Help with this amp? A patchwork product of simulation

[MJL21193]

My family had a 1967 Firebird and I agree the styling really is timeless. It really was a pony car as the handling didn't improve significantly until the Trans Am.

I love em, always have, always will. Another one of my favourites from that era is the "70 Mustang Mach 1. Optional 428 (thats 7 litres kids) just a monster of a motor. To put things in perspective, I had a full size school bus (briefly) with one of those engines.

I have been thinking about the case for this amp, trying to come up with a design that matches my taste and the things I already have. I've decided to make these mono blocks with the transformer in a separate case, supplying both (or three is what I really need).

I've built cases from aluminum before, but I'm really tired of this look. I want something sleek, distinctive. The baffles of my new speakers are piano black and I want to do the same with these amp cases.

I came up with this. The general shape and proportions are there but picture it gloss black. The entire body will be made from MDF, except for the back panel. I will line it with foil tape. Feet will be 3" diameter x 1/2" high stainless steel - polished.

 

[GK]

Very nice Cougar restoration there Radioman, I'm not a fan of Fords but I do like the Windsor motor. The car and motor look fantastic!

Pete B.

I hate to be a party pooper, but the Windsor engines were prehistoric, unresponsive and excessively heavy lumps even in their day. Here in Aus the Cleveland motor (a much better design) has legendary status due to the Ford GTHO.
The only thing I like about an old school Ford V8 is the fact that they put the distributer at the front (makes tuning a hell of a lot easier).

Incidentally, here is the worked high compression 384 cu. in. engine I recently finished bolting togeather for my 2900 pound daily driver.

 

[Radioman62]

Hrmm .. About this amplifier again

Thanks to John it's my turn to tinkering in the simulator.

First I would like to say that the resistor R38 from the latest schematic (post #582) doesn't do anything more then upping the distorsion a thousand of a percent. Both in the simulation file from John and in my slightly fiddled one for lower Voltages.
The meaning with this resistor has been mentioned in this thread before by both Pete B and Andrew T. There has been mentioned some article or book by Mr. Self Is there anything I can read on the Internet? Can someone link me so that I can edjucate myself in this mystery.

John, you said earlier (I think) that your simulator showed the contrary!? That the distorsion is less with R38 in place.

Second What happend with the two pole compensation? Was it really the cause of oscillation. Maybe it just was the 10pF feedback capacitor all the time, that Donk figured out? (Donk also suggested the two pole compensation if I remeber right)
John, have you tried with the two pole compensation again?

Third
Pete B, I have gone towards your constant current sources and without the bootstrap capacitor to gain.... nothing Many more components but not lower distorsion. Anyway it's very edjucating this simulation process. Have you brought your variant of the patchwork amp any further?

I hope I will be back soon with a scematic and working simulation with the new ThermalTrak device's i sampled.
As allways I have to much going on at the same time, f.e. a large RIIA and headphoneamp project... with tubes... that have been in my brain for 27 years ... So I probably should finnish that project before I throw myself into this? Maybe I'm not alone to have several unfinnished things going on at the same time ?

 

[MJL21193]

Thanks to John it's my turn to tinkering in the simulator.

First I would like to say that the resistor R38 from the latest schematic (post #582) doesn't do anything more then upping the distorsion a thousand of a percent. Both in the simulation file from John and in my slightly fiddled one for lower Voltages.

John, you said earlier (I think) that your simulator showed the contrary!? That the distorsion is less with R38 in place.

Hi Ove,
First:
I forgot to mention that for some reason Multisim doesn't handle the Vbe multiplier the same as in the real world. You will need to adjust the resistor values in a prototype to put the quiescent current in the right range. This depends upon the transistor that you use for this operation.
In Multisim, check the quiescent current through the output - it's seems to like being in the 60-100mA range. I'm using ~70mA in the sim and in the prototype.

With this set, read the distortion at 20khz near full output with or without the bootstrap resistor (R38) in place. I see a reduction in distortion by 0.002%. At other frequencies it doesn't seem to make a difference.

Second What happend with the two pole compensation? Was it really the cause of oscillation. Maybe it just was the 10pF feedback capacitor all the time, that Donk figured out? (Donk also suggested the two pole compensation if I remeber right)
John, have you tried with the two pole compensation again?

The two pole compensation was added when the sim was showing high distortion at the higher frequencies (0.029% @ 20khz). Changes in component values and devices have reduced this number to the point where two pole wasn't needed.
In the course of tracking down the source of the instability, I eliminated it and see no reason to put it back in.
I can't say for sure if it had any effect on stability.

I hope I will be back soon with a scematic and working simulation with the new ThermalTrak device's i sampled.
As allways I have to much going on at the same time, f.e. a large RIIA and headphoneamp project... with tubes... that have been in my brain for 27 years ... So I probably should finnish that project before I throw myself into this? Maybe I'm not alone to have several unfinnished things going on at the same time ?

Good luck with your version.
I also have too many projects happening at the same time. It's good though - you get bored or frustrated with one, switch to working on the other.