Hello,
I am trying to build an electrical stimulator (not for being used in living things) for a biomedical project, some of the parameters to control are waveform (pulse,triangular), voltage around 100v, waveform period (500 us-250ms) and frequency (1Hz-30Hz)

The approach i´m trying is to use a "high voltage" PWM signal + RC filtering. Now i realize that obtaining a "high voltage" PWM signal is not that easy, even though i´m using a pulse transformer.

To avoid DC bias in the pulse transformer i tried this:

Every stage goes as planned until the rectifer bridge before the RC filtering. I think that maybe there is
a more elegant approach for the task, so i would like to know and suggestions/ideas on the project.

I really appreciate any help you can provide.

 

What's wrong with using a boost regulator to generate 100Vdc and then just using a high voltage FET to switch the output to the filter? Are you trying to generate just a voltage at low current or at some high current level?

 

I am trying to build an electrical stimulator (not for being used in living things) for a biomedical project

Inasmuch as 'biomedical' translates nicely to 'life+health', your assertion that the project is not for application to organisms is a tad incongruous -- IOW, while I hesitate to suggest that the original post is 'oxymoronic', the fact remains... --- Thus it seems that, dependent upon the organism{s} involved, "VI AAC III" may prohibit substantive response to your inquiry -- Elucidation, however, may improve my perspective...

With marked jocularity
HP

 

In other words, what HP is saying is that discussion of your project is potentially in violation of AAC rules as stated in the Terms of Service.

 

Inasmuch as 'biomedical' translates nicely to 'life+health', your assertion that the project is not for application to organisms is a tad incongruous -- IOW, while I hesitate to suggest that the original post is 'oxymoronic', the fact remains... --- Thus it seems that, dependent upon the organism{s} involved, "VI AAC III" may prohibit substantive response to your inquiry -- Elucidation, however, may improve my perspective...

With marked jocularity
HP

Yes you are right, the post is somewhat 'oxymoronic', but that was not the intention.
I'm a biomedical engineer student, and this project is just for learning purposes.
The idea of using a transformer came from the university professor, working with the project
i realized that maybe there was a more elegant solution. I just wanted ideas, suggestions to make a
better project and to learn different approaches so maybe one day i'll be able to build a real one.

 

A pulse transformer is better at delivering things with a rectangular pulse shape than other transformers, but don't think that its output will have nice square corners, like a clock source only bigger. As suggested, a better way is a 100 VDC source powering a linear wideband amplifier with pulse, triangle, whatever as its input. You'll have much better control over the output amplitude and load impedance matching.

ak

 

Are you trying to generate just a voltage at low current or at some high current level?

Low current, max. 40-50 mA throug a ~220 Ohms load.
I search for dc-dc converters and found R24-100B. As for the FET , is there any particular suggestions?.
(I have mainly worked with BJT transitors so i don't claim to be an expert).

 

50 mA across 220 ohms is only 11 V. More to the story?

ak

 

50 mA across 220 ohms is only 11 V. More to the story?

ak

My mistake, it should said ~2,000 ohms. It's the analogous of skin impedance, that in not way is 220 Ohms!