bipolar transistors and pulse amplification

[fairchild1]

hello
i have been having trouble working with bipolar transistor. i have been taking pulses that at less then 1us long and amplifying them with transistors. how ever if i have a bipolar transistor as part of my amplifier whether it is my inputs are as part of another stage the pulses always are outputted as a >1us time pulse that has been amplified. i have used the bipolar transistor in may ways to try to eliminate the problem. but they all have the same outcome. the output is always a fast rise and a slow drop which would indicated current flowing easier one way. so i use pnp and a npn to see if the shape would change but it does not. can anyone give me advice.

[uart]

Sounds like a saturation problem. Show an example of the circuit you're using and I'll show you how to fix it.

[skeptic2]

Does your transistor saturate while amplifying the pulse? If so, try reducing the input level or gain to keep it out of saturation.

[fairchild1]

attached it my circuit. it is a standard inverting amplifier.

[skeptic2]

Can you tell us the source impedance, the load impedance and the high and low voltages of the pulse at the input?

[fairchild1]

as an input of the amplifier the charge pulses not voltage pulses so the duration of the pulse is the time it take for the charge to go from source to ground. in addition the voltage is dependent on the resistance to group so the higher the resistance the greater the voltage and greater the time of the pulse. these pulse are from a PMT.

[uart]

You need to prevent the transistor from saturating by limiting the base current.

You can use either of the two fixes shown. They both work by shunting away excess base current as the transistor nears saturation. They both require that there is a some amount of source resistance present so as to limit the available drive current, hence the extra resistor may or may not be needed depending on the impedance of whatever is currently driving the circuit.

[fairchild1]

i saw that before how ever i can not increase the impedance of the input other wise i would be increasing the the time of the pulse but i could use it as a second or thirds stage of the amplifier. so i will try it out

[uart]

i saw that before how ever i can not increase the impedance of the input other wise i would be increasing the the time of the pulse but i could use it as a second or thirds stage of the amplifier. so i will try it out

In that case then you probably don't need the series resistance so just add the clamp diodes, they will not greatly alter the input impedance of the circuit shown.

[fairchild1]

by any chance do you know what the diodes i should use

[uart]

I gave example part numbers on the diagrams (1n1418 and bat85)

[vk6kro]

I have attached a diagram of what should be happening with your amplifier.

Because it is an inverter, the input pulse duty cycle is inverted and you will get a negative-going pulse as your output if you have a positive-going pulse as your input.

So, the positive-going output pulse could be a lot wider than the positive-going input pulse. I wonder if this is what you are seeing?

Maybe you could modify this diagram to show what effect you are actually getting?

[Bob S]

as an input of the amplifier the charge pulses not voltage pulses so the duration of the pulse is the time it take for the charge to go from source to ground. in addition the voltage is dependent on the resistance to group so the higher the resistance the greater the voltage and greater the time of the pulse. these pulse are from a PMT.
Are you using these pulses directly into a discriminator for NIM (nuclear instrumentation module) logic. or going into a pulse height analyzer? I have used photomultiplier tubes (10 stage and 14 stage) a lot, and usually get very fast (~10 ns) risetimes and ~30 to 40 ns fall times into 50 ohms for fast events. Nai(Tl) (sodium iodide) is very slow. We rarely needed any amplifiers. I cannot remember whether anode or last dynode signals are faster.
Bob S