High Power Chip Amp

[star882]

Originally Posted by danielwritesbac

I'm not talking about a complex and dizzying time analyzing all of the possible options of a discrete preamp/predrive. What I'd like to see is a ready solution involving a chip-based bridged predrive running 8 output transistors per channel (bridge+parallel). Preferably, it can be "bugsplat" wired without a circuit board, in 20 minutes or less to result in a very powerful high fidelity amplifier. Design? Anyone?

8 transistors per channel is overkill. 4 is enough for H bridge unless you're really looking for high power. Using the Fairchild MOSFETs I mentioned earlier, using a B+ of 150v and a RMS current of 20A (into a 4 ohm load) means it is good for 1600w per channel! Not bad for just under $6 of transistors. (Just be prepared to burn out a lot during design, so have plenty of spares.)

And point to point becomes too messy for anything even remotely complex.

[a.wayne]

Originally Posted by danielwritesbac

I was going to try to argue with this and say that output transistors cost less for big amps. That idea failed somewhat.

However, parallel does make for cancellations, in this case used to quite good effect in making LM3886's fight off their own noises (that don't occur precisely at the same time). Unfortunately, if you were to cause enough cancellations then you also erase audio that is similar to the cancellations.

That is the point (beyond a carefully made pa150/bpa300) where output transistors start to get obviously more competitive on making clean audio.

The cancellations from paralleling can be reduced as well as the heat when the output devices are matched. Matched Lm3886's scale to pa200/bpa400. UNmatched output transistors scale to pa200/bpa400. Matched output transistors scale upwards of 1kw. Heatsink expense is decreased in proportion to cleaner audio from output transistors. And, the thermal interface area is considerably larger.

I'm not talking about a complex and dizzying time analyzing all of the possible options of a discrete preamp/predrive. What I'd like to see is a ready solution involving a chip-based bridged predrive running 8 output transistors per channel (bridge+parallel). Preferably, it can be "bugsplat" wired without a circuit board, in 20 minutes or less to result in a very powerful high fidelity amplifier. Design? Anyone?

I'm behind you on this , give it a shot, after all it's only 20 mins , to see if we could make a true voltage source amplfier ...

[Mark Allen]

It had better be a pretty good output stage to translate over the sound characteristics of the chip itself. Things usually go all to hell in the output stage. Transistors would need to be tightly matched to compete with same die transistors as in a chip, you'd also loose much of the built in protection of the chip itself where as the chip may shut down to protect things but the output stage may not. I'd still be interested in this sort of thing but I don't even have time to build the second amp and bridge them until early December...

Mark

[star882]

Design the entire signal path from scratch. Use high speed opamps to make the Delta-Sigma modulators (or use a FPGA for pure digital) and integrate overcurrent protection into the output drive stage.

I don't think matching is much of a problem.

[danielwritesbac]

Originally Posted by star882

Design the entire signal path from scratch. Use high speed opamps to make the Delta-Sigma modulators (or use a FPGA for pure digital) and integrate overcurrent protection into the output drive stage.

I don't think matching is much of a problem.

Some of that was greek to me, but I got the overcurrent protection part and that the matching won't be important (for indoor use).

For outdoor use matching is important in order to reduce reliance on output resistor, for a harder, stronger drive. But, a harder, stronger (harsher) drive is more-than unnecessary indoors (at point blank range) and we can just increase the output resistance slightly, instead of matching. This also applies to your PA150's--Use a tighter match on those LM3886's if the rig is for outdoor use or if the venue is quite large.

[danielwritesbac]

Originally Posted by a.wayne

I'm behind you on this , give it a shot, after all it's only 20 mins , to see if we could make a true voltage source amplifier ...

Um. . .per channel, not counting enclosure fitments. And, the first time though, I'm going to need help.

Can't do it tonight. I've either got to make a class-D or a hyper-efficient linear, run on 14.6 volts to facilitate a 3-way active system, with only one stubborn woofer. And, this is due before 10 a.m. tomorrow. There is also a 3a limit on amperage along with big expectations on performance and a very high heat environment that also involves occasional splashes of water. Egads!

[star882]

Originally Posted by danielwritesbac

Some of that was greek to me, but I got the overcurrent protection part and that the matching won't be important (for indoor use).

For outdoor use matching is important in order to reduce reliance on output resistor, for a harder, stronger drive. But, a harder, stronger (harsher) drive is more-than unnecessary indoors (at point blank range) and we can just increase the output resistance slightly, instead of matching. This also applies to your PA150's--Use a tighter match on those LM3886's if the rig is for outdoor use or if the venue is quite large.

I think some imbalance in the low and high side MOSFETs is acceptable. The two sides of the H bridge don't need much matching anyways. Just use the same part number MOSFETs.

Whether it's indoor or outdoor does not matter.

[danielwritesbac]

Originally Posted by star882

. . .

Whether it's indoor or outdoor does not matter.

I meant that there's a difference between "cohesive at long distance" or "laid back for indoors".
The sound you need to cover great distance may not be as pleasant at point blank range.
This relates to the output resistors--a closer match allows less resistance and the speaker will carry signal a longer distance (a bit harsher too).

[star882]

MOSFETs do not need any output resistors. They will self-balance. You don't even need to parallel them until you get to extremely high power levels. (The MOSFETs I mentioned earlier are good for some 1600w, and they're certainly not the biggest ones available.) It might be possible to use high speed IGBTs, which handle even higher power.

Check out Inverter modules for some of the largest (and most beautiful!) IGBTs you will probably ever see. Note that the ones that are paired up are handling a total of up to 50kW! (I wonder what Toyota does with the defective ones... They would make great necklaces...)

[AndrewT]

Originally Posted by star882

(The MOSFETs I mentioned earlier are good for some 1600w, ......... It might be possible to use high speed IGBTs, which handle even higher power.

which chipamp are you recommending that has these MOSFETs or IGBTs integrated into them?