Building a new Amp

[sbrook]

I'm in the process of planning to build a new amp system (multiple channels instead of just Stereo )

I was SO impressed with the Blomley amp that I built back in the early 80s that I want to "play it again Sam". 20Watts is more than adequate for my listening levels.

I had lots of problems finding Transistors suitable for it way back when, that I ended up doing many substitutions that have all worked out OK. But, I'm now in a deeper hole with replacing "what works".

So far I've found what appears to be suitable replacements for the BC182 / 212 in the BC546 / 557 and I can lay my hands on MPSA05/55.

I used MPSU05/55 and BD131A/132A (in place of the MJ481/491) in the darlington output stages.

Now, I cannot lay my hands on any of them in user quantities.

I'm considering replacing the MPSU05/BD131A with a TIP120
and the MPSU55/BD132A with a TIP125, both darlington pairs.

Now, at one time 30 years ago, I could wrap my head around doing the substitution ... now with 30 years of rust (I moved to computing after graduating and have barely touched anything that isn't simple single transistor stages or just switching circuits since!

So, what I know is that looking at the TR9/11 darlington (same for the other half) in the Blomley R13 is 1K and R15 is 100R

Now, looking at the TIP120, the equivalent of R13 is 8k and R15 is 120R.

So, this looks like I'm going to have to alter the biasing up the DC chain!

I'm showing my rust and begging for some of the knowledgable folks here to make some comments ... like other options for the substitution!

After the Power amp is done, I'm looking at the LM 833 or 837 to build a new preamp with mixer style controls.

Thanks for any assistance

[sbrook]

Just so you don't have to chase over creation to find the circuit I'm referring to ...

[CBS240]

IMHO, you should use a double ended PS (+V GND -V) and loose the output coupling cap. I wouldn't use darlington transistors for the outputs, but 2 transistors in darlington connection as is in the drawing. With darlington packages, the temp-co is more positive(Vbe decreases W/temp) because the driver is on same die and has same temperature as output. Can be more of a pain to thermally stabilize bias in class A-AB.

There exists better transistors these days, with more gain, linearity, and Ft. If you choose a faster output device, you might include base stopper resistors or adjust the compensation for stability.

It would be interesting to see the results of this circuit recreated with more modern components.

[sbrook]

Good point about temp coeffs ... Certainly the MJ481 and 491s are as archaic as the 2N3055!

The problem then comes as to WHAT to choose ... I've been poring over data sheets till I'm blind!

[CBS240]

I generally look for linear gain, high Ft, low capacitance, and of course SOA.


There are a lot of folks here knowledgeable of this subject, they seem to be silent.

[aguantesoda]

The Darlingtons pair are 2 Tips(142 for NpN; and 147 for PnP)
Resistor in the emitter; is 0,47 ohms.
R3 is 47k
Supply is +25v 0 -25v
Rl: 8 or 4 ohms

You get MAX ~30 Watt power / 8 Ohm
And 40-50 Watt into 4 Ohm speakers

you must put Q4 on the heatsink; near or between the Tips.

I hope that it serves you!

*El par Darlington Son 2 Tips(142 para el NpN, y 147 para el Pnp).
La resistencia del emisor es de 0.47 ohms.
R3 es de 47k
La alimentación es de +25v 0 -25v
Rl: 8 o 4 ohms.

Obtendrás una potencia Máxima de 30 Watts, sobre 8 ohms
Y 40-50 watts sobre parlantes de 4 ohms

Debes incluir a Q4 en el disipador; cerca o entre los Tips.

Espero q te sirva !

[CBS240]

I'm kinda interested in the results of the single input transistor circuit instead of a differential such as in the first circuit, http://www.diyaudio.com/forums/attac...amp=1103230029

With a double ended supply, could a j-fet funtion better as the input transistor? Maybe someone will simulate it and compare.

[sbrook]

I've always been wary of small signal jfets as being generally noisier than a bipolar, and I don't want to inject noise excess noise into the signal splitter.

The only problem I've had with this amp is that under certain conditions, I can push it into RF feedback at about 250kHz depending on "noise" from the tone controls (active Baxandall). I only had limited time with a scope to try to diagnose what was happening and didn't get a resolution other than the Zen approach ... "Doctor Doctor, My arm hurts when I do this" "So, don't do that!"

[CBS240]

I have found that having a scope and function generator is very helpful. It is almost impossible to create a new circuit without it because you won't easily know if it behaves well at high frequencies(well above audio) across the intended load unless you can see that it does. Of course there are symptoms, but if you can visually verify what is going on, it makes it easier to correct, and less trial and error. IMO, a good investment.

[sbrook]

I think I've found a decent output pair ... the MJE243/253

Now to find something to replace the Uniwatt MPSU05/55