# High voltage to low voltage in ps/nanosecond time (Zener/opamp?)

So I have a signal that goes from 0 to ~800volts in 30ns. I would like to use this signal to trigger a camera (input impedance of 50ohm) which takes a max of 10 volts and requires the rise time to be 1ns for this 10 volts. I tried a 10 volt zener diode which works... but delays the signal a lot (not sure why) so I get 10 volts in 8ns instead of 1ns. [People said zeners are made for DC and not so much for high frequencies, so maybe thats why]

I then thought of using attenuators to make the signal go from 0 to 10 volts (instead of 800V) in 30ns then use an opamp as a comparator with a 10volt DC battery connected to +Vs and Ref, Vin-, -Vs connected to the negative side of the battery. (and ofc the main signal going to Vin+) Since different opamps got different operating frequencies and slew rates I got this one: THS3491IDDAT Which got a 320 MHz operating frequency and 8 kV/us slew rate

However, I was getting weird results that I wasnt able to understand and shortly after multiple legs of the op amp broke, so I wasn't able to capture the output signal

Here is the input signal which is attenuated using 38dB (div 80) [time scale is 10ns/div and voltage scale is 10v/div

Was I doing something wrong? or is there an easier circuit to use? Any recommendation for other opamps that are easy to handle? (this one was tiny)

The source is a Rogowski coil with a resistance of 4ohm and inductance of 1.4e-7 H. (1 turn coil).

• What is the output impedance of your source? Which Zener did you use, and what value of resistors? What is the minimum your camera can trigger on? Nov 13 '18 at 23:33
• You have a signal source that provides you with $\approx 30\frac{\text{kV}}{\mu\text{s}}$? What's the output impedance of your signal?
– jonk
Nov 13 '18 at 23:34
• @Ale..chenski The source is a Rogowski coil which got a resistance of 4ohms but not sure about its impedance. (calculated the inducatance to be about 1.4e-7 H) I used 1N5925B for the zener diode which is 10volts and 3 watt. the camera can trigger with 5 volts and max 10volt. Nov 14 '18 at 0:02
• @jonk well its not 30kv/us per say, it only goes to 800volts in 30ns then goes to 0 after. Not entirly sure about the source impedance, its a Rogowski coil with 1 turn and about 4ohms resistance (calculated induactance of 1.4e-7 H) Nov 14 '18 at 0:04
• First rule about using op amps as comparators: Don't use op amps as comparators. Get a real (very fast) comparator. Second, do not try to use floating, dead bug construction for nsec response times. You need a proper ground plane and good construction techniques. Nov 14 '18 at 0:08

Try this

simulate this circuit – Schematic created using CircuitLab

• +1. To add, I have some feelings that the camera triggers from a regular 5-V TTL signal. Which is 2.4 V. Nov 15 '18 at 4:59

Sounds like what you need is a "capacitance speed-up."

Take a look at the following. Blue is the source pulse / 100 (so it fits on the scale.) Red is what a resistor + Zener would output. Green is what is output if a "capacitance speed-up" is used.

Note that the 1N755 is only a 500mW device, and the peak pulse from C1 into it approaches 650mW. Not really a problem for "a few" photos, but it couldn't be run like this continuously. Also note that the speed-up creates a nearly -1v tail... careful.

You obviously have some high cost instruments. I wouldn't compromise their life span with experimental add-on circuits. Get an industrially built unit. High Speed Pulse Generators (search for them) are available as modules and as laboratory instruments.

If you want to build one, you can use Step Recovery Diodes to generate fast rising voltages. Here's one principle:

https://www.overunityresearch.com/index.php?PHPSESSID=bgbfjuvodm8ekjgusgka0r4ed6&topic=1556.msg26264#msg26264

The idea is to use the sudden reverse conductivity stopping of a step recovery diode to break the current in an inductor and thus genrating a pulse.