The Art of 1731

[fobi]

Hello,
I just finished summalysing my thoughts about Melcor 1731 amplifier, file attached. Comments are welcome.
I think i can do the same for API 2520 with help from you, fellows.
BR

 

[shabtek]

"Protection: These amplifiers work with transformers, meaning a highly inductive load when the output of the transformer is open. Inductive load can return back-EMF voltage. The danger here is having the emitter voltage higher than the collector voltage forQ4 and the opposite for Q7. Having a current trough D2 by excessive output voltage will block transistor Q4. The resistors R13 and R15 provide some guard to the transistors in such events plus additional current surge defense"

R15-- or R12?

 

[fobi]

"Protection: These amplifiers work with transformers, meaning a highly inductive load when the output of the transformer is open. Inductive load can return back-EMF voltage. The danger here is having the emitter voltage higher than the collector voltage forQ4 and the opposite for Q7. Having a current trough D2 by excessive output voltage will block transistor Q4. The resistors R13 and R15 provide some guard to the transistors in such events plus additional current surge defense"

R15-- or R12?

Yes, sorry, R11, R14.
And 'defense' must be defence.


These resistors have additional functions, alongside with the Miller capacitance they are reducing the gain at RF.

 

[TheseGoToEleven]

I recently took apart an original Melcor 1731. I need to complete a few more measurements before doing a full post about it.

I found that C2 and C3 were not connected to each other. C3 (as labeled in your schematic) was instead connected to the -IN pin.

Your comments about the use of a stabistor in place of the combo of D1 & D4 match what I found. The part number was 1N4829. The other two diodes were 1N4727. The transistor part numbers also vary from what you have in your schematic, and what I have seen in others, so it seems like the bill of materials changed over time.

EDIT: My 1731 teardown has now been posted here.

 

[TheseGoToEleven]

I love your analysis, by the way! I think a lot of curious minds can learn from it. If you want to do an analysis of the 2520 amplifier too, that would be fantastic. But be aware that some 2520 schematics have an error, where the base of the "push" transistor is not connected to the stabistor, but somewhere else instead.

About the 1731, it definitely took me a minute to figure out that the reason R6 and R9 have triple the value of R5 and R7 is that, for Q3, there are three PN junctions in series between the base and the emitter resistor, instead of just one. So the equation for Q3 collector current would be:

We know I1 = I2, so if we set R6 = 3 x R5 and R9 = 3 x R7, this turns into:

The 3's can be factored out of the top and bottom, giving:

This is identical to the equation for the collector current of Q5, forming a current mirror. I think it would be nice to add a similar explanation in your 1731 analysis, just to make it easier for readers. (I know it was implied but I, for one, didn't make the leap right away.)

 

[Golgoth]

Hello,
I just finished summalysing my thoughts about Melcor 1731 amplifier, file attached. Comments are welcome.
I think i can do the same for API 2520 with help from you, fellows.
BR

Incredible work

 

[fobi]

I love your analysis, by the way! I think a lot of curious minds can learn from it. If you want to do an analysis of the 2520 amplifier too, that would be fantastic. But be aware that some 2520 schematics have an error, where the base of the "push" transistor is not connected to the stabistor, but somewhere else instead.

About the 1731, it definitely took me a minute to figure out that the reason R6 and R9 have triple the value of R5 and R7 is that, for Q3, there are three PN junctions in series between the base and the emitter resistor, instead of just one. So the equation for Q3 collector current would be:
View attachment 125170
We know I1 = I2, so if we set R6 = 3 x R5 and R9 = 3 x R7, this turns into:
View attachment 125171
The 3's can be factored out of the top and bottom, giving:
View attachment 125176
This is identical to the equation for the collector current of Q5, forming a current mirror. I think it would be nice to add a similar explanation in your 1731 analysis, just to make it easier for readers. (I know it was implied but I, for one, didn't make the leap right away.)

Incredible work

ta

 

[fobi]

I love your analysis, by the way! I think a lot of curious minds can learn from it. If you want to do an analysis of the 2520 amplifier too, that would be fantastic. But be aware that some 2520 schematics have an error, where the base of the "push" transistor is not connected to the stabistor, but somewhere else instead.

About the 1731, it definitely took me a minute to figure out that the reason R6 and R9 have triple the value of R5 and R7 is that, for Q3, there are three PN junctions in series between the base and the emitter resistor, instead of just one. So the equation for Q3 collector current would be:
View attachment 125170
We know I1 = I2, so if we set R6 = 3 x R5 and R9 = 3 x R7, this turns into:
View attachment 125171
The 3's can be factored out of the top and bottom, giving:
View attachment 125176
This is identical to the equation for the collector current of Q5, forming a current mirror. I think it would be nice to add a similar explanation in your 1731 analysis, just to make it easier for readers. (I know it was implied but I, for one, didn't make the leap right away.)

Yes, it's good to have more ways to explain why this stage is a piece of art Thanks.
Also, 2520 is on my TODO list, I even have a template, but capitalism doesn't give people too much free time ...