Your driver transistor (the 2N3906) drives the output transistors.
The output transistors drive the speaker.

1) What is the maximum peak current of the output transistors into the speaker?
2) What is the minimum peak input current to the output transistors? That is the maximum output current of the driver transistor.
3) What is the minimum peak input current to the driver transistor? That is the maximum output current from the differtential amplifier.

Then use Ohm's law to calculate resistor values.
Simple!

 

Ok thanks alot i'll do my best to fix up and repost my results later, Thanks.

 

Ok so I've removed the caps to make it direct coupled and lowered my resistor values for my driver and differential. What do you guys think about my circuit now? Is it better then what I first had?

 

Where are the simple calculations for your new amplifier?

1) Why is the load is only 3 ohms? Nobody makes a speaker that is 3 ohms. Most speakers are 8 ohms but some car speakers are 4 ohms and 2 ohms.
2) If the output is 18V peak then the peak output current into 3 ohms is 6A! But your darlington output transistors have a maximum allowed collector current of only 1.5A so they will blow up.
3) The tiny surface-mount darlington output transistors you selected will get much too hot since they do not have a metal tab to bolt to a heatsink.
4) If you use 6A or more power darlington transistors in a power package then their minimum hFE is about 800 so the peak current in the 2N3906 driver transistor should be 6A/800= 7.5mA. But the peak current in your driver transistor is way too high at 38V/300 ohms= 127mA! The poor little 2N3906 transistor will melt with average heating of 19V x 63mA= 1.2W!
5) The collector resistor for the 2N3906 driver transistor should be 38V/7.5mA= 5.1k, not 300 ohms.
6) Since the driver transistor should have an average current of 7.5mA/2= 3.75mA then its average input current should be 3.75mA/100= 38uA.
7) Then the collector current of one differential transistor should be an average of 190uA, not 650uA.
8) The SINGLEcollector resistor of the differential transistor should be 0.63V/190uA= 3.3k, not 5k.

Sorry, I do not have time to correct your schematic.

EDIT: the amplifier will probably oscillate at a high frequency because a frequency compensation capacitor is missing.

 

Hello

So I've made some changes to the circuit like audioguru aswell as adding a current source to the circuit. I have yet to add a compensating frequency cap to avoid oscillation because still doing some reading on it. So what do you guys think?

 

Your new schematic looks very good except:

1) The absolute maximum supply voltage for the 2N3906 driver transistor is the same as your total supply voltage. Select a transistor with a higher voltage rating.

2) The tiny surface-mount darlington output transistors cannot be cooled so they will fry.

3) There should be a capacitor to ground in series with R13 to avoid having the DC input offset voltage being amplified.

 

Unfortunatly the program I am using does have the two darlington transistors that i want such as the TIP141 and TIP147 which supports up to 10 amps, overkil... sure. and I will find a better transistor to replace the Q1 the pnp transistor aswell as adding a cap in series with R13.

Thanks audioguru your help is much appreciated.

 

Overkill is OK in a one of a kind project. It's only when the manufacturer wants to save 3 cents each on 100,000 products that you squeeze all the excess quality out.

 

Ok so I've tested my circuit using 15V for VCC and -15V for VEE and some of my transistors have blown up one being the 2n3906 and the TIP142 and TIP147 aswell and cant seem to figure out why? Any ideas...

Thanks

 

Assuming a wiring error caused the output to go to +13V or -13V then the current in the TIP142 or TIP147 is only 13V/6 ohms= 2.2A. but they are rated for 10A so they should be fine.

If one of the diodes is open or is connected backwards then the 2N3906 and the output darlington transistors will probably be destroyed.

Maybe most of the transistors have their pins connected wrong.

 

Well I made sure that all the transistor were pinned out properly but didnt pay much attention to the diodes ill have to take a look at that.

Thanks

 

Good news...

So I rebuilt the circuit making sure everything was in working order and the results were great. The amp does work properly and the sound was great, the THD was at 0.6% which still high for some people but I'm not to picky.

Thanks for everybody's help!

 

One last question...

So I was testing my circuit some more and noticed that only my TIP142 Darlington transistor was getting hot and not the TIP147. So I thook out the TIP142 and wasnt getting any sound at all. But when I only took out the TIP147 the music sounded fine without it. So I believe that my last stage is not class AB right?

 

I have not looked at your schematic, but as TIP142 and TIP147is a Darlington. Darlingtons will drop 0.6V CE, even in saturation. This means they tend to get hot as a matter of course.

 

I am glad to hear that you fixed it.
0.6% distortion is not bad for a simple amplifier.
You can make a few changes to reduce the distortion.

I calculated the DC voltages without a signal. Post your voltage measurements for us to see what is not working properly.

EDIT:
Notice that the power supply is positive 20V and negative 20V. Then the average DC voltage at the speaker is 0VDC.
If your power supply has only a single polarity then you are smoking your speaker with lots of continuous DC current.

 

Hello,

You can also replace the three diodes with a "rubber diode" and a couple of small emitter resistors to have less cross-over distortion:
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/part2/page3.html

Bertus

 

"Rubber diode", hee hee, hee.

 

Hello
Ok so I tested my circuit again and looked at a 10 KHz signal though an osciloscope and I got 28 Vpp which is good. So now I'm looking at speakers and not quite sure exactly what I should get. I found a pair on ebay but not sure if there the right speakers that I need. They take up to 400 watts max which is great even though I'm no where going to get near that.
Here's a link if you want to look at them, thanks.

http://www.ebay.ca/itm/200759145438?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649

 

Your link to very cheap little car radio speakers is a joke.
They are very small (my cheap TV had one that size) and the 400W peak per pair rating is 100W RMS each which is absolutely exaggerated! They might survive a low frequency playing 10W each for one or two minutes.

The tweeter in a speaker is designed to play high frequencies that are never continuous and are not frequent. So a continuous power rating for the little tweeter might be only 2W , not 400W.

The last amplifier schematic you posted produces about 22W into the 4 ohms of the cheap speaker but it will probably survive if you play music that peaks at 22W.
The average power of music playing 22W is 2.2W to 4.4W at low frequencies and is less for high frequencies.

The cheap little speakers will not produce enough bass response. They are rated for 50Hz without a spec that says how low level the sound will be. They might barely respond to 50Hz (-60dB) that can be measured with a sensitive instrument but not be heard.
Go to a store that sells car radio speakers. They probably will not have any that small. If they have them then listen to them play.

A good speaker is plus and minus 3dB (all frequencies sound almost at the same volume) from 30Hz to 22kHz.

A speaker needs an enclosure that has an adequate size and maybe with a certain size vent port. The cheap little speaker is probably made to fit into the door of a small car to play mid-range and high frequencies then larger rear speakers produce bass sounds.

 

So I'm back...

The amp works somewhat the way I want it to. Sounds good but I dont like the fact that theres 2 Vdc going to my speaker should I be concerned?