Sanken A1673/C4388 transistors

Status
Not open for further replies.
Hey all,

I've been seeing a lot of Sanken transistors showing up in things I've been ripping apart in order to expand my elementary knowledge of amplifier design. They are in my Sony 6.2 receiver, and in the subs that came with it, and a few other things I've cracked open and looked inside.

The last victim to my screwdriver is a Pioneer "760 watt" stereo car amplifier with 4 pairs of A1673/C4388 transistors (PNP and NPN respectively, so am I right to assume this amp is in a push-pull config? Like I said... very new). Also inside are 6 Sanken FKV550N MOSFETs.

For the sake of experimentation, is it possible to assemble even a simple class-A amp from one of these transistors? The simplest schematic I've seen online calls for 3 resistors, 2 capacitors and a power transistor, and nothing more (other than a healthy supply of volts and the associated wiring of course). I'd love to throw this together on a breadboard just for the hell of it, even if it ends up eating one of my sacrificial speakers for breakfast. The only problem is the schematics I've found don't mention WHICH power transistors are suitable.

Experimentation aside, are Sanken transistors any good for a DIY build? e.g. a nice class A stereo amp, along the lines of Nelson Pass's ACA (though I imagine these transistors are good for a fair bit more than 5 watts - data sheet says 100watts for both transistors as a complimentary pair).

Bowing to your superior knowledge,

Richard
 
Richard, I suggest that for your first design you would be well advised to aim for simplicity and highest possible probability of success. If this calls for other transistor types that you are forced to purchase new, so be it. I recommend a conservative class-A or class-AB design involving an NPN Darlington (TIP122) and a PNP Darlington (TIP127), driven by an opamp (LM833) and biased by a VBE multiplier (2N2222 or 2N3904 or BC547).

The Darlingtons will be especially easy to drive thanks to their Darlington-ness (huge current gain). Just about anything you slap together, at any bias current, has a good chance of playing music.

However if you don't want to design a circuit your own self: never mind.
 
Richard, I suggest that for your first design you would be well advised to aim for simplicity and highest possible probability of success. If this calls for other transistor types that you are forced to purchase new, so be it. I recommend a conservative class-A or class-AB design involving an NPN Darlington (TIP122) and a PNP Darlington (TIP127), driven by an opamp (LM833) and biased by a VBE multiplier (2N2222 or 2N3904 or BC547).

The Darlingtons will be especially easy to drive thanks to their Darlington-ness (huge current gain). Just about anything you slap together, at any bias current, has a good chance of playing music.

However if you don't want to design a circuit your own self: never mind.

Thanks for going into some detail with the components - makes it a lot easier to get started and to know what to look for 🙂 I'm happy to have a go at building a circuit, so I'll take a look down that path.

If I were to go purely class A to begin with, which of the two Darlingtons would you suggest (from what I understand, using a PNP/NPN pair is class-B/class-AB territory?)

Thanks again for the reply.
 
I think the TIP122 and TIP127 are hard to beat, based on their low cost and wide availability. You could design a 10 watt, pure class A amplifier using one of each and get something that works very reliably (as long as you mount them on BIG heatsinks). Or you could design a 30 watt, class AB amplifier using one of each.

Very soon you will realize that the vast majority of the cost of building a power amplifier is in the power transformer, the heatsinks, and the power supply filter capacitors. None of which are made by Sanken. And the chassis if you get fancy and build an enclosed box. Numerous DIYers just construct their power amps on an aluminum "sheet pan" meant for restaurant kitchens, and to hell with boxes link
 
I think the TIP122 and TIP127 are hard to beat, based on their low cost and wide availability. You could design a 10 watt, pure class A amplifier using one of each and get something that works very reliably (as long as you mount them on BIG heatsinks). Or you could design a 30 watt, class AB amplifier using one of each.

Very soon you will realize that the vast majority of the cost of building a power amplifier is in the power transformer, the heatsinks, and the power supply filter capacitors. None of which are made by Sanken. And the chassis if you get fancy and build an enclosed box. Numerous DIYers just construct their power amps on an aluminum "sheet pan" meant for restaurant kitchens, and to hell with boxes link

Yeah, I noticed how quickly the cost ramped up once the power supply was brought into it... my first kit was a pair of Jaycar LM3876 kits (Jaycar is an electronics franchise here in Aus) which were relatively cheap ($32.95 each), but the toroid alone was over $100. The power supply board (also Jaycar) was an additional $34.95, and then enclosures... eugh.

Still, totally worth it. After firing up the Jaycar kits I could hardly wipe the smile off my face.
 
Yes I would use a Vbe multiplier. Be sure to put the adjustment potentiometer in the lower (base-to-emitter) leg. Why? Because when pots fail they fail with wiper-OPEN which means the variable resistance goes to infinity. If that's the lower leg of the Vbe multiplier then you get very tiny bias current. The reproduced music sounds bad but your transistors don't melt. If the upper leg of the VBE multiplier fails open, bias current goes to infinity. No es bueno sen~or.
 
Yes I would use a Vbe multiplier. Be sure to put the adjustment potentiometer in the lower (base-to-emitter) leg. Why? Because when pots fail they fail with wiper-OPEN which means the variable resistance goes to infinity. If that's the lower leg of the Vbe multiplier then you get very tiny bias current. The reproduced music sounds bad but your transistors don't melt. If the upper leg of the VBE multiplier fails open, bias current goes to infinity. No es bueno sen~or.

Wow - thanks for the heads up 🙂 at least I ordered quite a few of each part so if I cook them, I can learn the lesson and move on.

Thanks for the info, really helpful. I ended up with the 2N3904 and the parts have all arrived already, so looking forward to putting it together.
 
Status
Not open for further replies.