Simple Chip Amp for P to P wiring

[AndrewT]

Originally Posted by coldcathode

AND the CAPS were a little backwards too!! DUH

the diodes in the lower (-ve) half are still upside down.

Two 50VA transformers are not equivalent to one 100VA transformer.

25Vac is the rated AC voltage when the resistive load draws the full specification output current.
When the transformer is off load it supplies a higher voltage.
A 50VA EI could be ~20% higher voltage.
A 100VA EI could be ~15% higher.
A 100VA toroid could be ~10% higher.

Your 50VA EI will give ~38.5Vdc when mains is at low tolerance and ~45.5Vdc when mains is at maximum tolerance if the transformer regulation is 20%.

A 160VA toroid will give ~34.3Vdc with mains low and ~40.6Vdc when mains is high and transformer regulation is 7%

Note that your 25Vac transformer cannot use 35V smoothing capacitors. It must use 50V capacitors.

Finally, a 160VA transformer can be used to power between 80W of total power and 160W of total power, i.e. 50W+50W of stereo power would be a good target.

[danielwritesbac]

This can be handy:

[danielwritesbac]

Originally Posted by coldcathode

Assuming I work out the PS issues.

I revised the schematic on that project removing the bridging and SIM optional components.

Couple of questions

#1 Feedback resistors and caps are they R5,R4 and C3??

These determine overall gain and LF cutoff correct?

#2 How to increase Input Impedance (to about 50K min) and assuming I have no DC from the source and the source is cap coupled to prevent DC back into it can I remove C1? and what is the purpose of C2?

#3 Notes say that C4 determines HF responce this will be a SUB amp so what components determine that POLE so I can lower it?


Any help to a "Bottlehead" would be greatly appreciated!!

C2 blocks high frequencies that are above the audio band.

You can use a buffer at the input, if you like.

The D1 led is is a clear led that shines a green light when it is on.

There are at least a pair of 220uF missing from the power circuit of that amplifier board schematic.
It goes like +220uF(0v)220uF- (much like a 2-cell flashlight)
Normally those would each be paralleled with 100nF ceramic or polypropylene although a smaller value polyester is workable.
You can also use a 2 uF 250v plastic cap (the same type sold for speaker crossovers) from V+ to V- at the power connections to the amplifier board.

Capacitors located between speaker groundpoint and rectifier are exposed to a lot of voltage and noise, so those caps need to have increased tolerances in order to prevent explosion. 50v to 63v rating will do nicely, but 35v caps may explode or leak. Remember, what returns from the speaker is powerful AC audio.

AndrewT's grounding methods will help your safety, but will also dramatically help the bass of your subwoofer amp.

[ratza]

D1 has another purpose: it is a constant current source for the input differential. I would get rid of it and replace it with two 1N4148 diodes. The reason for this is that you can hardly find two LEDs with the same voltage drop.

[danielwritesbac]

Originally Posted by ratza

D1 has another purpose: it is a constant current source for the input differential. I would get rid of it and replace it with two 1N4148 diodes. The reason for this is that you can hardly find two LEDs with the same voltage drop.

The pair of diodes idea should be seemly in the subwoofer amp project.

Does this require any compensation to the values of other components?

[ratza]

No, it doesn't, everything else will remain the same.

[coldcathode]

Ok,

I want to put the PS "to bed". Here I have revised the schematic correcting all the polarity (I think).

Note: This is the supply for ONE channel other channel with be ENTIRELY separate built as two MONO blocks essentially. ie; two more trannies etc.

I realize that the supply voltage will vary +/- maybe 4 volts?

Only questions on the PS are the following.

On the schematic I marked two points "A" and "B". Can or Should I place an additional diode at each point biased so as to "isolate" the two "halves" of the supply? ie; reverse bias the + connection and forward bias the - one?

Excuse me if that is a "dumb" question it just seems to me that doing so would supply 2A's of Positive Rail and 2A's of Negative Rail rather than 4 amps across both? Could just be symantics just asking.

The amp schematic calls for 100nf caps at the inputs so those are left out of the power supply schematic.

All caps will be ATLEAST 50V

Grounding?

The "earth" connection of the power plug will be grounded to the chassis. What configuration should I use for the remaining ground points?

The amp schematic has four ground points.
Obviously I want to "star ground" the junction of (C2,R2 & R8) with C3
Should also C+ be "star grounded" with the junction of C-,C7, speaker earth and 0V?
If so then how should those be returned, joined etc?
Any advantage to using a resistor to chassis earth to "raise" ground above the chassis earth?

[coldcathode]

As far as D1 is concerned. It appears that it is used to Bias (about 2V?) and be a CCS of 1.5mA? So the diode idea sounds good. I can use a lighter rocker switch to know if the amp is on. Only question about the 1N4148 is 1.5mA enough to have a good, solid 2V bias?

[danielwritesbac]

Monoblocs?
Okay. At the power supply board, there needs to be cable between the 4700uF+4700uF section and the polyester 2uF. . . other way to say the same thing is that, for monoblocs, the 2uF goes onto the amplifier board directly at the attachment point of a short-length cable to the power supply.

At the amplifier board, C+ and C- are shown 100nF. That is insufficient. It needs to be at least 100nF//220uF (parallel 100nF with 220uF) for both C+ and C-.

C+ and C- ground points is the attachment point for the 0v (power supply ground) cable. If these two caps are physically separate, then you can use a 16ga solid core wire to connect their two groundpoints. A "jumper" is the term. Tap the centerpoint measure of that very short bit of solid core wire as the attachment point for the power supply ground cable. For loud dynamics, you can also ground the output RC and the speaker at this point as well.

[danielwritesbac]

If you're using the above post as the wiring scheme, you can add a 3.3R resistor to the groundpoint of C3 (ground lift) and cable that via 20ga solid core over to the ground point of R2. Next, cable R2, via 20ga solid core over to the short chunk of 16ga (power star ground) where everything else is grounded.