Effects of Recirculating BEMF to Coil

[gotoluc]

Luc,

Could I get a bit of info?

I'd like to know what that 22 Ohm resistor and the auto bulb each read on your inductance meter. Also, could you measure what the hot resistance is of the bulb? The "hot" resistance would be soon after disconnecting it from the circuit and measuring with your Ohm-meter.

One last thing, is the bulb the first thing in series with the + side of your storage capacitor for all your tests so far?

[Edit] I also assume that the 22 Ohm in series with the MOSFET Source lead is no longer in-circuit and not being used?

Thanks,
.99

Hi .99

the bulb seems to be unmeasurable on my inductance meter. The lowest setting on my meter is 2mH and it works its way down to .000 then the -.000 comes up. The 22 Ohm resistor is .015mH

The bulbs cold DC resistance is 4.5 Ohm and if I have it at the same intensity with 200vdc switching around 1.8KHz and disconnect it the fastest I can (about 3 seconds) it measured 5.5 Ohm and worked its way down.

No! the bulb was first thing in series with the - side of your storage capacitor to the MOSFET Source for test 5 and 6. In test 3 it was as you say above.

Yes the 22 Ohm in series resistor is no longer in-circuit on the input side.

Thanks

Luc

[gotoluc]

In fact it should be possible to construct a simple feedback loop that would vary input peak current to the bulb to keep the bulb at a constant repeatable brightness; then the bias current would be the dependent measure.

A good idea TinselKoala! can you design one. Even better, I wounder if it would be possible to make one that would be giving an Energy measurement also.

Luc

[gotoluc]

Hi everyone,

a YouTube user posted a comment which explains very well and my reply may also help some.

User: Can101276
I like your unconventional testing, it is crude, however it will alow you to see work being done and see how your changes effect the work performed.

Correct me if im wrong, but what I got out of your experiment was that you can get more work performed with high voltage, high frequency pulses, than with lower frequency, higher duty cycle pulses at the same voltage.
So in dummy speak, more work for same watts!!!
You are on the right track.

My reply:
Yes Can101276, you are understanding this correctly. However there maybe limits as my higher voltage tests over 250vdc the efficiency starts going down but that maybe an electronic component issue causing this. I also see efficiency starting to go down when frequencies over 5KHz are used. This could have something to do with the coil characteristics. More testing will be required to understand this better.

Luc

[TinselKoala]

A good idea TinselKoala! can you design one. Even better, I wounder if it would be possible to make one that would be giving an Energy measurement also.

Luc

I think the circuit would be very simple. A 2n2222 controlling the voltage to the bulb, the photoresistor in the gate circuit of the transistor, the ammeter in the same circuit, some appropriate trimming resistors and a cap or two just for grins, and Bob's yer uncle as they say around here. Bulb gets dimmer, photoresistor increases, gate current decreases, transistor lets more current thru to bulb, bulb gets brighter, photoresistor decreases, gate current increases, transistor lets less current thru to bulb.... trimmers set brightness point, microammeter in gate circuit can be calibrated to a known power setpoint. (So many microamps on meter, not directly caused by power circuit, but can be accurately correlated with power, because as the power circuit increases voltage to try to make the bulb brighter, the photocell circuit uses gate current to make the transistor cut the power to the bulb by the corresponding amount.) This is off the top of my head, it's probably backwards or something. (NPN, PNP, always confuses the heck out of me)
I'd breadboard one up in an instant (and a half) if I only had the photoresistor on hand. I've got some in another location and I'll pick them up tomorrow and try it, let you know.
But even better would be to just record the bulb's brightness level using the photocell and the software chart recorder function that might have come with your scope software.

Funny that you should mention the energy measurement. In the old days people would sometimes look at the oscilloscope screen using a photoresistor. The brightness of the trace picked up by the photoresistor could be correlated with the energy in the signal giving the scope trace if the scope was set up right. The photoresistor acted like an optical electronic integrator. If your scope has a "Z" input this is one thing it can be used for.

[poynt99]

A good idea TinselKoala! can you design one. Even better, I wounder if it would be possible to make one that would be giving an Energy measurement also.

Luc

Luc,

You've already got an instrument for measuring the "brightness" of your input bulb.

Replace the bulb with a 1W 10 Ohm resistor (or experiment with Wattage and resistance values between 3 and 15 Ohm), and use your laser temp gauge/meter to measure and adjust the pulse width to obtain the same baseline temperature in the resistor each time a change is made. This is probably more accurate than eyeballing the bulb intensity anyway.

Thanks for the component specs btw

.99