Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

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

6SN7 push pull flea amplifier project
6SN7 push pull flea amplifier project
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
Thread Tools Search this Thread
Old 20th October 2019, 01:44 PM   #391
pcan is offline pcan  Italy
diyAudio Member
Join Date: Dec 2015
6SN7 push pull flea amplifier project
The power supply I will use is the cheap SMPS board from China already mentioned on other forum threads. It is available from several sources: I bought it from eBay - auction ID 173034019336 . This board will allow to change the power supply and tube filament voltage very easily. I will mount the amplifier on a temporary chassis next to this SMPS board, and I will eventually build a traditional power supply after completing tests and tweaks. I bought the SMPS board a while ago out of curiosity but I never tested it. This project is a good opportunity to test my brand new scope. I apologize for the bad screen pictures: I still need to complete the scope setup and then I will get the screenshots directly on a USB stick or network share, but I rather spent my time on the amplifier project. I have chosen this SMPS board because it needs a 12V input to avoid my safety concerns about the insulation from mains supply: this is the duty of the upstream 12V power supply brick. On this tests I used the Mean Well GSM90B12-P1M medical grade power supply (RS components order number 831-9141). It does have full electrical specifications on the datasheet and it bears all the safety marks. It does not need a safety ground, this will avoid ground loops. Many people apparently tested this eBay SMPS board on tube amplifiers with good results, but my first test revealed a big issue. The board does not have any filtering, neither at input and on the outputs. The result is the ugly mess superimposed on the DC outputs you can see on the oscilloscope screen: up to 5Vpp of noise at the 300V output (mostly 80KHz, 320KHz and several harmonics up to several MHz) and a hylarious 1 Vpp of HF noise on the low voltage output. My dummy loads are a car lamp on the filament output and a resistor on the anode supply output. As stop-gap measure I added an empirically designed filter; it surely could be better but I believe is good enough for the next amplifier test. The ceramic capacitors directly soldered on the existing main electrolytic capacitors have cleaned up the HF garbage, and the RC / LC cells greatly dampened most of the fundamental. I also exchanged the original electrolytic capacitors (China branded) with good ones, and this further decreased the noise. The remaining spikes may be something picked up inductively by the scope probe or a ground noise from the Mean Well power supply, but they decay pretty quickly and don't are a concern for me at this time also because I don't plan to use a top quality output transformer. A last remarks: on this SMPS board there is a low value 1W resistor between the negative terminal of the high voltage output and the negative terminal of the input 12V power supply and output low voltage supply. I believe it is a current sense for overload protection. Care should be taken not to short circuit it.
Attached Images
File Type: jpg HT_factory.JPG (328.5 KB, 141 views)
File Type: jpg FilSupply_factory.JPG (408.5 KB, 137 views)
File Type: jpg HT_with_added_rc_filter.JPG (290.7 KB, 137 views)
File Type: jpg Filsupply_added_LC_filter.JPG (350.4 KB, 135 views)
File Type: jpg Filsupply_filtered_zoom.JPG (322.0 KB, 134 views)
File Type: jpg smps_board_173034019336.JPG (237.7 KB, 49 views)
File Type: jpg power supply gsm90b12-p1m.JPG (63.6 KB, 48 views)
File Type: jpg inductor.JPG (45.2 KB, 33 views)
File Type: jpg filter schematic.JPG (64.8 KB, 40 views)
  Reply With Quote
Old 27th October 2019, 08:01 PM   #392
pcan is offline pcan  Italy
diyAudio Member
Join Date: Dec 2015
6SN7 push pull flea amplifier project
I discarded the switched mode power supply board because it cannot supply the required current at 300V; i replaced it with a traditional power supply with CLC filter (4H choke with 240 ohms DCR). The amplifier worked perfectly at the first try with a 15 ohm resistor as load, ECC83 as driver tube and 6H8C as power tube. A quick scope check measures approximately 10Vpp (0.8W) at 1 KHz with clean waveform and a good symmetric clip of the waveform when the amplifier is overdriven. The minimum admissible load on the primary seems to be in the ballpark of 20K. No issues at all with 40K, but less than 20K gives severe distortion at the higher frequencies unless the output power is severely reduced.

I then tried the power and bandwith available at maximum output with several transformers, to select the best one for this circuit. The feedback connection has not been used on this test. 15 ohm load has been connected to the secundary to reflect 20K or more on the primary side. Unfortunately I was unable to find at reasonable prices the cheap Antek AS-05T240 toroidal power transformers specified by Lingwendil, but I tried other types:

- power toroid 15VA 2x6V (RS stock number 671-9053): 22Hz - 6KHz - This is a cheap generic part.

- power toroid 15VA 2x6V (Nuvotem-Talema 91935-P2S2): 22Hz - 7KHz - Good manufacturer, 2x price.

- power toroid 10VA 2x7v (Nuvotem Talema 70040K): 18Hz - 7KHz (recomended on this forum)

- power toroid 25VA 2x7v (RS stock number 124-3873): 20Hz - 8KHz (no-brand but twice the size of the previous one, at the same price tag)

- 70V line transformer OEM29A100: 110Hz - 17KHz (I honestly had better expectations)

- 100VA vintage power transformer (500V CT HT winding = primary, 5V winding = secundary): 50Hz - 12KHz but power output was half of the normal one.

Now the real output transformers:

- Vintage 6V6 PA amplifier (Bogen T260): 30Hz - 19KHz

- Cheap guitar amp transformer (for EL84 PP): 40Hz - 38KHz

- vintage Saba 10W "for ELL80", with gap removed: 15Hz - 24KHz

- big Hi-Fi transformer for EL84PP (price in the 100 euro ballpark): 8Hz - 40KHz

Maximum measured output has been 1.1W.

The vintage Saba transformer is a slightly bigger version of the one I used on my previous build described in this thread. Weight is 600g; size and lamination thickness are the same as Edcor GXSE10. I don't have a GXPP10 to compare, but it may be similar. Saba developed this transformer for ELL80 and some ECL86 radios. It has a paper gap on the lamination, this way it can be used both on PP and SE amplifiers. I removed the paper gap and pressed the lamination in the vice while checking the inductance, then I fastened the laminations with a nylon zip tie. This detail is important while dismantling this transformer. There is a separate feedback winding that I did not use.

The next step will be a test with headphones. The load needs to be 16 ohm or more with the transformers I already have, while my speakers are 6 ohm so I cannot use them. I was hoping to find a suitable toroidal transformer and maybe add a small winding for the speaker, but 7 KHz open loop bandwith will require a excessive amount of global feedback.

Last edited by pcan; 27th October 2019 at 08:09 PM.
  Reply With Quote
Old 27th October 2019, 09:21 PM   #393
kodabmx is offline kodabmx  Canada
Spreading triode love.
diyAudio Member
Join Date: Oct 2011
Location: Toronto
6SN7 push pull flea amplifier project
Triad VPT12-2080 passes 50kHz in my tests...
  Reply With Quote
Old 28th October 2019, 02:58 PM   #394
Lingwendil is offline Lingwendil  United States
diyAudio Member
Lingwendil's Avatar
Join Date: Apr 2008
Location: Oakley California
I wouldn't mind a pair of those Saba ELL80 transformers, the Saba stuff is pretty nice, and has a great sonic reputation

I've used the 70V line transformers in the past for some builds, and other than the low bass capability, they work well enough and sound pretty good. I was able to get very cheap units to pass 7-10 watts easily with no feedback, within their limitations. Good option for a garage amp or MI use.

Nice work so far! I would be perfectly happy with a 40Hz bottom edge, given the power of the amplifier. Looking forward to seeing how you complete it
Check out My 6SN7 Push-pull Flea Amplifier project! PCB COMING SOON!
  Reply With Quote
Old 28th October 2019, 11:42 PM   #395
pcan is offline pcan  Italy
diyAudio Member
Join Date: Dec 2015
6SN7 push pull flea amplifier project
I installed the Saba transformers and added a 6.3mm output: this little amplifier does really work well as headphone amp. I like it more than the cathode follower OTL I was using before. It does have more gain and power, and sound is more brilliant. Feedback inputs are still not connected, I will try with feedback later. I had to cancel my original compact chassis build plan because the SMPS power supply did not worked well enough, so I put the amplifier in a temporary clear plastic chassis with same shape, size and layout as the previous point to point octal build. This time I rotated 90 degrees the front and back panels, it seems more practical. The idea was to test the board and then transfer the whole setup on a wooden chassis of the same type as the previous one. Thankfully I made this test on the clear plastic box first because I discovered a flaw on my amplifier chassis layout. I was not obvious on previous tests with speakers. I've gone back to the octal build and it does have the same issue, despite the different power supply: there is magnetic coupling between the power and output transformers. The 90 degrees rotation of the power transformer and 5cm of space aren't enough to avoid a very faint hum while using the amplifier with a low impedence headphone directly connected on the outputs, with feedback disconnected. It becomes virtually unudible to me with the 250 ohm headphone but I know it's there and now I must remove it. After a few checks and moving things around, I confirmed that magnetic coupling is the culprit. I will rebuild the amplifier with output transformers at the front and board in the middle, but maybe I will do something completely different on a bigger square chassis, since the square shape does resonate with the way tubes are placed on this board and a bigger chassis will allow more distance between power and output transformers.

The simplifed input with omitted grid resistor on the board and 100k potentiometer seems to work well. The only issue is that such a large input resistence does pick up some noise when the potentiometer is rotated at the point of maximum resistence. I will try a 47K potentiometer.

I would say that there are no issues with this design when used as headphone amplifier. A provision for two resistors to be used as output attenuator for the headphone jack (next to the feedback input) would be the icing on the cake, but it works very well already. As general purpose speaker amplifier, the output transformer is a show stopper because the most common types cannot be used, and cheap workarounds are a compromise on quality, at least on my tests. Doubling the output tubes would solve the issue, and a noval socket for output tubes would provide more flexibility (ECC82 / 12AU7 / 6CG7 / 6FQ7 ...) with no board size increase.
Attached Images
File Type: jpg IMG_4477.JPG (303.8 KB, 30 views)

Last edited by pcan; 28th October 2019 at 11:56 PM.
  Reply With Quote


6SN7 push pull flea amplifier projectHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Chinese amp: 2A3 6SN7 6SJ7 5Z5P push pull Alexontherocks Tubes / Valves 16 7th May 2013 05:37 AM
6SN7 & 6V6 Push Pull - increase gain laosei Tubes / Valves 18 26th February 2013 09:01 AM
6sn7 driving Push Pull Parallel kt88's? ruben alstad Tubes / Valves 58 22nd November 2012 12:08 AM
VTV Push-Pull 6SN7/KT88 Amp: DIY or China? 88man Tubes / Valves 5 18th October 2009 03:27 PM
YAPPAP (Yet Another Push-Pull Amplifier Project) arnoldc Tubes / Valves 16 6th July 2006 03:51 AM

New To Site? Need Help?

All times are GMT. The time now is 01:12 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio