My task is to create a MOSFET switch that will provide pulses of at least 30V with rise time 5ns. I would be happy with 20ns i guess, but getting close to 5ns would make me proportionally happier.
Purpose of the switching circuit is to calibrate Tektronix P6015A probe. I can't find and i cannot afford the recommended calibrator, and so i am experimenting in the field of switches, where i am very much a beginner. For this, i need a steep square, the frequency is not that importatnt i guess. In reality the probe is used to measure a radar magnetron with 1/1000 duty cycle.
To drive a MOSFET (not yet chosen) i picked UCC2752 driver chip, because of the appealing 1ns rise time.
Now, this is a dual driver, and i will be driving one FET - is that still a good choice?
Can the input of UCC2752 be driven from a square wave generator i have at hand, which provides 0-10V/80ns rise time pulses?
Shall i feed just one input or connect A and B and drive them together? The same question stands for teh outputs A an B.
I imagine there is a lot of current from the output and so there resistors in series that limit the current - how shall i choose their size?
Would it be beneficial to use push-pull MOSFETS and put the probe between them?
Thank you.
Hello great to hear from CZ. What exactly is the scope and application?
The 10' P6015A is 75 MHz. A normal pulse gen can supply a 10..20V pulse but needed a 50 Ohm load for specified rise time.
TEK P6015A is tricky to compensate, with 5 interacting controls.
Only the LF comp and DC are needed for a stock probe that is just used on a new scope with different input cap. eg 15..22 pF.
The HF comp is not needed unless it has been repaired or fiddled with.
We used the 1kHz scope comp CAL out on out TEK 7904 which has a 40V output.
Be careful with the ground return lead and the tiny different sized trimmer pots on the comp box.
Bon Chance
Jon
Thanks Jon, all i have is an old pulse generator that is really slow (80ns), then i have R&S RTB2002 which has a weak calibration output and also very slow (80ns), R&S SMT06 which has only modulatuon pulse, which is faster than the previous (about 30ns leading edge), but i would like a proper thing. 5ns, max 20ns.
My scope is R&S RTB2002, which is very good and sufficient for the task. What i need is a calibration pulse, and that's what i am working on right now.
Hello: No fancy pulser or FETs needed.
You can use a 50..500 DC PS and resistopr, pulse with just asnall fast relay, mains >>diode>>realy coil
Runs like a vibratore.
See TEK P6015A mauanl p. 2-5...2-10 for CAL proceeudures
https://download.tek.com/manual/070822305.pdfBon chanc e
Jon