BJT diff amp

[sumochump]

We are told we must use an emitter coupled differential amplifier with diff. input and single ended output. We can use voltage sources of + and - 15 volts. Using only npn bjt's we must design a circuit that will have a gain of 10 v/v. The load is 1000Ω.

My question is why does my circuit clip at 1V no matter how I design it?
Is there another stage I can add to the output that will increase the voltage swing to say at least 8Vpp?

 

[crutschow]

How do you expect us to know what's wrong without a circuit diagram?

 

[sumochump]

That would be useful, sorry.

**broken link removed**

Here is my pspice coding:
BJT Diff. Amp

Vcc 7 0 15
Vee 8 0 -15
Vcm 1 0 0
Vd 101 0 sin(0 .001 100)
Ev+ 2 1 101 0 .5
Ev- 3 1 101 0 -.5
Is 4 8 1m

Q1 5 2 4 npnT
Q2 6 3 4 npnT

RC1 5 7 1k
RC2 6 7 1k

C1 6 9 .1m
Rl 9 0 1k

.model npnT npn(is=1e-14 bf=200 vaf=100)

.probe
.tran .01m 10m 0m .01m
.op
.end

here is a link to my circuit:
**broken link removed**

If I increase the input to say .1V the output is clipped at .2V

 

[sumochump]

I have submitted a diagram, and it is under review I guess, but in the mean time I think I realize what my problem is. I need to reduce the voltage dropped across the collector to emitter. Without looking at my circuit could you give some solutions to reduce Vce(DC value)?
(Sorry about the diagram not being up, it has link in it so it is under review)

 

[RCinFLA]

Are you using a tail current source or just a resistor? Is the 1000 ohm load in addition to collector resistor load? Is the gain of ten referenced differential input to differential output or differential input to single ended output?

For diff amp you are playing the net emitter resistance of the two diff devices against the total collector load resistance. You can play with just the BJT Re (= 0.026/Ie) or add resistor in each emitter leg to help swamp out the current dependent Re of the bjt devices. For such a low collector load resistance you cannot have too much additional emitter external resistance and must run enough current to get the bjt Re down.

 

[sumochump]

I would like to drive the amplifier with a current source if possible. And the gain is diff input to single ended output. For the small signal model my load would appear in parallel with Rc because I am using a coupling capacitor from the collector to the load.

 

[sumochump]

I have found my problem lies with the thermal voltage. The input cannot be larger than half of the thermal voltage. Is there a way to get around this?

 

[sumochump]

I figured it out, thanks for the help!!