COP 17 Heater | Rosemary Ainslie

[lighty]

Glen, Harvey and others can probably fine tune the best diode setup for the IRFPG50. I'm using a 1n914 for the parallel didoe and a 6A100 for the blocking diode.

With a 1n914 blocking diode position, it gets super, super hot. The 6A100 is a robust but slow diode but it is handling the heat. I use a 1n914 from source to annode of blocking diode as the parallel diode and it works fine.

No matter which diode you use as the blocking diode you will always get power dissipation at that point. Either they are low voltage drop but then they are slow of they are fast but don't have current carrying capability. You should consider either using BJT transistors or better yet if you want to just replace MOSFETs buy IGBTs without intrinsic diode. There are some out there which are not too expensive and I have used them in the past exactly for the reasons you wish to practically block intrinsic diode in MOSFETs. IGBT in this case can be direct replacement without touching driving circuit.

[Aaron]

reverse current

Thanks Lighty, but if we do want the reverse current conducting capability that the body diode provides - would the IGBT with the blocking diode and parallel diode modification outperform the mosfet?

I hope I'm asking this correctly.

[Joit]

Heat

When i run with the Quantum Timer i got no heat at the Transistor, only at the Wire,
maybe there is a Way to redesign the Timer?
I think a part of the Heat comes from the Values of the Resistors there.

Fuzzytomcat
thanks for the Link, i did dload the Prog and allready played a bit around with it,
but i am again headover in other Work.
But i found a Hairdryer with 220V/330Watts ~ 1,5A to glow the Wires in there, I think, i will try this one too.
I hate waiting, when i order Wires, till i got them.

[dllabarre]

My first post in this thread

I was looking for a circuit to add to something I was working on and was invited to look at this circuit that Rosemary published. I started reading a few of the first posts and got really interested.

Then after a few pages I saw where Rosemary was invited to join the discussion and actually joined in. So I had to keep reading.

Well, to my surprise the "bashers", "doomslayers" and "generally negative people" actually started attacking her! I was shocked!

I apologize for any of the Americans that were in that group of people.
Please don't judge all of us Americans by those few.

I'm looking forward to building one of these as soon as I get the components.

Did you all decide which 555 and MOSFET worked the best yet?

Thank you Rosemary for sticking around here.

[lighty]

Thanks Lighty, but if we do want the reverse current conducting capability that the body diode provides - would the IGBT with the blocking diode and parallel diode modification outperform the mosfet?

I hope I'm asking this correctly.

Ah, I only now realised that I didn't gave close enough look to the original circuit. I'm afraid that without some more complex circuitry you're stuck with blocking diode losses.

However, depending on the current consumption of the circuit (I really don't want to go through 123 pages of this thread) you could you use Schottky diodes as blocking ones. They are very fast and have a relatively low forward voltage drop in comparison with "normal" diodes so they should also dissipate less energy. Schottky diodes are usually made up to 20A but most usual values are in the range up to 2A so decide for yourself. As for the "parallel" diodes you could use one of the hyperfast types. Check out the Electrical motor secrets thread where I already discussed that with Jetijs.

[Aaron]

circuit components

Did you all decide which 555 and MOSFET worked the best yet?

Hi Don, welcome to this thread. My personal vote is for the IRFPG50 mosfet and 555 circuit simply because this is exactly what Rosemary used for her results.

There might be something better but I'll personally stick with this for now.

[Aaron]

diodes

you could you use Schottky diodes as blocking ones. They are very fast and have a relatively low forward voltage drop in comparison with "normal" diodes so they should also dissipate less energy. Schottky diodes are usually made up to 20A but most usual values are in the range up to 2A so decide for yourself. As for the "parallel" diodes you could use one of the hyperfast types. Check out the Electrical motor secrets thread where I already discussed that with Jetijs.

Thanks Lighty, will check out the Schottky diodes and will refer to the Electric Motor Secrets thread for the hyperfast ones.

[FuzzyTomCat]

Thanks Fuzzy. Sounds really good. How are those resistors coming on?

Hi Rosemary,

It's been fun narrowing down the small issues and getting better results each time, I was waiting until it got dark to take some photos and post more aperiodic oscillations wave forms using the finer tuning with the frequency counter. I don't think the spice will like the flavor .... well see

The resistors are coming along just fine, I have all the parts now including my sewing machine foot peddle just waiting for payday to get a "Barbeque Rotisserie" to do some coil winding for next week

Glen

[witsend]

Hi Fuzzy. Nice to know that our kitty cat is still on top of everything. Can't wait for the pictures. The problem with condiments is they can be varied to accommodate taste preferences.

One more week for resistors. That's brilliant. Hopefully you'll access the brilliant Tektronix. Yet again - thank you Lisa, in the event that you're reading this.

[witsend]

I apologize for any of the Americans that were in that group of people.
Please don't judge all of us Americans by those few.

Golly Dllbarre - there's nothing to apologise for. The attack more or less faded when TK couldn't get conclusive results - which is my take on it. No doubt he'd deny this.

And I love Americans. It's your accents. Just so delicious.

Am delighted to hear that you're also replicating. But the authority on the switching components - that's up to Fuzzy or Aaron to advise. I've often seen you on line btw.

[FuzzyTomCat]

Hi everyone ,

Here are some more scope shots of aperiodic oscillation wave forms using a "Frequency Counter" on the Battery (+) and (-) connections, the "Ainslie-Murakami Negative Dominant Waveform Generator" replication is running around 1.2 Mhz

Channel 1 ground is on battery negative side of 0.24 ohm shunt and probe is at 555 negative rail side of shunt. A1 trigger @ 34mV , 50 mV, Probe at X10

Channel 2 ground is on battery negative side of 0.25 ohm shunt and probe is at mosfet source side of shunt. 20mV, Probe at X10











Glen

[witsend]

Golly Fuzzy. It's looking just so good. Well done. and

[lighty]

@Aaron

I really shouldn't write things in the 2AM otherwise I don't think straight. You only want to let current pass in one direction like from drain through source, right? And that's why you using blocking diode, correct?

Well, MOSFET can basically conduct both ways (D-S and S-D), when gate threshold voltage is reached (G-S voltage). However IGBT behaves more like BJT transistor in the sense that it will block reverse voltage between C-E. Usually that voltage is something in order of 15-20V depending on particular IGBT (check out datasheets). What that practically means is that IGBT will block reverse current flow by it's very nature and it should do it much more efficiently that any blocking diode.

Please consider that using it like that is not what IGBTs were originally designed for. I don't see any practical reasons why it wouldn't behave as predicted by literature but this is not what IGBT designer had on their minds. IGBTs are not so expensive so you could try doing it this way. Of course be sure that you use IGBT without freewheeling diode (parallel diode) and to check out datasheets for reverse C-E voltage. For example if you're using 12V system then you could use IGBT with peak reverse voltage of 20V as it's continuous blocking voltage should be around 15-16V which should give you 3-4V margin of safety.

[poynt99]

An IGBT will work as a basic switch and will conduct any reverse current if an external blocking diode is used to mimic the MOSFET body diode, but there is no guarantee that the aperiodic oscillation effect will be possible to attain due to the substantially different construction of the IGBTvs. a pure MOSFET. It would have to be tried, but I doubt there would be any advantage to going this route.

An IGBT is basically an input MOSFET in a Darlington configuration with a BJT output transistor. Also typically, IGBT's switch slower than MOSFETs do, but depending on the application, this may not be an issue.

lighty, be careful of the voltages. If a flyback diode is not used a 20V device will likely die a quick and nasty death.

.99

[witsend]

Lighty - I tried to read your link. Is there any chance of getting a translation here? I'm sure there are a number of us who'd be interested.