A rebuilding Push Pull project.

the thermionic watercooler

A rebuilding Push Pull project.

Postby ZepZep » Sun Aug 26, 2007 10:18 pm

Hi all,

It’s my first post in this forum. I have a very bad sounding China PP amplifier in my place. and I’d to upgrade it. China amplifier has global negative feedback, a 12ax7(phase splitter), a 12au7(driver), 2 x EL34 (fixed bias) and run in an ultra linear mod. I’d to upgrade it. I’m not electronics pro so if you could help me I appreciate a lot.

My project is:
In original circuit cut as little as possible PCB wires;
Don’t swapped tubes sockets;
Available tubes: 12ax7, 12au7, 12at7, ecc88, e182c, EL34
My system needs less than 20 w (so class AB1 it´s ok)
Triode mode
NO negative global feedback (if it´s possible)



Comments are very welcome.


PS: Sorry my bad English.
Last edited by ZepZep on Wed Nov 07, 2007 9:23 am, edited 1 time in total.
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby Shannon Parks » Mon Aug 27, 2007 6:07 am

Hi ZepZep,

How much different is the new schematic from the original design? Could you post that too? If it is very different, then we might just want to start out making a couple small changes.
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby EWBrown » Mon Aug 27, 2007 7:15 am

Is the "China Amp" one of the "Ming Da" units that I've seen offered up on E-bay? From what I've seen they are long on looks , short on performance.

I had a chance to play with one that a friend had, it was definitely lacking.... I swapped out the China "Sino" tubes for him, installed JJ ECC83s and ECC82s and Svetlana EL34s and that made a noticeable improvement. Another suggested upgrade is to replace the four EL34 grid coupling caps with Auricaps, either 0,22 or 0.33 uF. I didn't bother with that, he didn't want to lay out the $$$ at the time.

The thing that bugged me most about that amp was that there was a persistent low level hum in the left channel. Yellow_Light_Colorz_PDT_21 :o

/ed B in NH (back from 3 weeks vaca in NC)
Last edited by EWBrown on Tue Aug 28, 2007 5:58 am, edited 1 time in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby ZepZep » Mon Aug 27, 2007 7:57 am

separks wrote:Hi ZepZep,

How much different is the new schematic from the original design? Could you post that too? If it is very different, then we might just want to start out making a couple small changes.


A very good amp …….. to kick. Yellow_Light_Colorz_PDT_21
Last edited by ZepZep on Wed Nov 07, 2007 9:24 am, edited 2 times in total.
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby parabellum » Mon Aug 27, 2007 2:40 pm

Those china amps sure do look pretty sometimes and I always wondered if they make them real crappy and I guess I got my answer.

Maybe it would be for the best to cut your losses and start from a decent platform?
User avatar
parabellum
 
Posts: 195
Joined: Wed Apr 25, 2007 1:37 pm
Location: Bay Area, CA

Postby ZepZep » Mon Aug 27, 2007 3:08 pm

parabellum wrote:Those china amps sure do look pretty sometimes and I always wondered if they make them real crappy and I guess I got my answer.

Maybe it would be for the best to cut your losses and start from a decent platform?

I agreed with but I have a great diy 300B SE amplifier. So I though why not upgrade and learn with my China PP amplifier. I have my PP in a box since 2001.
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby Gingertube » Mon Aug 27, 2007 9:40 pm

ZepZep,
Its worth persisting with that Chinese Amp.

Your proposed mods are all heading in the right direction.
The Triode Mode Output will make a big difference.
Direct Coupling the DiffAmp/Driver from the front end is a good idea BUT use a current source in place of that final resistor to 0V.
I notice that you are changing the front end (to a diffamp) as well with a current source in the tail - all that should help.
The change from 12AX7 to ECC88 for the front end should be fine - the gain you loose from the lower mu tube will be made up by ditching the global feedback.

Here is a link to what I did to a Chinese Music Angel EL34 PP which is a slightly different circuit. In this mod I used some trick balanced shunt feedback from the output tube anodes.
http://www.diyaudio.com/forums/showthre ... genumber=3

Cheers,
Ian
Gingertube
 
Posts: 245
Joined: Thu Oct 05, 2006 10:27 pm
Location: Adelaide, South Oz

Postby ZepZep » Mon Aug 27, 2007 11:24 pm

Gingertube wrote:Your proposed mods are all heading in the right direction.
The Triode Mode Output will make a big difference.
Direct Coupling the DiffAmp/Driver from the front end is a good idea BUT use a current source in place of that final resistor to 0V.
I notice that you are changing the front end (to a diffamp) as well with a current source in the tail - all that should help.
The change from 12AX7 to ECC88 for the front end should be fine - the gain you loose from the lower mu tube will be made up by ditching the global feedback.

Here is a link to what I did to a Chinese Music Angel EL34 PP which is a slightly different circuit. In this mod I used some trick balanced shunt feedback from the output tube anodes.
http://www.diyaudio.com/forums/showthre ... genumber=3

Cheers,
Ian

I don’t if I have understood you. Is something like this?


CCS tuned at 5mA and second at 16mA.
Last edited by ZepZep on Wed Nov 07, 2007 9:24 am, edited 1 time in total.
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby ZepZep » Mon Aug 27, 2007 11:31 pm

ZepZep wrote:
Gingertube wrote:Your proposed mods are all heading in the right direction.
The Triode Mode Output will make a big difference.
Direct Coupling the DiffAmp/Driver from the front end is a good idea BUT use a current source in place of that final resistor to 0V.
I notice that you are changing the front end (to a diffamp) as well with a current source in the tail - all that should help.
The change from 12AX7 to ECC88 for the front end should be fine - the gain you loose from the lower mu tube will be made up by ditching the global feedback.

Here is a link to what I did to a Chinese Music Angel EL34 PP which is a slightly different circuit. In this mod I used some trick balanced shunt feedback from the output tube anodes.
http://www.diyaudio.com/forums/showthre ... genumber=3

Cheers,
Ian

I don’t if I have understood you. Is something like this?

CCS tuned at 5mA and second at 16mA.




How many power output (W) could it achieve?
Last edited by ZepZep on Wed Nov 07, 2007 9:25 am, edited 2 times in total.
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby Gingertube » Tue Aug 28, 2007 5:54 pm

Yes,
Use that a starting point and adjust on test once its running. With 420V rail 16mA will be a bit to much on the driver diff amp as you will get 0.008 x 33K = 264 volts drop across the anode resistors. You will need to drop the current or the anode load resistors or both.

The 12AU7 grid voltage will be set by the anode voltage of the front end, (probably about 85 to 90).
That means 12AU7 cathode voltage of about 95 to 100 V.
That leaves 320 volts across the tube and the anode load resistor. About 1/2 and 1/2 with possibly a bit more across the tube than the loads is right.
Say 180 across the tube and 140 across the load resistor - so that means anode voltage of 280 volts.
You would need 18K anode loads if running 8mA per side. with 33K anode loads the current would need to dropped to 4mA per side which is perfectly adequate. If you don't like the 12AU7 at 4mA there are other tubes you could try (ECC99 for example).

Cheers,
Ian
Gingertube
 
Posts: 245
Joined: Thu Oct 05, 2006 10:27 pm
Location: Adelaide, South Oz

Postby ZepZep » Tue Sep 04, 2007 3:36 am

Gingertube wrote:You would need 18K anode loads if running 8mA per side. ...
Cheers,
Ian


Hi,
Gingertube & al. thanks for you help.

I simulated what you advise me but seems not work at Ltspice. Is it a spice 12au7 model problem or it could happen in real world?
Last edited by ZepZep on Wed Nov 07, 2007 9:26 am, edited 2 times in total.
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby ZepZep » Thu Sep 27, 2007 10:04 am

Yellow_Light_Colorz_PDT_13
Anybody?
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby Shannon Parks » Sat Sep 29, 2007 6:04 am

ZepZep,

Some of the ImageShack jpegs aren't coming up. Could you post the original schematic again and then another of what mods you did (or propose to do)? I got lost re-reading the thread. Yellow_Light_Colorz_PDT_09
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby ZepZep » Sat Sep 29, 2007 3:28 pm

Hi,

The original schematic is in post 4, some Gingertube phase splitter mods (post 10) seem don´t work at Ltspice. Is it true or is it only at Ltspice?
ZepZep
 
Posts: 10
Joined: Sun Aug 26, 2007 5:50 am

Postby Shannon Parks » Sun Sep 30, 2007 8:22 am

Still a little confused, but are you planning to have DC on the phase splitter grid from the voltage gain tube's plate, and then a CCS in the phase splitter tail? This won't work for a couple reasons - 1) your low B+ isn't high enough for a DC coupled phase splitter, and 2) your CCS (per the schematic) probably doesn't support the voltage drop across it.

You can remove R21 and replace it with a LM234Z and current set resistor (say 10 to 15 ohm range) from a stock design. You will need that AC coupling.
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Next

Return to diy hifi

Who is online

Users browsing this forum: Google [Bot] and 14 guests