I rather like Marley's design since it is simple, and uses a differential input for negative feedback to stabilize the output DC bias point and gain, and reduce distortion.
Is there some advantage to Mr Al's circuit I don't see?

 

I rather like Marley's design since it is simple, and uses a differential input for negative feedback to stabilize the output DC bias point and gain, and reduce distortion.
Is there some advantage to Mr Al's circuit I don't see?

The circuit I posted also uses negative feedback, a differential pair is not necessary for that, though it is likely better.

 

You can trust @crutschow 's circuit.
If you want better THD, then improvements are just modification of that circuit.
Change R2 for a constant circuit circuit. Add a current mirror load for Q1 and Q2.

Hi Ian, when you put a constant current in place of R2 then the emitter resistance re = 26 / Ie becomes more constant and the amplifier more linear. Is the current mirror only to reinforce this, making re more constant along both legs of the LTP or does it also reduce the matching requirements of Q1 and Q2?

 

Hi Ian, when you put a constant current in place of R2 then the emitter resistance re = 26 / Ie becomes more constant and the amplifier more linear. Is the current mirror only to reinforce this, making re more constant along both legs of the LTP or does it also reduce the matching requirements of Q1 and Q2?

It increases the gain by making the load of the ltp transistors a constant current source and it also keeps the ltp in balance where its distortion will naturally be lower.
It's also one fewer resistance value to calculate!
In days gone by, manufacturers would put the internal circuitry of their op-amps in the datasheet. The input circuit was always loaded with a current mirror.

 

Hi
I am getting 55.3 mA.
RS

Hi again,

If you and @crutschow are getting a higher zero signal current in the output transistors then that tells me that the transistor models are different, and that tells me that the circuit bias would need to be adjusted as it was used. The original design probably teetered on the exact transistor specs, which means that one probably would have needed adjustment also. Some of these amplifiers do require that kind of adjustment, it depends on the main bias method. You can see that one transistor with that one resistor I changed to a lower value, and that probably isn't the best way to get this biased. We can look into the theory behind that method (transistor + single resistor) and maybe substitute with another method if any of you think this circuit is worthwhile.

 

Hi again,

If you and @crutschow are getting a higher zero signal current in the output transistors then that tells me that the transistor models are different, and that tells me that the circuit bias would need to be adjusted as it was used. The original design probably teetered on the exact transistor specs, which means that one probably would have needed adjustment also. Some of these amplifiers do require that kind of adjustment, it depends on the main bias method. You can see that one transistor with that one resistor I changed to a lower value, and that probably isn't the best way to get this biased. We can look into the theory behind that method (transistor + single resistor) and maybe substitute with another method if any of you think this circuit is worthwhile.

Can anyone recommend a substitute for the 2N2907? Since I can't get my hands on this device
should I use a replacment for the 2N2222 as well since these are complementary?
Thanks
RS

 

Can anyone recommend a substitute for the 2N2907? Since I can't get my hands on this device
should I use a replacment for the 2N2222 as well since these are complementary?
Thanks
RS

In an audio amplifier, you can use any small signal transistor that has sufficient voltage rating.
For output transistors, you can nearly always use dull, slow, old TIP31/32; but you can make a better amplifier with faster transistors or MOSFETs.

 

Can anyone recommend a substitute for the 2N2907?

A Google search listed several equivalents, such as the 2N2905, 2N4405, NTE129.

should I use a replacment for the 2N2222 as well since these are complementary?

Not necessary.

 

A Google search listed several equivalents, such as the 2N2905, 2N4405, NTE129.
Not necessary.

Actually, now that I look at the circuit in post#33 9 "LilTiger" I need easily found
replacements for THE

A Google search listed several equivalents, such as the 2N2905, 2N4405, NTE129.
Not necessary.

Thanks!
RS

 

The old 2N2907 was in a metal case like the original old 2N2222. Their max current was only 600mA.
I always use modern 2N4401 NPN and 2N4403 PNP that also have a 600mA max current and can dissipate more heat in their modern small plastic cases.

 

The old 2N2907 was in a metal case like the original old 2N2222. Their max current was only 600mA.
I always use modern 2N4401 NPN and 2N4403 PNP that also have a 600mA max current and can dissipate more heat in their modern small plastic cases.

Thanks Audioguru again. Those work much better in my sim than the 2N2222 and the 2N2907.
RS

 

Thanks Audioguru again. Those work much better in my sim than the 2N2222 and the 2N2907.
RS

Before I start bread boarding some circuits I need to know how to determine the required voltage
rating for capacitors and the required wattage rating for resistors. I assume that we apply a margin
of safety. Must we look at both ac and dc factors when determining these ratings?

RS

 

Before I start bread boarding some circuits I need to know how to determine the required voltage
rating for capacitors and the required wattage rating for resistors. I assume that we apply a margin
of safety. Must we look at both ac and dc factors when determining these ratings?

RS

Hi,

Power dissipation is P=R*i^2, but that's only with the proper surface area.
Power radiated is related to the entire surface area of the resistor. This may include the bottom or not if the bottom is right on the board, but then if the board is a good thermal conductor that helps too on the bottom.

It is somewhat customary to use a resistor that is twice the rated power as R*i^2, but i find that you should go as high as you can so you maintain a cooler resistor temperature, especially if it is in an enclosed area, but you may need venting also, and in extreme cases, a fan as well.

Ceramic power resistors seem to work pretty well for higher power circuit requirements. These will get super hot if run at their rated power too though. Even with one-half power the resistor could get very hot.
The power resistor will most likely survive, but you don't really want a space heater in the same box as your other parts like transistors, diodes, caps, etc.
What this all means is you have to do a thermal analysis, which would come after the circuit is built, and after you've chosen some reasonable power ratings for the resistors. Check and see if any parts get much hotter than you would like to see and make them bigger. This could be when you do a breadboard for example, possibly with a mock up case for everything, or the actual case to be used which is best.
You'll also have to choose heat sinks for the output transistors most likely because of the power levels involved. You do need venting for these as a minimum.
This is why class D started to become so popular. When I first got into audio work long ago there was no class D that I can remember. If we wanted "class D" we'd have to invent it ourselves. The only class D I was aware of at the time was pure sine voltage converters, which I did a lot of work with, and that was only for a fixed or a little variable AC frequency output, not the entire audio spectrum.

 

Before I start bread boarding some circuits I need to know how to determine the required voltage
rating for capacitors and the required wattage rating for resistors. I assume that we apply a margin
of safety. Must we look at both ac and dc factors when determining these ratings?

RS

You can't go far wrong with specifying the voltage as being the supply voltage.
Most of the resistors are in class-A parts of the circuit (ie long-tailed pair and voltage amplifier stages) so the power dissipated is the same with a signal as without, so they can be calculated on the DC conditions.
The only exceptions are the bass-emitter resistors of the power transistors, which will have almost 600mV across them at most times.

 

A solderless breadboard is not suitable for building an audio amplifier:
1) The contacts cannot handle the high current.
2) The many rows of contacts and wires all over the place add coupling capacitance that cause oscillation and pickup interference.

Solder the circuit compactly on a small pcb or stripboard instead.

 

Hi,

Yeah, they make PC boards with multiple dip patterns already etched. You do have to jumper everything, but it works out pretty well and everything is soldered. They make boards with other etched patterns also.

When I do this I bring out connections to the edges of the board. I'll show a pic when I get a chance of something I built up sort of recently (earlier this year). Many people use connectors that have pins that fit right into the 0.1 inch spaced etch patterns near the edges of the board. All this stuff is widely available, online.

It's also easy to drill out the holes to larger diameters for thicker component leads.

 

Hi,

Yeah, they make PC boards with multiple dip patterns already etched. You do have to jumper everything, but it works out pretty well and everything is soldered. They make boards with other etched patterns also.

When I do this I bring out connections to the edges of the board. I'll show a pic when I get a chance of something I built up sort of recently (earlier this year). Many people use connectors that have pins that fit right into the 0.1 inch spaced etch patterns near the edges of the board. All this stuff is widely available, online.

It's also easy to drill out the holes to larger diameters for thicker component leads.

Thanks to all!
When considering the voltages for caps do we look at the dc volts only or the total volts (ac + dc)?
When considering the wattage rating for resistors do we look at total volts and total current or just dc in order to determine wattage?
RS

 

Thanks to all!
When considering the voltages for caps do we look at the dc volts only or the total volts (ac + dc)?
When considering the wattage rating for resistors do we look at total volts and total current or just dc in order to determine wattage?
RS

See post #74

 

See post #74

I read that but ac passes through the capacitor. I am a novice, I have to ask these questions.

 

Ian’s point is that no resistor or capacitor is ever going to see a voltage larger than the supply voltage, so rating them for that voltage is safe.