1
\$\begingroup\$

For example, in a datasheet I am looking at, the critical rate of rise of off-state voltage is given as $$2000V/{\mu s}.$$

If I have, say, five of these thyristors in series, would the total critical rate of rise of off-state voltage then be $$5\cdot 2000 V/{\mu s}?$$

Thanks!

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Doubtful. The parasitic capacitance of each may be different leading to dv/dt that's different from 1/5 of the total applied dv/dt. There's probably some safe factor like maybe 4*spec based on worst case process differences+margin for physical parasitic differences but I don't know what it is. \$\endgroup\$
    – John D
    Jun 6, 2019 at 15:31

1 Answer 1

Reset to default
1
\$\begingroup\$

Note that I'm not an expert in thyristors specifically, which is why this answer sounds general.

The general rule for stacking things like this is that it's your responsibility to make sure that the voltage is balanced (or current, if it's a bunch of devices in parallel). This means that you either need to look for the manufacturer's recommendations, that you need to read their data sheets with a careful and critical eye and do some analysis, and possibly that you need to add components to balance the string (like a cap, resistor, or both in parallel with each device to help them stay balanced, at the cost of worse parasitics in the stack).

\$\endgroup\$
3
  • \$\begingroup\$ Hi Tim, thanks for the answer. So what you mean, then, is that if the thyristors can stay strictly balanced (which, if a large dv/dt is required, may be difficult) then the max dv/dt will scale linearly with the number of thyristors? \$\endgroup\$
    – Lerbi
    Jun 7, 2019 at 15:10
  • \$\begingroup\$ Yes. But it's almost certain that you can't just blindly stack any old set of thyristors and have it work out right. This is, I suspect, worth at least a chapter in a book on using thyristors in HV power electronics. (Note that I'm not an expert in thyristors specifically -- I'm going to edit my answer with that, lest I lead someone astray). \$\endgroup\$
    – TimWescott
    Jun 7, 2019 at 16:43
  • \$\begingroup\$ @Lerbi scaling linearly with the number of thyristors only work in an ideal world where all components are identical. In real life things are made in batches with varying parameters. You should try doing some monte carlo simulations with capacitors in series with varying capacitance (like 20% tolerance), or resistors, and see how different voltages you get across each of them. The important aspect that I want you to understand are the tolerances, the nonlinear real life. \$\endgroup\$
    – Harry Svensson
    Jun 7, 2019 at 17:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.