Too big C means overheating the transformer. In a short time a huge current will flow, so the internal R of the transformer will grow hot.
We have the formula dU/dt = I/C
With 4A average current, 10ms for double rectifier 50Hz, if you accept 1V ripple, which is not the minimum!,
C = 4A . 10 ms /1V = 40mF = 40.000 uF Do I calculate well?
So I agree that 10mF is not too much. It gives about 4V ripple at full load/power, which means that you perhaps get more distortion.
Alternative is a high frequent PSU (like in your desktop) or a voltage regulator. With a smaller C, you get less ripple.
Some handbooks tell you must buy your transformer 150% the average current value, so 6A in this case.
As stated in this thread: the PSU is the most expensive (and perhaps underestimated) part of a modern amp. It costs about EUR 60, where the chipamp is only EUR 4 to 8.
We have the formula dU/dt = I/C
With 4A average current, 10ms for double rectifier 50Hz, if you accept 1V ripple, which is not the minimum!,
C = 4A . 10 ms /1V = 40mF = 40.000 uF Do I calculate well?
So I agree that 10mF is not too much. It gives about 4V ripple at full load/power, which means that you perhaps get more distortion.
Alternative is a high frequent PSU (like in your desktop) or a voltage regulator. With a smaller C, you get less ripple.
Some handbooks tell you must buy your transformer 150% the average current value, so 6A in this case.
As stated in this thread: the PSU is the most expensive (and perhaps underestimated) part of a modern amp. It costs about EUR 60, where the chipamp is only EUR 4 to 8.
thx dr frost dk 
It seems ok at the moment, though it needs definitely tidying.
The cpu coolers are indeed some nice pieces to work with and also with fans for some extra cooling.
I used my old pc case, which just got left over. It has got a couple of temp sensors and a small display. An old pc psu for it, and the fans.
It seems ok at the moment, though it needs definitely tidying.
The cpu coolers are indeed some nice pieces to work with and also with fans for some extra cooling.
I used my old pc case, which just got left over. It has got a couple of temp sensors and a small display. An old pc psu for it, and the fans.
JaanusJ i did however notice that the wires to transistors (collector & emitter are quite thin), you should use thicker wire to get all AMPS out of the transistors, and of course as short wire as possible.
And as i said im all for using what you have laying around, that's why i used the ATX-Power case as it's size was good for the Transformer i use.
Next AMP will be a 4x TDA7293 with a SMPS and also in an ATX-Power supply case, just a different layout as the SMPS takes up very little space.
When i get on the project ill make a new thread,
And as i said im all for using what you have laying around, that's why i used the ATX-Power case as it's size was good for the Transformer i use.
Next AMP will be a 4x TDA7293 with a SMPS and also in an ATX-Power supply case, just a different layout as the SMPS takes up very little space.
When i get on the project ill make a new thread
,That's what diy is about
What gauge should the wires be?
Looking forward to this new project of yours
what gauge are you using now?, as you can see in my amp i also use the output from directly from emitter to resistor to out
The wires used in the ATX case are bigger then this, this is just to show the setup of the direct and VERY short wires
An externally hosted image should be here but it was not working when we last tested it.
The wires used in the ATX case are bigger then this, this is just to show the setup of the direct and VERY short wires
Looks nice and especially the wires. In so many gear they have expensive parts but cheapass cables with just one core , often too long, so twisted around..., i'm busy with a technics mk-5 turntable(1200mk5) and though it's not old , it has the open variable resistors and shitty unshielded cables and offcourse the connector, like in active speakers , which are crap . Don't replace them for one core silver ones, cause silver doesn't like to be bend at all, just copper, not solid which has way longer molecules.
Yes that's the problem you'll get when you turn it on and the caps are fully onloaded, but , 1,5x times the ouput value is the absolute minimum, also soundwise you better take then times 2. I've read an article with the figures and they were right. But let's not forget the quality between transformers. I would never go to the limit, since a somewhat higher power transformer doesn't cause a lot more and things depends offcourse between schematics , since the full load ain't coolable plus the sound is way better with max 60 watts, just use them parallel or together with a toshiba 5200 and the pnp version.
Most nice schematics also use that option and biasing them is a bit harder , plus they use a transformer for the tda on a lower voltage , but compared to the price of a good sub, an extra transformer ain't expensive. Just search on google images
Most nice schematics also use that option and biasing them is a bit harder , plus they use a transformer for the tda on a lower voltage , but compared to the price of a good sub, an extra transformer ain't expensive. Just search on google images
Its been a while, just want to report that the LM3886+5200/1943 amp at +-43.6V rails i doing just fine, good gain,good tone the chip is runnin just warm, the transistors a bit warmer-without the fan- sorry no pics, lost mycam.
the schem is the same gclone ptp with the 22k-1k-47uf fb string, 2x470uF at the chip, PS has single bridge rectifier then 2x2200uF parallel each polarity, then 3x1ohm parrallel each polarity then 2x2200uF parallel each pol. then +-V out and PS star ground.it sounds really good direct to rin or thru a 2.2 or 3.3 film cap.
still working well?
what a cool and efficient idea, it is nice to know that you guys tried it and sounds perfectly.
I once had a similar thought and too dumb to think about how to fit them though,
I also have some questions, will this very design work to other chip amps like 7815, AN7161, TA7205, etc?
and will it matched to other power transistors like TIP, 2N3055/MJ2955, or even on sanken 2SA1216/2SC2922? since I have several pairs of those Trs
I have scrolled over from the first to this last page I barely found you pals discussed about the last transistors pair I mentioned above?
sorry for my bad english, thanks
regards, bruariwan
I once had a similar thought and too dumb to think about how to fit them though,
I also have some questions, will this very design work to other chip amps like 7815, AN7161, TA7205, etc?
and will it matched to other power transistors like TIP, 2N3055/MJ2955, or even on sanken 2SA1216/2SC2922? since I have several pairs of those Trs
I have scrolled over from the first to this last page I barely found you pals discussed about the last transistors pair I mentioned above?
sorry for my bad english, thanks
regards, bruariwan
merry CHRISTmass to you all 
Hehe i suffered that kind of failure once also, but in my case it killed the TDA chip, i did not short the AMP but was testing some speakers and have bad connections that had the speakers connect and disconnect at high power, and this is not good for ANY amp as the B-EMF is hitting VERY hard doing this.