Can’t find thyristor specs: how big of a problem is it?

[Michelle]

This is a bit of an odd question but here it goes.

Basically I’m going to build a few coil guns and one of my plans is to built a multi phase linear induction launcher.

To begin with I’m learning more about electronics and switching so that I hopefully don’t kill myself; and to that end, I found a “good deal” on three thyristors for $50 that I’m going to use just to have something to start off with and practice.

Said thyristors are BISCR6601818 which are 660v 1800a units that I think are used for motor starting.

Generally I wouldn’t buy something like this without more information but like I said this is just to get a working contraption going before I design a potentially more optimized circuit. These are normally $400-800 and similar thyristors are also expensive which is why I want some beater units for practice.

What I’m wondering is how limiting will it be to not have the full spec sheet as far as simulating a design (another goal) or making a first pass at the launcher.

Are there assumptions I can make about them? Does it matter very much if I’m not pushing the limits of their voltage and current? Can I derive important info by experimenting with an oscilloscope for example?

I realize switching speed and di/dt is an issue however I’m beginning with a small projectile and readily available electrolytic capacitors which means a maximum of 600-700v anyhow. I don’t see approaching the current limit or even close.

I’m also aware that the series resistance of the capacitors can be another limiting factor but I’m not sure how this relates to the thyristor capabilities.

By limiting factor I’m referring to delivering a short enough full power pulse at a given projectile speed. I realize the thyristor must fully discharge and the launcher won’t be as effective if the rate is too slow, but again this may come down to the capacitors.

Sorry for the long winded question. I already know what I’m trying is dangerous and ridiculously ambitious but feel free to tell me that if you want. I have a good idea of the overall theory behind what I want to do but little in the way of electronics knowledge.

[Konkedout]

Looks like an obsolete device.

Thyristors are relatively good at turning ON but (except maybe for GTO thyristors) need to be turned off by the current turning off elsewhere.  That is why a lamp dimmer output waveform can have a steep rising edge but the current turns off only when the AC input passes through zero.

I did not find this part at Digikey.

When I googled it I encountered the name Powerex.  I found a Powerex website but that did not help either.

I think that designing with an obsolete device makes your life more difficult.  I wonder whether your design might work better with an IGBT or maybe a Silicon Carbide FET.  I think many electric cars use Silicon Carbide FETs. 

[Michelle]

Looks like an obsolete device.

Thyristors are relatively good at turning ON but (except maybe for GTO thyristors) need to be turned off by the current turning off elsewhere.  That is why a lamp dimmer output waveform can have a steep rising edge but the current turns off only when the AC input passes through zero.

I did not find this part at Digikey.

When I googled it I encountered the name Powerex.  I found a Powerex website but that did not help either.

I think that designing with an obsolete device makes your life more difficult.  I wonder whether your design might work better with an IGBT or maybe a Silicon Carbide FET.  I think many electric cars use Silicon Carbide FETs.

Thanks,

IGBTs may be the best choice and is also an avenue I’m pursuing, especially for the reluctance coil gun I want to build (but that’s a different topic). I’ve heard of sic fets but am not sure if they’re affordable or available, or can handle as high of currents as an igbtb

As far as turning off the thyristor, the operating principle in this case I believe is that one would design the coil in such a way that the length of the pulse is determined by the ring down of the waveform, and just letting the capacitor fully discharge.

The question then is what effect does the thyristor have on the waveform.

Another thing I’m unsure of is whether (for an induction launcher) I can use the initial positive “hump” of the waveform to induce the eddy current that will accelerate the projectile (and allow me to use electrolytic caps with the negative voltage cut off) or if I need to use the full ring down wave form in its Ac format, at a given frequency, to induce the eddy current (which would mean I can’t use electrolytic capacitors, I think).

Based on my research nobody has a home built induction launcher so I’m sort of in uncharted territory. The good news is I have a good barrel and projectile with minimal air gap. I have a lot of scientific papers about these being built in labs but they don’t always go into detail about their switching circuits; in one case they used ignitrons!

Again I’m pretty ignorant and need to learn a lot here, and that’s part of why I’m looking into multiple options.

[bdunham7]

How fast do you need them to turn on?

[Michelle]

That’s a great question that’s really difficult for me to answer right now but ballpark for starting with a single stage 1ms for the peak of the main/first sinusoid?

[bdunham7]

I have a few Soviet-era thrysistors that I've never found the time or a reason to conduct any experiments on, but I've been told that they're not fast and like large IGBTs, the require time and gate current to turn on fully which can mean that power dissipation can be higher than expected just due to parts of the device turning on before others.  They really are intended for HV 50/60Hz power control.

Here's a datasheet from a similar device:

https://asenergi.com/pdf/semiconductor/thyristor/t_disc/t553-800.pdf

I didn't read the entire datasheet in detail (there's a lot to look at) but one issue I see is the critical di/dt of 1kA/ms.  Will your inductive drive have enough inductance to slow the current rise to that rate?  With a 50-hz sine, the current comes up gradually whereas in your device you are hitting them with capacitive discharge.

Just out of curiosity--and perhaps to get an idea for mine--how are you planning on mechanically mounting and heat sinking them?

[Michelle]

Thanks,

At this point I haven’t designed the lc circuit so theoretically I could limit the current rise however I wanted to. I intended to simulate that but without a specific goal I’m kind of taking a flyer. On the other hand my projectile choice imposes some design limitations to start off with as well as voltage and capacitor limitations. I figured that whatever I’m going to come up with for the first prototype would be less demanding than what the large thyristor can handle but maybe that isn’t the case.

How would I know if I exceed the critical di/dt? Will it get hot? Explode? Just limit the output current after a certain point?

I don’t have a plan for mounting and heat sinking them but for me that’s the least of my worries lol as I have access to a machine shop and plenty of materials. Just looking at them I would be inclined to sandwich them with the flat sides touching heat sinks and being held together by screws about the perimeter of the thyristor.

I'm thinking something like these would be a better choice I just wanted to play with some thyristors too:

https://www.infineon.com/dgdl/Infineon-IKY75N120CS6-DS-v02_01-EN.pdf?fileId=5546d462636cc8fb016388021f2616cf

[bdunham7]

How would I know if I exceed the critical di/dt? Will it get hot? Explode? Just limit the output current after a certain point?

I don’t have a plan for mounting and heat sinking them but for me that’s the least of my worries lol as I have access to a machine shop and plenty of materials. Just looking at them I would be inclined to sandwich them with the flat sides touching heat sinks and being held together by screws about the perimeter of the thyristor.

I'm not sure exactly what happens if you exceed the critical di/dt, but I imagine it is thermal damage to the portions of the device that turn on first.  Whether that results in a failure to short or increased voltage drop IDK.

For mounting, etc keep in mind that they need to be clamped with several tons of force in order for the internal die to fully make contact and in normal usage they can dissipate kilowatts of power. 

[Michelle]

How would I know if I exceed the critical di/dt? Will it get hot? Explode? Just limit the output current after a certain point?

I don’t have a plan for mounting and heat sinking them but for me that’s the least of my worries lol as I have access to a machine shop and plenty of materials. Just looking at them I would be inclined to sandwich them with the flat sides touching heat sinks and being held together by screws about the perimeter of the thyristor.

I'm not sure exactly what happens if you exceed the critical di/dt, but I imagine it is thermal damage to the portions of the device that turn on first.  Whether that results in a failure to short or increased voltage drop IDK.

For mounting, etc keep in mind that they need to be clamped with several tons of force in order for the internal die to fully make contact and in normal usage they can dissipate kilowatts of power.

wow! I had no idea they needed to be clamped like that lol. Thanks for that tidbit. It's not really an issue to make clamps with some bolts however if I don't know how much force it needs since I don't have a spec sheet then I suppose that could be a problem! Maybe using thyristors is just more trouble than it's worth when I can use IGBTs...

Thanks this is exactly why I made this thread. You don't know what you don't know.