You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
D
duivendyk
Guest
The good thing about this ptp circuit is that the cap is a high impedance (takes little current) at cranking speed.Your results are more in line with what I expected.Because of the high
instantaneous currents (short duty cycle inherent in ignition circuit) you need gutsier diodes.The current requirements on the diodes also increase with decreasing batt. voltage (higher currents for the same power).Schottky diodes have lower drop about 0.3 V less, but need also 3-5 amp rating,inclined to doubt wether worth the money.The caps (ESR) is a different story you don't know what kettle of fish you're in bed with (clammy metaphor),beyond a certain minimum the total value does not really matter.But parallelling a bunch of them gets you lower ESR.Because of the high peak/average ratio, these losses, the 2.8Ohm in the winding and the diode drops and the cap ESR all affect the output to some extent.In any case your results are pretty close to what I would have expected.
Based on the likely shape of the pulse and some simplifying assumptions, I have concluded that the best battery voltage is probably in the 7-10 V range.This is a fairly broad optimum.With your Nicads to cannibalize you could fashion something in that range, of course with the 6V job,allready around you could try that first.But you won't be able to measure because of your meter issues.The lower the voltage the more likely we are to mess up the ign. circuit (higher currents).So far so good,go to it.Plse read following post!!
instantaneous currents (short duty cycle inherent in ignition circuit) you need gutsier diodes.The current requirements on the diodes also increase with decreasing batt. voltage (higher currents for the same power).Schottky diodes have lower drop about 0.3 V less, but need also 3-5 amp rating,inclined to doubt wether worth the money.The caps (ESR) is a different story you don't know what kettle of fish you're in bed with (clammy metaphor),beyond a certain minimum the total value does not really matter.But parallelling a bunch of them gets you lower ESR.Because of the high peak/average ratio, these losses, the 2.8Ohm in the winding and the diode drops and the cap ESR all affect the output to some extent.In any case your results are pretty close to what I would have expected.
Based on the likely shape of the pulse and some simplifying assumptions, I have concluded that the best battery voltage is probably in the 7-10 V range.This is a fairly broad optimum.With your Nicads to cannibalize you could fashion something in that range, of course with the 6V job,allready around you could try that first.But you won't be able to measure because of your meter issues.The lower the voltage the more likely we are to mess up the ign. circuit (higher currents).So far so good,go to it.Plse read following post!!
Last edited by a moderator:
D
duivendyk
Guest
ET,there is a way to extend the range of your meter.The basic idea is to shunt part of the current to be metered around the meter by sticking a resistor across it,if say only half the current went through, it the full scale reading would represent 600mA instead of 300.But you would have to calibrate it somehow.Here's how.
You have a 12V battery,hook up your meter in series with a variable resistor (50 ohm +) start with max resistance,reduce resistance till you read full scale.Now stick another variable resistor across meter,my guess pretty small, couple of Ohms. You could try a 10 ohm fixed job first and see what this does and take it from there, if it for instance reduces the reading by 10%, the meter resistance is about 1 Ohm and you would need 10 of them to get the 1 ohm you need in order to divert half the current around the meter.I have no good idea what your meter resistance is, depends on the sensitivity of the meter movement (more sensitive= $$, and lower R of meter).You could combine a fixed one with a variable (25-50 ohm)for fine adjustment.If the required resistance is very low, which I doubt ,you may have to resort to paralleling fixed ones.You could give this a try.Do you understand the principle?.If you want to extend the range still further you can of course shunt it even more and go for 1 Amp full scale,i.e. have roughly 1/3 of the original resistance that is, put a bit less than 1/2 of that value across it.Then you'll be a good shape for lower batt. voltage tests.
You have a 12V battery,hook up your meter in series with a variable resistor (50 ohm +) start with max resistance,reduce resistance till you read full scale.Now stick another variable resistor across meter,my guess pretty small, couple of Ohms. You could try a 10 ohm fixed job first and see what this does and take it from there, if it for instance reduces the reading by 10%, the meter resistance is about 1 Ohm and you would need 10 of them to get the 1 ohm you need in order to divert half the current around the meter.I have no good idea what your meter resistance is, depends on the sensitivity of the meter movement (more sensitive= $$, and lower R of meter).You could combine a fixed one with a variable (25-50 ohm)for fine adjustment.If the required resistance is very low, which I doubt ,you may have to resort to paralleling fixed ones.You could give this a try.Do you understand the principle?.If you want to extend the range still further you can of course shunt it even more and go for 1 Amp full scale,i.e. have roughly 1/3 of the original resistance that is, put a bit less than 1/2 of that value across it.Then you'll be a good shape for lower batt. voltage tests.
Last edited by a moderator:
I think i understand, but have a big headache at present so will need to read it over when im a bit more clear headed. I have 3 mm's so i could manually test the resistance of the meter, if that will work.
I have one variable pot, ill grab another during lunch today.
I have one variable pot, ill grab another during lunch today.
D
duivendyk
Guest
Be careful that you don't bust the analog meter that you try to measure the resistance of!.You have to put it on the 300mA scale to check it First put the meter you use for the resistance measurement on the 1-10 ohm scale, and take a digital one and measure the current the first meter puts out to see if not above 300mA.Take it easy, Rome wasn't built in a day .
ok, i have read over and understand how the circuit is meant to work - makes sense to me. I'll have to raise the idle of the bike until it reaches 250 (limit of meter), put variable resistor in series and tweak until needle just deflects lower than 250 to show limit of meter. Then put resistors in parallel across meter to divert current through resistors - if i reduce to say, 125ma then im still getting 250ma through circuit, only half is though meter, so if i increase to 250ma by reving engine im really reading 500ma since 250 will be going through the resistors.
what i dont understand, is how the resistor in series comes into it - will you calculate that out based on the resistance it applies? Also, I imagine ill need resistors that can handle 5w+?
also, i wont bother reading the meter using dmm - probably not as accurate. Also, the electronics store i go to have shunts for around $7us, will probably be more reliable than me doing it, and almost cost the same (www.jaycar.co.nz) - could you look at see if its worth buying one of those instead? 5W resistors + pots will quickly add up in cost, i would imagine!
what i dont understand, is how the resistor in series comes into it - will you calculate that out based on the resistance it applies? Also, I imagine ill need resistors that can handle 5w+?
also, i wont bother reading the meter using dmm - probably not as accurate. Also, the electronics store i go to have shunts for around $7us, will probably be more reliable than me doing it, and almost cost the same (www.jaycar.co.nz) - could you look at see if its worth buying one of those instead? 5W resistors + pots will quickly add up in cost, i would imagine!
Last edited:
D
duivendyk
Guest
I was trying to save you the trouble of having to use the WW as a power source and thought it might be more convenient to use a battery and a variable resistor to set up the 250 mA.If you think it's easier to use the engine to do that,go right ahead that's fine, no problem .All we'r really interested in is to put enough resistor across the input to get down to the 125 reading.You get it now?
D
duivendyk
Guest
I thought so,take it easy,Tylenol 500mg,2 max.You haven't been breathing exhaust fumes I hope.Carbon monoxide poisoning produces headaches !
Last edited by a moderator:
nope, definitely not - have been doing everything outside in wildly differing weather, i blame it on the change of season (tends to do it to me).
Right, well, i stuffed up + blew the pot i had (doh!) so i had to do the testing on the engine itself, which ended up working ok, i think (results seem to make sense, hopefully they're reliable?)
circuit as before, with 4 1w 1ohm resistors in parallel with the meter, measured it to be the same resistance as the meter - meter wasnt showing that way though on current flow.
Original current flow = 175ma. With resistor shunt circuit running in parallel, dropped to 40ma.
so, i would assume 175/40 = 4.375 multiplier.
reving engine took meter to 75ma. 75ma*4.375 = 328ma peak current produced, 13.9v battery, peak wattage 4.56w. Considering you can get **** good LEDs @ 3w, should be a good light even with 12v set up (id be happy enough, at any rate). not sure if my maths is right here, but the result is right about where i expected it to be.
Going to do the '6v' battery test now with the butchered battery. Im getting some fairly inconsistant results, i blame this on the weather, the fact im using connector blocks + breadboards for this, and the generally roundabout idle that the happytime seems to have. IE idle current last night was over 200, now its 175 - perhaps im idling it lower, or perhaps it was warmer last night and the metals were fractionally closer? who knows.
Will be interesting to see the results with the 6v battery though.
Right, well, i stuffed up + blew the pot i had (doh!) so i had to do the testing on the engine itself, which ended up working ok, i think (results seem to make sense, hopefully they're reliable?)
circuit as before, with 4 1w 1ohm resistors in parallel with the meter, measured it to be the same resistance as the meter - meter wasnt showing that way though on current flow.
Original current flow = 175ma. With resistor shunt circuit running in parallel, dropped to 40ma.
so, i would assume 175/40 = 4.375 multiplier.
reving engine took meter to 75ma. 75ma*4.375 = 328ma peak current produced, 13.9v battery, peak wattage 4.56w. Considering you can get **** good LEDs @ 3w, should be a good light even with 12v set up (id be happy enough, at any rate). not sure if my maths is right here, but the result is right about where i expected it to be.
Going to do the '6v' battery test now with the butchered battery. Im getting some fairly inconsistant results, i blame this on the weather, the fact im using connector blocks + breadboards for this, and the generally roundabout idle that the happytime seems to have. IE idle current last night was over 200, now its 175 - perhaps im idling it lower, or perhaps it was warmer last night and the metals were fractionally closer? who knows.
Will be interesting to see the results with the 6v battery though.
Last edited:
Similar threads
- Replies
- 3
- Views
- 360
