Dx Blame MKIII-Hx - Builder's thread

Status
Not open for further replies.
Where was your bulb tester at the time? In the cupboard, maybe?

I will use that bulb now Andy..

though this happen, I got no blown fuses, no burning trannies or smokies parts..maybe bias problem?


Will be back on that later, let me pick the wifey first ;)
 

Attachments

  • Testing.JPG
    Testing.JPG
    684.4 KB · Views: 751
Hi mister Junie! did you use a 45+45 toroidal transformer? is about 63.63 45x1.414 close enough to 64VDC rail to rail, maybe is a solder, I always check that is so damn small that solder can ruin connection between tracks well that can not be because that can cause burning tracks, now you got me thinking what can that be uhmmm?

regards
vargasmongo
 
Is your bias trimpot adjusting?... say... changing current when adjusted?

Your current changes when you tweak bias trimpot?...if yes, then increase base to emitter resistance in order to decrease your current....if not, then search because you made a mistake..... you see that both amplifiers presents the same problem....seems the same mistake was made.

Be sure i have tested and i have tested with 84 volts supplies.... also i have tested with 4.5 volts, 12 volts, 20 volts, 35 volts, 45 volts, 55 volts, and 64 volts.. and worked fine in all conditions.

Check bias transistor position, check insulators, measure transistor colector and check if they are really insulated from the heatsinks.

Use a lamp in series with your transformer as your resistor will smoke soon and go searching...the mistake will be there waiting for you.... i am sure sooner or latter you will find it... relax...take a breath... take a rest...and go searching.

Picture is celebrating we now have a lot of oil...Brasil is becoming rich....80 percent of the population have high speed internet connections (120 millions with 10 Mega ADSL connections)... our payment was multiplied by 5 in the last ten years...we are giving money the the International Monetary foundation.... you see we gonna be the next important countries providing oil (also Ethanol) to the whole world.... we gonna be the next "Arabian Petrol Scheik"

regards,

Carlos
 

Attachments

  • img_6082.jpg
    img_6082.jpg
    232.3 KB · Views: 640
Last edited:
Carlos, I was wondering; for the initial testing, can we use a low-current say 12V+12v 20VA transfo, or even two 9V batteries, just to see if circuit is functioning properly?

It's kinda risky to connect the "big engine" pushing 500VA into the board for the initial shakedown.

If we adjust the bias using a 15VDC+15VDC, will we need to re-adjust it at full 64VDC+64VDC ?

Regards,
Martin.
 
Yes, you can use two 9 volts batteries if you want

The problem is that if you start having high drain of current, your battery will discharge very fast.... your 15 volts low current supply will be better...but be aware you may suck more current than your regulators (If LM7812) can allow you to use...but give a try if you have a non regulated (without IC regulators in the output as series pass voltage regulators).

I cannot remember, but i suppose the presence of a CCS will alow you to adjust once..... as i said, i am not totally sure about....maybe a small fine tuning will be needed in the stand by bias trimpot....because the voltage will be really a lot different than the standard voltage... CCS can help but cannot make all that magic...maybe will be out of range of CCS to keep things in order.... and we have a bootstrap following the CCS and this one produces voltage drop too.

Sorry, i cannot remember.....but i do think you will adjust only once.... or will need a small touch after first low voltage adjustment.

After your future testing, please inform your results to confirm or dennie what i am saying in order to help others.

Man..... i cannot remember....terrible when we reach 60 years old and we start to forget important things.

regards,

Carlos
 
Last edited:
A friend from other forum told me these two heatsinks are reaching 52 degrées celsius

these units are too much hot... 52 degrées Celsius (Centigrades)... this is too very good.... better if you can keep these ones at 45 to 47 degrées.

A taller, or bigger heatsink will be needed for him to cool down these two transistors mounted in separated heatsinks.... they are lovely but small, not that tall as i was dreaming for.

The guys is using 68 volts supplies...... and the higher the supply voltage the bigger will be the problem related these heatsinks.

In my construction, home prototypes using standard supply voltage, i have not noticed heat there because these transistors where mounted into the main heatsink...but seems you may face some trouble because heat there.

Be aware you may have to replace these heatsinks for larger ones.

regards,

Carlos
 

Attachments

  • DSC_5053.jpg
    DSC_5053.jpg
    300.2 KB · Views: 753
Dear Vargasmongo,
Explain! what is different?
thanks,
Francois

No, is about the MKIII green board and the red one the 220pF cap is oriented, in a different way that is all, mister Junie show me, that is way I say that, nothing to worry about I'm sorry I'm Puertorrican I get exited for everything

regards
vargasmongo
 

Attachments

  • red board.jpg
    red board.jpg
    69.7 KB · Views: 702
No, is about the MKIII green board and the red one the 220pF cap is oriented, in a different way that is all, mister Junie show me, that is way I say that, nothing to worry about

Yes, the board is very crowded. I recall moving some things around to increase the space between tracks and between pads a bit. the schematics are the same. I apologize for any panic I may have caused.:cannotbe:
 
Understanding the trimmer resistor

I'm getting to that point, so I'm re-reading Carlos' old posts - carefully this time. I want to make sure I understand Post #3 in this thread. Of course I could just blindly follow Carlos' instructions, but I want to understand why.

His instructions say that with a 64V supply, we should adjust our trimmer resistor so that the temporary 100ohm resistor in series with the rail voltage should have a ~4.4V difference across it, and result in 0.044A flowing from positive rail to ground. (Presumably -0.044A flowing from negative rail to ground.)
Using Ohm's law, I calculate that 64V=0.044A*(100R+XR), where X is the resistance from supply voltage to ground in the absence of the 100ohm safety resistor. Some high-school algebra tells me that X=1355ohms. Is this correct? Though I can't figure out where this 1355 ohms comes from, it seems that much of it comes from our trimmer resistor. We are setting the resistance of the trimmer to set the resistance from supply-to-ground to be ~1455ohms in the presence of, and ~1355ohms in the absence of, the temporary 100ohm resistor. If the voltage across the 100ohm safety resistor is 4.4V, then the resistance of the rest of the circuit must be ~1355ohms. Is this correct? And if we can't find any setting of our trimmer resistor which sets the voltage across the 100ohm temporary resistor to = 4.4V, there must be something wrong with our amp.

the problem is that when I check my math by calculating voltage drop from known resistances, I come up with different numbers.
In the absence of the 100ohm resistor: I=64V/1355ohms=0.0472A
In the presence of the 100ohm resistor: I=64V/1455ohms=0.0440A
Voltage difference due to the 100ohm resistor: V=(0.0440A-0.0472A)*100ohms=-0.323V. This value should be -4.4V:cannotbe:

Well, it's too late at night for me to concentrate on this anymore:sleep:

-Byron
 
Yes, more or less your calculations Byron...too late to my old brain too

People that feel these 2 small vertical heatsinks too much hot can reduce VAS current in order to reduce heat there.... i have simulated and i found the original schematic fine... with the modification you will have changes in performance, but not that huge.

The 22 ohms resistor that sets the CCS current can be increased to 47ohms in order to reduce current...then you will have these transistors mounted into vertical heatsinks (CCS to VAS and VAS transistor) less hot.

I found power there not that big...the left one is the CCS transistor, CCS to feed VAS (Voltage Amplifier Stage) and it is dissipating 1.65 watt.... the rigth side one is the VAS transistor, the second stage, the most active unit, following the buffer and the power there, dinamic power, will be around 1 watt....not that huge power and the small heatsinks should fit in most of the cases... exception is when people is using higher supply voltage... say... really higher voltage.

You can replace the 22 ohms resistor by a 27 ohms, 33 ohms, 39 ohms or 47 ohms.... i suggest you to go to 47 ohms because will really cool down in order to work fine with tiny vertical heatsinks... with this resistor replaced, the performance when using low impedance output loads will clip earlier.... some loss in power...not that huge, not that important.

Some high current. high voltage and high speed diodes installed into the output, in the power transistors, from colector to emitter, can be used too as we gonna face some back EMF there.... so, if you intend to push the amplifier hard, then search for high speed, high current diodes used to fly back purposes there...what model to use i cannot suggest you as i do not use them, so they are not familiar to me... search for high power amplifiers and copy the diode model to apply to yours....this is to play heavy metal, full power, distorted sound and connected to 2 ohms loads... an extra protection if you intend to operate your amplifier above the limits, entering distortion and draining huge ammounts of current and generating enormous output having a lot of back EMF in your output line.... normal civilized use will not need these diodes.

If you decide to use alike dB competition of bass, operating in clipping mode..for sure you will need these protective diodes.

Enjoy your amplifier boys...sound is awesome and you will perceive that instantaneously.

regards,

Carlos
 

Attachments

  • Small vertical heatsinks.jpg
    Small vertical heatsinks.jpg
    440.8 KB · Views: 682
Last edited:
I'm getting to that point,.......................

Well, it's too late at night for me to concentrate on this anymore:sleep:
what you have calculated is basically correct.
The big difference between your assumptions and the real circuit is that the 1355r is not a resistor, but a complex combination of resistors and semiconductors.
The other assumption that does not hold, is that the +ve and -ve currents do not need to be equal. If they were then the ground would not pass any quiescent (no signal) current.

If you apply the correct supply voltage to the amplifier when the output bias is turned down to minimum then you can start setting up your amplifier.
Using Carlos' method.
Power down and insert a 100r in place of the +ve rail fuse. This 100r could be soldered permanently across the fuse holder terminals so that all you need to do to set up the amp is remove the fuse.
Power up using the bulb tester to prevent damage if something has been changed by accident.
Check the bulb is off or has a VERY DIM GLOW.
Measure the voltage across the power supply terminals.
Measure the voltage drop across that test/set-up resistor.
Now slowly increase the bias voltage across the Vbe multiplier. Use small increments and measure the bias voltage, measure the test resistor voltage.
Continue increasing the bias voltage. You will probably find that the bulb will become increasingly bright. Do not change the bias setting. Change the bulb to a higher wattage, or if you are really sure you won't make a mistake power up directly from the mains.
Recheck the bias voltage and the test resistor voltages. They will have changed since moving from the low value bulb wattage.
Why have the test voltage changed, I only removed the bulb tester. They change usually increase because the Power supply voltage will have risen.
Now start increasing the bias voltage checking both those test points. Keep increasing until you reach the final setting given by Carlos. Check device temperatures check for burning or overheating smells. Check device temperatures. Wait a while and recheck device temperatures.
When the amp reaches it's final operating temperature, 30 to 60 minutes say, recheck the test voltages. Readjust the setting is necessary. replace the fuse. You have now changed the PSU voltage. The test voltages will have changed, but you can't measure them, because you have bypassed the 100r resistor with a fuse.
Job done.

BUT DO NOT connect your speaker yet.

Short the input signal hot to signal ground. measure the output offset at the speaker terminals. Measure the output noise at the speaker terminals. Power off. Let the amp cool a bit.
Power up and recheck the output AC and DC voltages. Leave the amp running with shorted input for many hours with the occasional OFF/ON to stress the components. When you are sure that the amplifier is behaving itself, then it is time to connect a cheap speaker.
Listen for any hum.
Power OFF. remove short from input. Power ON. Is the speaker still working? Can you hear any noise? Is the cone still at it's centre travel position? Measure the AC & DC voltages at the speaker terminals. Are they similar to those from before.
Power OFF, connect an audio source. Power ON. Recheck your voltages. Is the speaker still OK?
Now listen to some quiet music. Does it sound "clean".

Job really done.
OFF, connect your real speakers, ON, enjoy.

Do you want to hear about the correct way to adjust the bias of DX amplifiers?
 
Last edited:
Status
Not open for further replies.