Emitter resistor values??

[Zero Cool]

Just out of curiousity, when designing an amp, how do you determin what values to make the emitter resistors? and what effects would there be if you altered those values within an amp (not that i would, its just for discussion).

For example. an amp i am working on that has 4 pairs of output devices and uses 5 watt 0.47 ohm emitter resistors. How critical is this value? what if i substituted 0.33 ohm, or 0.70 ohm etc.

What effects would there be if the value was changed. up & down???

again, this is for discussion only.... I found a source for some 0.50 ohm resistors which i figure is close enough to 0.47 to not have any ill effects. but this got me to wondering what effetcs various changes in values would have??

Zc

[Leolabs]

Well,i never put much attention to this issue,just stick to what available.But 0.1ohms is my favourite

[Mr Evil]

Higher emitter resistors add a little feedback, helping to keep the quiescent current constant with temperature, and in the case of parallel devices they help share current equally. If you're having difficulty setting a stable bias you may need to increase their value.

They also increase distortion and output impedance, so you want to use the lowest value you can get away with. For BJTs 0.1 is a good choice. I would avoid going below that, or above 0.47. With MOSFETs it's often possible to do without any at all.

[Leolabs]

Yap!MOSFETs make things much easier

[anatech]

If you parallel mosfets, you need source resistors to current share. Not unless you plan to match the darn transistors really very closely. Take it from me as I've had to do this. If you add a little feedback, the source resistance becomes less of a problem. I'm talking 6~12 dB, depending on the circuit.
I'm kinda partial to 0.22 ~ 0.33 ohms for BJT's. Again, depending on the circuit and voltages. The heatsink size also determines how stable an amp is for bias.
-Chris

[BrianL]

Doug Self did an extensive analysis of this in his EW/WW
articles and in his book(s). It also might be on his web site.
He arrived at 0.1 Ohms as the best compromise, as I recall.

[Mr Evil]

Originally posted by anatech
If you parallel mosfets, you need source resistors to current share...

Lateral MOSFETs will surely be okay without.

[X.G.]

although I had a idea of non emmitter resistor for power BJT a few years ago ,I had no more time to go forward....I found a Japanese very expensive power amp(IIRC,its type is TP ONE..I lost its trade mark) realized my idea a few months ago.

the method of the Japanese amp is applying current mirror in output BJTs( like my initial idea),one power BJT for constant current set and temperature compensation ,other power BJTs(5 pcs?) for SEPP output as mirror

I think that non emmitter resistor is the best,liking the wire is the best 'capacitor'

cheers

X.G.

[skmenon]

this depends on the ckt config. Is this an emitter follower or common emmitter ckt. the emitter resistor is usually chosen to set up the ckt current and a value of 0.1 is usually the choice.

[Zero Cool]

The Amps i have been repairing are Bass Guitar amps, so the value was probably chooses for stability under extreme conditions VS. Ultimate sound quality.

The value sounded High to me. I am used to seeing 0.1 ohm or 0.33 ohm commonly. so i wondered what effects the values would have.

How does an emitter resistor add feedback to the circuit? I understand how current sharing happens, but not feedback per se.


here is the circuit i have been working with.

So far i have 2 of the 3 spare amp channels up and running that i had laying around. I am waiting on parts for the third.


Zc