300w Boost converter

[oztules]

My diaries entry for "300w boost converter" can be found here.

http://www.fieldlines.com/story/2006/5/4/31543/52124

Somehow it has disappeared from my "my diary" page, and so this hopefully will stay as a signpost to it.

Good idea Dinges

[willib]

glad to see i'm not the only one confused with what your up to..

why are you using dc rectified mains voltage ?

on your new converter?

[oztules]

Willib,

I seem to be lacking a high voltage wind generator at the present, so the easiest way to get that kind of voltage is to rectify the mains. This will give me an unforgiving environment to test its mettel. (as the very low impedance will find any fault in the control and protection circuits pretty quickly). If it can be idiotproof on the mains.... it should be robust enough on a mill.

It will also serve as a battery charger with float as well, so had to test it with the mains anyway. (set it for 14.5, 28.8, 42v etc.)

It will currently charge a 12v, 24, 36v, battery pack at <20A, simply by turning the potentiometer.

So it's a test bed for the pwm control system, as well as a useful device. Whether it can be useful on a HV mill (such as a motor conversion or a f&P remains to be seen.

In reality, I could probably squeeze a Power Factor Correction circuit where the mains filter circuits are, and may look at this as well.

How is your pwm coming along?

.................oztules

[willib]

i got a quarter amp out of it .

i linked two coils in series

the coils are fairly heafty around 130 to 150 turns each of 16 gauge wire.

i havnt found a suitable core yet

[oztules]

any idea of the duty cycle / pulse width

[willib]

do ya mean at maximum power?

 a whole 6 watts lol

[oztules]

Yes Willib

When it's thundering alone at full throttle (6W), what is the waveform of the:

a. output of the pic.

b. accross the resistor you have in series with the battery.

Result from a: we will know the duty cycle of your max output condition.

Result from b:....

The instantaneous current drop over the safety resistor you have, may be killing your output. If the amplitude of the voltage drop over that resistor is larger than a volt or so, you may be better able to explain why you have not turned the world upside down with your massive power output:)

The only way to get this info is by using the scope, and checking the voltage drop with each pulse. A meter will only give you average current, which doesn't tell the story.

If your safety resistor (in the donor battery line) can't get enough instantaneous current through it to set up a decent mag field, then you will only get token output, and a sloppy / slopey square wave.

Perhaps use a largish electro-cap after the resistor to help get enough instantaneous current to the switchers. this may bring up your power without losing the safety effect of the resistor.... too big a cap will start to negate the effect of the resistor in some circumstances.... if your confident in your device, you could do away with the safety resistance all together.... but only if you can control the duty cycle from the start condition.

...........oztules

[willib]

scope is set at 0.2 mS & 2V/div

this is the pic output at max power

this is the voltage across the resistor ( three 1 ohm res in series) the resistors are the only part of the whole setup that get warm to hot when she's turned on.

same scale as before .2mS/ div & 2V/ div

[willib]

it looks like the 16F88 (aka pic) is on for three quarters of the period , and off for one quarter

it also looks like the input current is a heafty 5A peak to peak

the shape of the second wave kind of looks like a capacitor charging and discharging

[willib]

i have been very reluctant to reprogram the pic16F88 , mainly because i have to reprogram the programmer.

Because i lost the newest version , when i shorted the keyboard out trying to help someone out with his real world flight simulator - simulator! its a long story....

the upshot of it is the origional program in its sort of final stage is still on the HD ,BUT the HD is an old  200MB 3 1/2 " full hight drive ,which has its own isa HD card ,  which i havnt been able to get at, to get the program out of it...its a mess .

so until i feel like dragging out all the specs on the 16F88 , using the older version , and changing it to make it work with the 16F88 ,im stuck in 1.25Khz land as far as the pic pwm is concerned..

[willib]

i'm sorry i meant to say it is a 32MB seagate ST138 .....

the reason i put it on that machine was the motherboard had a isa slot which i needed for my homegrown pic programmer .

of course i could download the pic programming software , and build/or buy  another programmer , that would be too easy

[willib]

knock knock * opens door *

helllloooooo

hmmm an echo !

no one home

* pops open a brew *

hmmm nice chair , real comfy

guess i'll just watch a little tele till oz gets back

then zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz

[oztules]

Well Willib,

It looks like what i alluded to in the previous post is what is occuring. The resistor is causing the current build up in the inductor to be slow, and so you loose the effect of the steep square wave to drive the inductor. Try an electro after the resistor, and check the wave form again.

You have severe losses, as even with half pulse width, you should double the voltage, and so drive the power more than 6w.

The resistor, and the excessive length of the coil, both add up to destroy your square wave.

A higher frequency will help with the coil length. Current limit will solve your dependance on the resistor.

remember to use high speed diodes.

..........oztules

[kita]

not sure if this is a bit of a piggyback, so sorry oztules if I am trampling on your diary but I have a question which I think is relevant... After my post here (http://www.fieldlines.com/story/2006/7/27/171012/024) dan or dave mentioned that you had been experimenting, with others, on variable dc vooltage boosters. I have read up the diaries and its fascinating butI don't think I am ready to take on a project like this. But I also haven't found any commercial solutions: a 12v (or 13.4V) to 24Vish converter that can handle 100A @ 12V. Is it conceivable that your work could be scaled up to handle these currents, so that I could winkle out a retired electronics engineer to help me, or should I just design around the problem in the ways that others have suggested?  

thanks for any help, and please let me know if I should move this somewhere else.

[oztules]

Hi Kita,

I think Dan (maker of toys) may have been referring to this story

http://www.fieldlines.com/story/2006/6/10/13381/2616

rather than the 300w boost converter. It describes a voltage regulator .. simple but useful for any voltage you wish to set it to. It has a handful of components and is still working flawlessly in its new alternator. It is the kind of thing you may want if you need to change any alternator to another voltage, or the same voltage if you blow the internal regulator of your current one.

The 300w booster may also solve your problems, but rather than up-grading it, it may be more useful to duplicate it 3 times.

This will give you redundancy if any of them fail, and allows for easier wiring thickness etc. The one shown with the commentry should give you enough detail to play with. You can modify it's behaviour easily, and descriptions of changes are in the story and posts... (somewhere amongst the drivel).

If it were me, I would probably see if I could fit a second (24v) alternator on the motor, which would be dedicated to the 24v battery pack.... and  maybe fiddle with the regulator to change the charging regime a little as well. (this will require some space under the bonnet, and an adapter plate and belt and possibly new pulley on the current alt or steering pump or air con pulley etc, to hang the new belt off.)

If that is a no-go, then change the present alt to 24v, charge the 24v batts, and then  charge the 12v battery with a buck converter or similar. There are many after market "trucker" step downs 24 to 13.8v. to run all manner of appliances at 13.8v from the 24v truck systems.... At least this uses off the shelf solutions.

For a down and dirty 13.8v voltage source for your 12v battery, a dozen LM317's in parallel will give you 36A @ 13.8v (or whatever)  fully voltage and current regulated for almost no other parts,(two resistors will do it) and can be picked up for very little. This will solve the 12v from the 24v pretty quickly.... not for the purist, but for the animals amongst us will work fine for charging the 12v battery.

At least if you change the alt to 24v, a quiet idle may see enough current going into the batts to get you out of trouble on long films showings or a smaller battery pack.

Whatever you do, I hope you have fun doing it....

......oztules