diyAudio (
-   Solid State (
-   -   Overheating on NEW DX Blame super (

DIYGEEK 12th February 2012 03:09 AM

Overheating on NEW DX Blame super
Looking for a little help on a DX Blame build, it’s the supercharged version.

I etched the board and soldered in the components, it took me about 4 days to do this, I loved the build all the way. Looking forward to using the laser printer method, this board was laid out by hand. Also love the community that designed and release something wonderful like this for me to enjoy.

This amp worked right out of the box, I sent 10 volts on a dual supply, 1 amp max output. It worked and seemed to have lots of gain, had the MP3 player on volume 2… When pushed it distorted due to supply voltage. I set the voltage to 27 volts 1 amp supply and again it worked better, the transistors seemed to heat normally and could run like this for quite a while. I measured the current draw at 10mA

I upped the power supply to 32 volts 4-5 amp max draw… and added two 10,000 uF capacitors to the power supply. The amp worked perfect and sounded pretty good, base was still lacking at this point and the output transistors heated quickly and I had to shut it down 45 sec in. Transistors are mounted on a large insulated heatsink. I measured the current draw at 3.87 per side. Voltage was 32 volts then dropped to 17 when connected to the amp, volume did not change voltage or current draw by more than just a fraction.

Variables – I am using two 2SC5200 & 2SA1943 transistors and MJE15034 & MJE15035 ALL component values are identical to the ones published except for input capacitor (I have .47 instead of 1 uF) and I guessed on the output coil. My estimate would be the coil is probably closer to 2 uH instead of 1.

Any ideas of this overheating problem?

I do not hear any oscillation but input gain seems high. I have some noise when 100k volume is turned down, less volume more input noise.


jaycee 12th February 2012 07:23 AM

You wont hear oscillation as it is often at a frequency well out of the normal hearing range.
Check R28 with no input and see if it is getting hot. If it is, the circuit is oscillating.

Also check that you have biased the amp correctly. I noticed this is not described in the guide. First set P1 for maximum resistance, then connect a voltmeter on millivolts range across R26. The output bias current is given by (R26*voltage) so for example 6.6mV would correspond to 30mA of bias. You should see a little more than twice this being drawn from the supply when the amp has no input.

AndrewT 12th February 2012 11:06 AM

set the amplifier up for testing before you connect any speaker.

Short the input.
Leave the output open circuit.
Insert a bulb tester into the mains feed.
Turn the bias adjust pot to maximum resistance.

Power up.
Is the bulb off/on/dim/very dim ?

DIYGEEK 13th February 2012 02:08 AM

I’m sure it has to do with the biasing. I set P1 to 160 ohms like schematic said and that’s it. Everything on the amp seems to work fine when ran at lower voltage.

I just measured R26 at 11mV with bias set all the way at 500 ohms. Then I cranked bias and this changed to 12.9mV

I also pulled the fuse and measured 29.8v across the 100 ohm resistor, after several moments of running the resistor got hot and I had to shut it down.

The input was shorted but I just jumped + and – input, Do I need to also short it to ground also?

My biasing does not seem right this should be around 2.5 mV right?

Here are pictures of my amp, I cant post here yet due to missing security ring? Think I am too new. 1c481481

I also used these vids to help set measure my results.

Any ideas of what I should do next with this bias?


jaycee 13th February 2012 06:51 AM

From your pictures, you have substituted that NTE375 in place of the bias device BD139. The TO-126 and TO-220 devices have different pinouts - one is the reverse of the other.

Also your 2SA1943/2SC5200 transistors look like they may be cheap copies or fakes. If so, that could be causing the oscillation.

DIYGEEK 13th February 2012 02:49 PM

Thanks, this site and the members are the best...

I was doing some recherché last night and discovered that minimally I would have to use a different bias setup due to the transistor change using NTE375 would be different than the original schematic. I did not check pin out though and that will be the first place that I start today. When I asked the local electrical store for a BD139 sub you would think things like reversed pins would stand out in the cross book… a sub should be equal to or better than the original device… but that’s not what NTE cross does…

Thinking I might use a MJE15034 or MJE15035 because that’s what I have left to use… I will certainly order a bunch of BD139, seems like it’s a favorite of many designers. My bias will still be off and I may have to play around with values to get the right mV setting if I use one of the MJE (correct npn or pnp of course I have to look it up)

Once I have the correct gain THEN I will find out if these are fakes or not… I was concerned upon ordering them, I know fakes exist and probably should not have ordered from Ebay even though it was a US seller.

I would appreciate further comment if someone can indeed tell if these transistors are fakes due to what they see.

Its been years since I picked up the soldering gun, I’m pretty happy with the results so far and the fact it worked right away… and now I will learn more while getting it to work right… the learning is what I am after and its been good so far.


Stormrider 13th February 2012 03:58 PM

MJE340, MJE182, 2SD649 among many others would be good subs for the BD139. NTE parts are typically not recommended.

The MJE15034 will not work, as the pinout is reversed. It needs to be a TO126/TO225 NPN.

*Edited to correct my stupidity...

AndrewT 13th February 2012 04:24 PM

the bd139/140 are To126

It's the bigger To220 that are reversed.

butch_vlad 13th February 2012 06:37 PM

Output transistors look fake
Brand is Japan? :)
Make sure you replace them, you might burn other components when your fakes burst into flames. If you have a spare one brake it and post a picture with the die inside it. Maybe you are lucky and they are not too worthless.

DIYGEEK 13th February 2012 07:58 PM

Scratch that, a quick look and the best choice is the MJE340 as bias transistor replacement, and I have those.

The amp works much better now in this configuration…. A good catch, thanks!

Now I am sending 31.2v per rail and 20k storage caps everything is cool with inputs crossed.

The 100 ohm across removed fuse is not hot and I measured 3.2 volts so that gives me 32mA.

Then I measured bias voltage at 7.3mV across 0.22 ohms resistance = 33mA.

I then played music and it distorted pretty quick. I adjusted bias on the fly until it sounded good, this is at the halfway point pretty much like the schematic stated.

I am still not sure what bias voltage I am trying to achieve with these given values?

Last I measured the 5200 & 1943 base to emitter voltage… The NPN had 590mv and the PNP 630mv. After bias I will need to check offset voltage these should be the same right?

All times are GMT. The time now is 10:36 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio