Simple Chip Amp for P to P wiring

[danielwritesbac]

Over at Decibel Dungeon, there is this photo:
http://myweb.tiscali.co.uk/nuukspot/...roundlift2.jpg

The ground loop breaker needs the thickest cable of all; however, it may be a stranded cable.
One end of the ground loop breaker is for amplifier enclosure and the other end of ground loop breaker is for power star ground. Attach VERY securely. Its for both safety and pretty bass notes.

It shows the ground loop breaker schematic, but they just don't make bridge rectifiers that look like that. Here is a photo that illustrates the circuit:

[danielwritesbac]

Once I had used cable that was too thick at the small signal ground. . . the power supply ground is thicker yet. . . and by the time I got to the ground loop breaker it was impossible to solder the (thickest) cable without cooking the components. However, the simple solution was to double-up with cable of identical length, like this. . .

The purpose of this device is to ground the metal amplifier enclosure

Edit: An almost identical unit grounds the electronics in my work truck--its chassis was too rusty for reliable connections, and after adding this ground loop breaker, now it starts in a flash each morning and the battery has never run down.

[ratza]

Originally Posted by 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?

Actually... not really. At that current the voltage drop will be somewhere near 0.65 - 0.7V/diode. Regarding PS, it will work perfectly as you drawn it now, there's no need for other diodes. For groundings and alot of other aspects you should read the articles written by Mr Rod Elliot. You're building his amplifier, afterall.

[coldcathode]

Thanks all,

After going round and round and upside down, I think my original intent was a "minimalist" design. The "excercise" of looking at different options opened up my mind to the SS stuff versus the tube stuff.

During my research for this possible project I found that "Gainclones" seem to have a "cult" following. This appeals to me in some way, however it appears that most if not all of these implimentations seem to be focused on a much broader frequency range than what I intend to be reproducing.

So what I seem to have "settled on" is the following

PS has been determined, I intend to use "Chipamp.com" PS boards (ala Carlos FM) with the transformers that I have. 25VCT @ 2A. (2) per channel

The chassis will be of similar construction to my Tube amps which is a steel box chassis with wooden base. One chassis with 2 independant "monobloc" amps. I have a rather substantial aluminum heatsink from a "junkpile" find that will be "overkill" but aesthetically fits with the layout. The heatsink will be approximately 3" high and run the full width (10") across the back of the chassis.

The amplifiers themselves are now the question to be answered. While playing with this I have come up with an idea that might help me in accepting the SS circuitry. What if I were to lower the gain of a gainclone and drive it with a relatively high voltage signal from a tube? Given the very small physical size of the power supplies and amplifiers a 9x13" chassis would leave me enough room to use a medium mu triode as a driver for the amp.

Because the triode gain stage is inverting I would then want to use an inverting gainclone to correct the signal phase. I am hesitant to call this a "hybrid" amp because in essence it's just a "tube preamp" into solid state amps. But for purposes of the overall piece it is a "hybrid". I can place (2) more of the tranformers (I have 8) back to back to give me roughly 120VDC to run the tube.

Any thoughts on this? It really fits with my overall aesthetic and in essence does not add to the complexity since the "overture" circuit can be used AS-IS by coupling the plate output of the tube directly to the input of the chip amp. The low input impedance of the chip amp is OK with the tube since I am not asking it for a lot of gain anyway and a 6CG7 tube would have an output inpedance of about 7K anyway.

Here is a quick sketch of the circuit.

[danielwritesbac]

The minimum required gain is much different, depending on non-inverting mode or inverting mode, and inverting mode can go much lower. As far as I know, that's generally true of non-unity-stable op amps.

Um, I think that schematic needs an output cap for the subwoofer?

[AndrewT]

Hi Cold,
what are the values of the feedback resistors?

How did you arrive at the 8uF for the NFB capacitor?

Why no RF filters in front of the buffer/amplifier and again at the front of the power amp?

The 75k between the stages should be connected to Signal ground. Remove it from the common ground tying in Pin7 & pin8.
The 8uF should be connected to Signal ground. Remove it from the common ground.

You have omitted the HF decoupling on the power pins.
You have omitted the output Zobel.
You have omitted a high pass filter at the input (=DC blocking capacitor).

[danielwritesbac]

input cap at .5uF x minimum gain of 20 x 1.43 = minimum nfb cap size of 15uF to get bass like a jukebox, so double that cap size to 30uF (or larger!) for a decent result. That's a guess, but it should work.

On that schematic, the speaker needs a more significant ground. . . otherwise it is the opposite of a subwoofer amplifier.

[danielwritesbac]

OH! yes. Well, with the chipamp.com power supply it has four leads for power.
The V- and pg- of the power supply cable directly to the 100uF cap that you have pictured on the V- rail.
Likewise, the V+ and pg+ of the power supply cable directly to the 100uF cap that you have pictured on the V+ rail.

In this case, the common ground is directly between the pair of 100uF caps, as pictured in the schematic. The ground loop breaker and the speaker may be attached together at this point (midway, 0v, between the 100uF power caps at the amplifier board).

After hookup is complete, the speaker will have been grounded.

[danielwritesbac]

I think you need a resistor. Approximately 3.3 ohms will do it. One end attaches to common ground. The other end of the resistor is the new signal ground.

That way you can attach the signal ground components to a signal ground per AndrewT's specs at post 76

[danielwritesbac]

OMG! What was I thinking? I just bought a vacuum tube! Philips 6922