COP 17 Heater | Rosemary Ainslie

[Aaron]

update

update - changed 0.01 to smaller then put it back... (left cap of bottom 3 in the bottom right).

This schematic gives 2.419kHz @ 11.6% on.

I need to change something to get 3.7%.

The range this does is good anyway, but I want exact. down to 3%~3.7 - coil saturation level.

[Aaron]

200kHz+

In oscillations, frequency read by scope has been over 200kHz.

[Aaron]

quantum schematic update

Good range now.

Will tweak as necessary but we're very close. This gives 10% at 25kHz.
Maybe too fast, but anyway, better than the stock schematic.

[Aaron]

diodes on timer circuit

http://www.oup.com/us/pdf/microcircu...14B_philip.pdf

http://www.nxp.com/acrobat_download/...8_1N4448_5.pdf

The 4148 is just a little beefier as far as current and continuous reverse voltage - but in this circuit they are very compatible, especially in their recovery time - both are 4nS at identical rated test conditions.

Ok, good to know. The 1n914 was a tad bit warm.

I had hundreds of them so used them. The package of 4148's at radio shack also had the number 1n914 on the package the guy said they are identical. Guess not.

Will post updated schematic soon.

[Harvey]

p.s. the 5.2k resistor on the quantum circuit - I actually have 5.05k. Closest I could match it.

I used 1N914's instead of 1N4148 - but they are the same thing.

My gate pot is a 10 turn precision 2k.

I want a 100~200 ohm pot but can't find one at radio shack. Will get one, make it easier. I can see the gate pot tuning is a bit touchy but MAKE SURE TO HAVE THE RIGHT DUTY CYCLE AND FREQUENCY for your particular circuit. Jut do trial and error.

MAKE SURE THE FREQUENCY ISN'T SET TO MAX. Back it out to about 1/2 your range, then increase gate pot. If no oscillation - keep gate pot increased a bit then slowly increase and decrease off time - you should see your timer signal pulsing the mosfet across the screen suddenly turn into many oscillating samples all across the screen instantly.

http://www.oup.com/us/pdf/microcircu...14B_philip.pdf

http://www.nxp.com/acrobat_download/...8_1N4448_5.pdf

The 4148 is just a little beefier as far as current and continuous reverse voltage - but in this circuit they are very compatible, especially in their recovery time - both are 4nS at identical rated test conditions.

[Joit]

Wistend i agree with, that the Output from the Timer depends at the Load.
I saw the Waveforms different at a Coil or a Heating Element.
Seems there depends more at the Signal, what comes in.

And i agree too, that there is kind of Self ordering from the Coil/Load, as you see it at a SG.
I saw this too at couple Coils, right now, i run a Bedini, what speed up with lower Resistance at the Basis to a certain Point,
first then i can turn the Pot higher, and it speeds up more.
It dont speeds up, when it runs a low RPM, first it need adjustment to this Point and a certain RPM, or the Frequency turns over.
And well, when you wanna have a efficient Bedini, put 4,5 or more Diodes parallel for the charge, and you will see a different.

A Conclusion.
We pulse and saturate a Coil/Load to a certain Point, where we dont waste Energy.
When there is the right Amount of Energy in, you can get a certain Amount back, where it depends at the Setup, how much this will be.
Each Load has a certain Point, where it resonate.
Mainly you need first to find this Point over the Dutycycle and the Frequency.
You can try it with a bigger Coils, it seems it s more better to see there,
because it is easier to find the Point.
About the Heat, it is created from the Energy, what moves through the Lead.
You can put a balance there, like a Resistor, and can move the Point around, where the Heat appears.
I could figure a Setup like 2 Coils and at the middle a heating Element, that you can move more Energy through the Circuit.
But maybe it works with Caps at the same Way.



Just to bad, that this Doorman, what worked 25 Years at the Industry did not do his Homework.
Maybe he would had have a life now, and could tell something.

[Harvey]

Hi everyone,

Here are some documents I posted in the other threads you may find useful ...

"International Rectifier" Application Note AN-1005
Power MOSFET Avalanche Design Guidelines

"Advance Power Technology"
Understanding the Differences Between Standard Mosfet's and Avalanche Energy Rated Mosfet's

Best Regards,
Glen

Thanx Glen, very helpful links

[Aaron]

misc

p.s. the 5.2k resistor on the quantum circuit - I actually have 5.05k. Closest I could match it.

I used 1N914's instead of 1N4148 - but they are the same thing.

My gate pot is a 10 turn precision 2k.

I want a 100~200 ohm pot but can't find one at radio shack. Will get one, make it easier. I can see the gate pot tuning is a bit touchy but MAKE SURE TO HAVE THE RIGHT DUTY CYCLE AND FREQUENCY for your particular circuit. Jut do trial and error.

MAKE SURE THE FREQUENCY ISN'T SET TO MAX. Back it out to about 1/2 your range, then increase gate pot. If no oscillation - keep gate pot increased a bit then slowly increase and decrease off time - you should see your timer signal pulsing the mosfet across the screen suddenly turn into many oscillating samples all across the screen instantly.

[Harvey]

Aaron, can you post all the numbers you see on your 555's?

And have you tried swapping the chips to see if they are consistent?

Thanx.

[Aaron]

quantum circuit

If the fact that TK and I (and a few others) were correct all along regarding the duty cycle issue for the Quantum circuit doesn't bolster our credibility with you folks here (the recent subject of discussion), then I don't know what would.

There was such a big deal made out of the circuit being wrong I never wanted to use it because I trusted.

TK as absolutely right about the duty cycle issue not getting 3% but I have no idea how he was stuck on 97% ON time - because the circuit tunable with the pots. With frequency - it is limited to 500Hz.

However, with all the claims about oscillation being red herring, it won't oscillate at all, etc... I found the opposite to be true and proved in on the video I posted early this morning.

It not only oscillates the mosfet but with the exact proper waveform where it doesn't default to zero. It stays above.

It won't happen by simply increasing gate resistance. The duty cycle and frequency has to be in the right range - and will be different for each individual circuit. Then increase gate resistance and boom. You got it.

[FuzzyTomCat]

Hi everyone,

Here are some documents I posted in the other threads you may find useful ...

"International Rectifier" Application Note AN-1005
Power MOSFET Avalanche Design Guidelines

"Advance Power Technology"
Understanding the Differences Between Standard Mosfet's and Avalanche Energy Rated Mosfet's

Best Regards,
Glen

[Aaron]

water heater element

Interesting we are probably all using 40% more energy than we nead to, just to have the same result!

Hi Michael,

You mentioned the water heater elements are actually inductive resistors. I never knew that. I thought they were some straight resistive element.

It that the case in the states? Are all the elements simply inductive resistors?

If so, that explains a lot!

If it is operating at 60Hz for example at 220v, then it probably has horrible power factor and simply correcting with the right capacitor will probably drastically increase the efficiency.

My tank probably uses 3500 range when it is on. By simple power factor correction, it will probably only use 2500.

I always thought they were 100% efficient at making heat because I thought it was some straight non-inductive heating element.

[Aaron]

Updated Circuit

I was up until 5am enjoying with circuit.

In red, look at the one modification I found by switching out caps and watching what happens, it allows a much smaller duty cycle. Will do more.

@Harvey - I saw your post and will have to play with some mods to see what else I can get out of that circuit. Much higher frequency would be nice.

Anyway, I don't know why this circuit gives me the oscillation that doesn't settle to zero and the normal 555 circuit I've been using doesn't. In either case, all I have to do is sneeze and my mosfet oscillates. They are all authentic oscillations.

The pic I posted last night of the oscillations, THAT IS THE EXACT WAVEFORM EVERYONE NEEDS to get. On the shunt, you can see spike - then going positive and looks on for the whole cycle then spike and repeat. You don't see it come down to zero at all. That IS what you want. It appears to be on at full duty cycle - but it isn't - the real on time that the battery is giving corresponds to the on time of the timer.

When you zoom in on the coil, you'll see it doesn't settle at zero either.

When looking across the load, LITERALLY, go across the load. Ground of probe to negative of resistor and probe at positive where battery connects to it. It should be across exactly like you put the scope across the shunt.

Do NOT look at the waveform and rms reading on coil by placing the probe ground on common ground like Quantum article shows. That is ONLY IF you want to use both channels at same time to do a comparison.

Anyway, the circuit below I never built because I thought it was completely wrong. Not! Even without my mod below and at 60% duty cycle, the mosfet will oscillate with proper waveform just fine!

Try it, you'll like it.

This will be the circuit I'm using from here on out. I'll post any other useful mods as I find them - will play with some of Harvey's analysis of it.

[gotoluc]

EDIT WELCOME BACK LUC. What a pleasure

Thanks Rosemary for the warm welcome

Welcome back Luc, hope everything is OK. The post I made was just a bit of pent up frustration, I suppose the safety valve let off a little steam.

The experiment that has been running at my neck of the woods is just an idea I had and probably at the end of the day it will show that when an item such as a heating element has a rating of 1.5kw and connected as recommended, it is very inefficient in the use of that energy. Interesting we are probably all using 40% more energy than we need to, just to have the same result!

More results will come when my partner in crime returns from his hols:

OH, perhaps TK should shave off his mostach

Mike

Hi Michael,

thanks also for the warm welcome The trip went well and I'm also doing well. Looking forward to getting back to my bench tests and development.

Luc

[witsend]

Golly Harvey - it seems I've way underestimated your abilities here. EDIT Apologies. I see you've already suggested the problem on our 555. My 'help' in preparing the paper was not the same as the builder of those 555's and I have no idea how any such is constructed. It's just way more than I ever wanted to know. Actually regret this now.

But there's something unique in the properties of this. TK complained that he built it to spec and found that it defaulted to 90% on. Aaron can adjust it to 60% ON at the smallest duty cycle. But his actually resonates at 100% on - albeit that the switch is still switching at 60%.

Our own resonance waveform pattern is also effectively permanently on. We get a clear 3% periodic waveform but with multiple oscillations inbetween. This automatically compounds the frequency. And the waveform only defaults to zero with momentary ringing - replaced by a near perfect but slightly varied copies of the two primary waveforms between each switching cycle. So that ringing is the only moment that it actually gets to zero and it really is momentary. The multiple oscillations between the 'duty cycle' waveforms was identified as a parasitic Hartley effect by an acknowledged boffin at one of your leading universities. He was also going to do the tests. He has never communicated the outcome of those tests notwithstanding my applications. I would have thought he'd be more than ready to deny anything significant here if, indeed, that's what his experiments found.

Whatever the cause of this, the results are extraordinary as it relates to the performance of the battery and to the energy dissipated at the load. And I'm reasonably certain we were nowhere near optimising this.

Regaring the degrees separation. I see this now. I took the trouble to read the links. It is indeed interesting. And I guess it's better that everyone stay in their comfort zones when it comes to disclosure of their names. I still find it remarkably brave when people do disclose this. Maybe not sensible. But wow.