Retro Amp 50W Single Supply

1. I think it will be too difficult to find out a reason why particular set up is not working.
My suggestion would be to try a known set up.
If you have a trafo used for some other project like quasi amp, just put in the linear supply and see. let it be giving only 35v or so.. Atleast it will help in seeing if SMPS is causing issues or its something else.
2. try and move the speaker , if its too close

this ofcourse, after taking into consideration and acting upon 44250' suggestion, if the noise is still there.
regards
Prasi
 
you also might have neon or other lamp (modern power-saving) at your work desk. I had a lot of trouble finding a reason why I had fat line on my osciloscope - while amp works perfectly stabile and with wide bandwidth. reason was modern power saving light bulb at my work desk,when I removed it and changed for old wolfram light bulb everything was all right. just another guess...

Thanks for your suggestions,

I found this out the hard way twice previously - one time an overhead fluorescent light was inducing noise and another time a $1 usb plug was putting noise onto the ring main!
 
1. I think it will be too difficult to find out a reason why particular set up is not working.
My suggestion would be to try a known set up.
If you have a trafo used for some other project like quasi amp, just put in the linear supply and see. let it be giving only 35v or so.. Atleast it will help in seeing if SMPS is causing issues or its something else.
2. try and move the speaker , if its too close

this ofcourse, after taking into consideration and acting upon 44250' suggestion, if the noise is still there.
regards
Prasi

Thanks for the suggestions Prasi.

Seems to be the Sony SMPS in this case. I used a bulk SMPS instead of the power brick type and the noise level is substantially lower.

Maybe I can filter the noise off the SMPS otherwise I need to buy $100 transformer and try to find space in the enclosure..
 
Member
Joined 2010
Paid Member
If you shop somewhere other than retailers like Jaycar or Altronics here in Oz, you can save a lot on toroids for linear power supplies. Try RS or Element14 for their occasional runout specials on ratings from 80-250 VA, where I've saved more than 50% overall on several purchases in recent times. The manufacturing quality of even their budget lines is usually better too.

I find a dual 30V linear bench PS almost essential for testing audio gear. There is so much EMI buzzing around the home and neighbourhood these days that it becomes near impossible to test open audio assemblies on the bench. You can buy in many grades and qualities but importing a heavy power supply sure is expensive on airfreight. Get average quality TENMA from E14, RS etc, which is not much better than Ebay's junk but you can save a worthwhile amount even on their economy prices when specials are offered. Watch that the displays may be dim or inaccurate - not all models are created equal.
 
Thanks for the suggestions Ian. I'll have to keep an eye out for those sales!

I have a single 0-30v PSU at the moment. I think the ammeter lies though as I was running a JLH board on there which apparently idles at 400mA instead of the expected 1.25A (if the built-in meter is to be believed). Seemed to be putting out more than 400mA worth of heat too..
 
25W Retro Amp
 

Attachments

  • 25W Retro Amp.jpg
    25W Retro Amp.jpg
    142.7 KB · Views: 1,109
I'm using the pcb layout posted by wiljj78 at post #72. Transformer is 250va dual 28v secondaries, which gives a rectified 42v DC. Mid point voltage is set to 21v DC and bias is set at 25mA. Output Transistors are 2N3055H (Magnatec), these were £10.40 for 10 so I knew they were going to be good, cos you have to pay for quality!

Here's the thing though, it's cool, like the John Candy film (movie) "Cool Runnings". I expected the things to be finger meltingly hot but the transistor casings remain cool so I know I've done something wrong. I really should put it on the scope to see if I can work out what the problem is (I have to work out how to use the oscilloscope first though) :)

Had it singing the "blues" for an hour this afternoon, it has a certain warmth and depth to it, and more than adequate volume for my apartment. I think the AX6 is going to replace the MX50 (ebay kit).

Thanks to Apex, wiljj78, avtech23 and Prasi.

Cheers,

Graham
 
Measurement is required after building an amp. TO3 transistors work well with about .02 v across the .51 ohm output emitter resistors, or about 40 ma. This is measured with the amp silent. This is the proper amount of idle current to prevent crossover distortion. A DVM is adequate for this measurement.
These work best with a heat sink of 0.6 deg C/watt or better. Mine are just a 1/8" aluminum angle 4" long, so I installed two fans, one on each channel. Mine now run cool enough to run 18 hours a day.
 
Last edited:
Reference post #212. I had to take that AX6 (mod) board out of service because a couple of loose solder joints (p2p build) had come loose causing large popping. A new hotter soldering iron fixed that: while I had it out I addressed the perceived lack of high frequency response versus the original dynakit ST120 PC15 board still in the case with 1970 RCA TO5 drivers. I replaced the TIP31c/32c I had used as drivers with MJE15028/29. Voila! AX6 high frequency response is great, just as good as the PC15. Test tracks included top octave piano (Peter Nero Young & Warm & Wonderful, When I fall in Love) and Martin Denny Hawaii LP with high freq percussion instruments. So the 3 mhz Ft of the TIP31c/32c was really holding back the NTE181 outputs ( rebranded MJE15003 possibly), which were set free on high frequencies by the 30 mhz Ft MJE15028/29. I'm still using a GE Semi D44R2 TO220 VAS transistor, which is not a problem my ears tell me.
I'm still using the original dynaco 6.5 amp 72 vac transformer, regulated by 5 parallel TIP147 transistors to 70 vdc. That was the point of the project for me, not buying a transformer & case from a source I disrespect. Also protecting my $600 each speakers with a cap instead of a relay rated for AC loads which might weld shut under DC fault current. The original PC15 board had become damaged beyond repair due to TO220 transistors twisted in place to replace to unavailable TO5 driver transistors blown before I bought the amp.
I find the whole 30 pages of SMPS posts rather a distraction. My whole music room is SMPS free, no buzzing lamps or lamp dimmers either. The PC + flat screen TV etc with SMPS live 30' away on a different circuit. The OEM dynakit steel case with perforated cover helps keep AM radio & police/CB band out of the circuit, and is important in my location to stress free listening. Back to using this amp 14 hours a day, 1/4 W base level, 70 W or higher peaks on 1812 Overture etc. Using 101 db 1W 1m speakers.
 
Last edited: