Routing pulses

[Kerrowman]

 

What is the best way to divert some 0.5-1.7kV inductively generated flyback pulses from one destination to another and then back again after a short period without using a relay, which would probably be too slow and inaccurate to set the duration of the ‘divert’? Can this be done with an FET/ IGBT or is there a better and more obvious way?

Thanks

[HarryA]

What types of short period are you interested in?  What divert times are you interested in?

An inductor  in series with the source and load would retard the current going out and when the magnetic field collapses there would be a back emf towards the source.

A length of coaxial cable would form a delay line. If not terminated to match the cable at the load a pulse would be returned towards the  source. see: analog delay line

 

[Kerrowman]

Hi Harry,

It will be a lot easier if I attach a two page design brief that explains what I’m trying to do in modifying my current setup. As you will see the duration of the ‘divert’ needs to be adjustable.

Julian

 

 

Design Brief (Combined pulses).pdf

[Kerrowman]

Hi Harry,

I appreciate that you probably don't have time to read design briefs like the one I sent so I have proposed a design that hopefully you can comment on.

To clarify, the circuit needs to switch an input from one output to another for a set duration (30-200ms) starting when a trigger signal arrives. The sequence of events is as follows - a trigger signal arrives to operate the relay which directs the ‘Input’ to a different output (from ‘Output 1’ to ‘Output 2’) and then, after a preset interval of between 30 and 200ms, directs the input back to ‘Output 1’.

For this, I am thinking of using a solid-state relay (LCC120) triggered by some form of timing circuit that will send the trigger signal through to flip the relay at the start of the interval but then send out a low after the set duration to flip the relay back i.e. the input back to output 1. (see attached)

If I use a decade chip like to 4060BE, then it will only go high after a set number of oscillations from its inbuilt oscillator and so won’t operate the relay at the start, so I need a timer that is enabled by the ‘tigger signal’ to operate the relay and then, after a preset interval, turns off the relay so it reverts to its original output.

Can you suggest an alternative timer method or any additional components to make this work?

Thank you

Jules

 

 

[HarryA]

Is the action  like this?  I got to many cd's

Are you comfortable using a micro-controller like an Arduino UNO?

time:   action                      switch 1    switch 2     
0:     hv pulses charge C1    at ca        at cd S2's
1:     C1 is charged to 24v   at cd         at cc
2:    hv pulses applied to V  at cb         at cc
3:    C1 drops to 17v            at cd         at cd S2's
4:    hv pulses charge C1     at ca        at cd  S2's

  cycle repeats

[Kerrowman]

 

Hi Harry,

That’s good of you to do a spice test and if I was starting from scratch then using a microprocessor would have been the way to go. However, I should have made it clear that I already have the cap discharge circuit in operation, that charges up a bank of capacitors to 24V then releases the stored energy as a high current pulse down to 17V. You can see the unit in the bottom left corner of the unit in the pic.

 

So I presently have a system of delivering the HV pulses to the batteries directly (with swapping) or the high current pulses from the ‘cap dump circuit’ but not mixed in the way I have described in my brief and post.

 

Therefore I’m looking to revise the main PCB with some additional elements to do the timed routing I described and not change the cap dump circuit. Hence my suggestion of an SSRelay with a timing circuit. Using an Arduino would take up too much space whereas an SSR and some timing component’s I can fit onto the present PCB format.

 

Incidentally I credited your help in the past, amongst others, in the Acknowledgments in my ‘replication manual’ for this device. This is part of the scientific process for results publication in the future so that any one can replicate the findings.

 

You can see the manual here:

 

https://www.dropbox.com/sh/td55b8675vvqtbg/AADzPSKMOI8q_YM1cFUT2T07a?dl=0

 

(I have temporarily removed the PCB files while I explore the circuit revision)

 

Assuming the SSR idea is sound then I’m trying to figure out how to set the relay with a short 4ms pulse (as per the earlier diagram) and then reset it after the adjustable period (30-200ms). Perhaps I need a latching relay instead but haven’t found a small DIP 8 format one.

 

Your thoughts are appreciated.

 

Julian