipping occur. The positive swing on the grid can reach this clipping but the operating class is define by the negative swing, when it don't reach the tube "cutoff" where the current is 0ma, the amplifier operate in class "A". If it cross this limit, it operate in class "AB1", the power is then greater but the distortion is usually higher ...

You told me in your message you dont like to use any bypass capacitors but like I told you, if you don't use them, distortion will reach about 3,5% at 1Khz and the power will be much lower. But it is possible to use a LED bias instead and higher bias and current for the 807 tubes, up to 60ma for a 25W plate dissipation.

This will be my next test, maybe for tomorrow is I am not to much busy.

Alain.
  Reply With Quote
Old 29th November 2011, 07:51 AM   #9
diyAudio Member
 
Alain Poitras's Avatar
 
Join Date: Aug 2011
Location: Laurentides QC
Blog Entries: 1
oshifis,

I just found out I was wrong about something ...

Your amplifier don't operate in class "A" but in class "AB1"

Click the image to open in full size.

In fact, it operate in class "A" just below 2,73W ...

My mistake was I monitor the voltage at the plate of a 807 and I see a nice sinuso´dal because the push-pull transformer, it is the current I have to monitor !

Alain.
  Reply With Quote
Old 29th November 2011, 08:40 AM   #10
diyAudio Member
 
Alain Poitras's Avatar
 
Join Date: Aug 2011
Location: Laurentides QC
Blog Entries: 1
oshifis,

I just found out I was wrong about something ...

Your amplifier don't operate in class "A" but in class "AB1"

Click the image to open in full size.

In fact, it operate in class "A" just below 2,73W ...

My mistake was I monitor the voltage at the plate of a 807 and I see a nice sinuso´dal because the push-pull transformer, it is the current I have to monitor !

Alain.
  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
Driver for a 807 em triode mode...help !!! aldovan Tubes / Valves 7 4th October 2011 06:15 PM
Triode connected 6LW6 for SE wrenchone Tubes / Valves 2 5th July 2011 05:10 PM
Curves for Triode Connected 6EW6 wrenchone Tubes / Valves 0 15th March 2010 12:52 AM
Triode Connected 6BK5 wrenchone Tubes / Valves 2 5th February 2009 03:59 PM
8417 triode connected tube data rhcclark Tubes / Valves 0 5th May 2007 03:05 PM


New To Site? Need Help?

All times are GMT. The time now is 12:48 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ę1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2