My 100 watt amp

[gaetan8888]

Hello

Here is a schematic of an amp that I've made for a friend a year and a half ago, I prefer simple circuits since more complicate one don't allways sound better.

I've re-simulate it (but this time using LTspice) at 100 watt, it's a Linn topology with ccs and LTP but no CM. it use a boot strap and no CCS for the vas, and a phase lead capacitor so the miller capacitor (lag compensation) can be lower with the same stability. It look good at sim thd and should sound very good.

It sound better if you matched the two LTP input transistors for HFE and VBE.

You adjust the offset using R7, for minimum dc output. Bias are adjust with R19 at 60 ma.

You need to use a low cob and fast speed for Q7 the vas transistor, at least 120 mhz with a cob lower than 10 pF.

The miller cap (lag comp) C12 and the phase lead cap C13, should be silver mica.

If you use faster driver transistors, you can lower the value of the suckout capacitor C11 to arround 175 nF.

All capacitor should be rated at 100 volt.

For those who would suggest a CCS for the vas, I founded the boot stap to be better sounding, so I use boot strap.

This amp will sound very musical with a very good sound-stage.

Btw, R2 are no more to adjust the offset, so forget the text saying that R2 are for offset adjusting. I forget to delete it and I can't modified the posted images.

I include images of the circuit, the thd spectrum and also a very simple vbe test for the matching the two LTP input transistors.

I thank Hugh Dean for his good help and teaching since few years, without him I would never done this amp.

Bye

Gaetan

[gaetan8888]

Hello

Here is a rev.1 of the circuit of my amp, I've deleted the the text saying that R2 are for offset adjusting.

You adjust the offset using R7

Bye

Gaetan

[AKSA]

Congratulations, Gaetan!!

Nice amp..... you might need a bit more adjustment on R7 to achieve offset control. I'd suggest making R30 820R, and R7, the pot, 200R.

When output offset is adjusted to within +/-20mV, the two arms of the LTP should be in almost perfect balance if the matching of the diff pair is good.

This does mean matching beta, of course, but also the Vbe with 1mA flowing through the device.

Ciao,

Hugh

[Bonsai]

Nice one Gaetan! Happy listening!

[gaetan8888]

Originally Posted by AKSA

Congratulations, Gaetan!!

Nice amp..... you might need a bit more adjustment on R7 to achieve offset control. I'd suggest making R30 820R, and R7, the pot, 200R.

When output offset is adjusted to within +/-20mV, the two arms of the LTP should be in almost perfect balance if the matching of the diff pair is good.

This does mean matching beta, of course, but also the Vbe with 1mA flowing through the device.

Ciao,

Hugh

Hello Hugh

R7 and R19 value was for the sim testing, but R7 trimpot value should be 200 ohm, and R19 trimpot value should be 500 ohm and forgot to bring them back.

Sorry but I have problems to see details on my computer screen, so I forgot details quite often.

I include a rev. 2 of my circuit with right trimpots values.

Thank

Bye

Gaetan

[gaetan8888]

Hello

I've forgot to say that this 100 watt amp was made by me for a friend a years and a half ago and test with a Denon cd playe with Dynaudio loudspeakers and it was very good sounding.

Bye

Gaetan

[soekris]

And maybe lower the R15/R23, a common value is 2R2, will give you a couple watts more output power....

Soren

[destroyer X]

And you have include some new features.... very good.

regards,

Carlos

[tessier]

Hi

I've done the rev 2 schematic and used 10p for the c13 cap.

This amp sound delightful, Diana Krall and Norah Jones sound so good, even orchestral music are a bliss with this amp.

Thumb up for you sir

Btw, his there anyone who know a simple DC detect circuit ?

Thanx

Paul

[Eugene Dvoskin]

Originally Posted by gaetan8888

Hello

Here is a rev.1 of the circuit of my amp, I've deleted the the text saying that R2 are for offset adjusting.

You adjust the offset using R7

Bye

Gaetan

R7 is to set current through input differential pair. Adjusting offset this way (it will make some adjustment, because diff pair transistors are not ideally matched) is more or less pointless.