Honda/Huasheng/Titan,Lighting & Electrical System.

[skyl4rk]

I just ordered the parts, I will have a better idea after I build it.

I'm fairly certain it will produce a small trickle charge.

I am certain that if either of the diodes are installed the wrong way around there will be sparks and smoke.

 

[duivendyk]

Haleluja someone made some measurements !,fantastic,but I think you're jumping the gun a bit with that charging circuit though, a lot of that small amount of available power is going to end up in that current limiting resistor instead of charging the battery.That is wasted power you can ill afford.As far as the measurements go,I think we can prob. get rid of that power robbing resister.To begin with, the test voltage readings,these are ac. readings I assume ?,the peak values are 40% higher, if it is a sinusoidal output voltage (it probably isn't).secondly there is no guarantee that this is a symmetrical output (equal amplitude pos.& neg. cycles,) that depends on the loading by the ignition circuit ,which could well be unsymmetrical.This matters because the rectifiyng circuit you propose is also an unsymmetrical load.For this reason it is entirely possible that a negative going rectifier could safely load the black wire more than one of opposite polarity.All that needs to be done really, is to repeat the loading test, but use a DIODE + potentiometer to ground,do it for both speeds and repeat the test for the opposite polarity (reverse the diode). If they are different we know which way to go with the polarity of the rectifier setup.It would be useful to also measure the dc voltage at the resistor.Don't get alarmed if it is low, this is an averaged voltage over a whole ac cycle not the peak value,remember the battery is charged during the peaks
I think that by driving one of these constant current buckpuck regulators used to drive 12V LED's from a battery from the rectifier circuit and using this in turn to charge the battery we should be able to get rid of the current limiting resistor and get more juice into the battery.The regulator would need to be preceded by a sizable storage cap ( few thousand microfarads at 25V.There may be other usable charging regulators around but they are usually not designed for high efficiency at low charging currents.I don't know how much current the buckpuck regulator itself takes either,nor what it costs.The charge current is controllable which is nice,that also means that it can readily be turned off!.A 14 V zener to protect the battery would still be worth having.

 

[skyl4rk]

Good point about measuring voltage behind a diode, I will try that in a few days when I get back to my workshop. I measured DC and did not think to measure AC, which I will also do. I assume the pulse is made by passing a magnet past a coil which would be a DC pulse?

I believe the battery will act to smooth pulse dc, at least as long as the battery capacity is so much larger than the small currents we are dealing with. If not, an LC coil-capacitor could be used to shave peaks and fill valleys. The concern is not to allow peak voltages to damage the light bulbs. I doubt that the voltages we are dealing with here would hurt the battery.

Perhaps the current limiting resistor could be put on the ground side of the zener, which means that only current over 14V would be limited. That should still protect the zener.

Since the motor starts and idles at 9V, the battery charging circuit may not have an effect on the spark at starting and idling speeds, except when the battery is badly discharged. In that case a switch might be necessary to disconnect the charger circuit in order to start and idle the motor. Under normal conditions the battery voltage will oppose current flow as long as the spark voltage is lower than the battery voltage.

 

[duivendyk]

Plse. read my post of today you'll find it informative.

 

[Irish John]

Hi Irish John, I would welcome a good source for springer with cantilever studs I already have a use for it!

Thanks, Mike

Mike, everyone would like to have the option of brakes on a springer but the law in the US isn't very big on bike brakes. Because of this we get everything that is made for the US market and although we have to have front brakes by law they usually bolt on a cantilever brake which is less than useless. These ones have V-brake studs because the importer insisted on it.
They are available through me but they are really only for the Australian market which is a lot smaller than the US market. If I started sending them to the US I'd need to get a lot more in and the order went out yesterday for 12 new ones because that is what I reckon the MBc guys in Oz might buy.
If you want them in the US you really need to lean on the bike companies to get them to change their cruiser products so they have better brakes. You need something up front to compensate for the pathetic coaster brakes that seem to infest almost every cruiser. I had a bad accident on a new bike last week and it wouldn't have happened if there hadn't been a coaster brake on the bike. If you really want a springer with V-brake bosses I could send you one but the postage would be quite dear for an item costing very little. I'll start a thread on these forks to see who's interested in them.

 

[skyl4rk]

More measurements and observations:

When grounding the kill switch wire, it sparks in rythym with the motor exhaust, so the current pulses. The power is not like normal ac or dc. There will definitely need to be an inductor in line to absorb current and then release it over time.

Voltage Measurements
All measurements from X to ground.
Wire = Kill switch wire

DC
Bare wire idle: 1V
Bare wire high rpm: 2V
Diode at idle, Wire-->|--X: 1V
Diode at high rpm, Wire-->|--X: 2V
Diode at idle, Wire--|<--X: 1V
Diode at high rpm, Wire--|<--X: 3V
Bridge Rectifier at idle: 1.8V
Bridge Rectifier at high rpm: 2.8V

AC
Bare wire idle: 10V
Bare wire high rpm: 20V
Diode at idle, Wire-->|--X: 2V
Diode at high rpm, Wire-->|--X: 3.5V
Diode at idle, Wire--|<--X: 0V
Diode at high rpm, Wire--|<--X: 0V
Bridge Rectifier at idle: 0V
Bridge Rectifier at high rpm: 0V

Not sure what it means yet. I am less certain that this will work, but I do believe an inductor is needed if it is to work.

 

[duivendyk]

Don't give up yet,although the black wire is unlikely to be a real power house,it may still provide a usable dc output to charge a battery with. As I have mentioned before the dc readings after the diode wouldn't look all that great because the meter reads averages and the pulses are pretty short,so the peak to average ratio is quite high.For instance if you had a rectangular dc pulse with a 10% duty factor and you read a 2V average,the peak value would be 20V, enough voltage to supply current to a 12V battery, how much that remains to be determined.
One question about the readings, did the polarities of the dc outputs made sense,I mean when you had a negative going rectifier the reading was negative (you don't mention the polarities of the measured voltages to ground).Another more important question, when you refer to a bridge rectifier what exactly are you talking about,a 4 diode full wave bridge circuit?,or a 2 diode bridge circuit.In either case you cannot consider the output with respect to ground the actual dc output !.In a 4 diode full wave rectifier circuit for instance,either the input has to be floating (both ac input wires are NOT connected to ground). In that case one side of the output,of the dc. side that is,may be grounded, if you should elect to do so,but if the input has one side grounded,like in our particular case with the black wire, then the output HAS to float,grounding one side results in a short circuit.So any dc measurements with respect to ground are not meaningful.Please tell me what your rectifier circuit looked like (preferably some kind of schematic).I would recommend that you take a look at www.pilom.com This deals with b... that would be jumping the gun at this stage.

 

[skyl4rk]

I spent some time today experimenting and measuring. The most important thing that I found out is that my el cheapo voltmeter battery apparently loses power as it gets cold, so my measurements may be flawed. There were some measurements that I was getting different results on, now I know why, the voltmeter got cold.

Regarding the bridge rectifier: The bridge rectifier is a four diode rectifier, with the ac input from the motor floating and not connected to the dc output ground. I used four diodes in a breadboard, and later I also tried out a rectifier in a case, a part that I had for another project. Oddly, the four diode system worked fine, but the rectifier in a case always shut down the motor when I made the dc connection to measure voltage. Using the four diode bridge rectifier, the dc output was 14V at idle and 24 volts at higher rpm.

A schematic for a diode bridge rectifier:
http://upload.wikimedia.org/wikipedia/commons/1/13/4_diodes_bridge_rectifier.jpg
from: http://en.wikipedia.org/wiki/Diode_bridge

The rectifier in a case looked like this:
http://mouser.com/Search/ProductDetail.aspx?qs=sGAEpiMZZMtcLAek5QF0i9Td9SJjjJn6hPywd2%2bgQew=

The diode bridge rectifier was just four diodes stuck in a bread board. As far as the schematic, I don't have a way to draw one, but there are four diodes connected in a box shape. The motor wire (ac) and motor ground (ac) both have one diode pointing towards them and one diode pointing away from them. The dc positive output has two diodes pointing toward it. The dc negative has two diodes pointing away from it.

Some other interesting experiments:

Inductor: I wound a coil, 234 turns of 22 gauge enamel coated magnet wire around a bolt (40 ft of wire). I put this in line with the positive output and did some measurements. There were no significant measurable differences between the voltage with or without the inductor in line. However, when I tried the rectifier in a case, it killed the motor when hooked up to make a dc measurement on the dc output, but when I put the inductor in line with the positive dc output, it allowed the motor to run.

Battery: I connected a battery to the dc output side of the diode bridge rectifier, pos to pos, neg to neg. This was a small 12V lead acid battery, 12 aH, the size that is typically used for electric bikes. The battery was cold and had not been charged for months. With the battery in place, the voltage stayed stable on the battery side of the circuit. The battery started out at 12.7. After reving the motor up for a few minutes, I saw the battery move to 12.8V. This may be coincidence but it may mean that a small charge was added to the battery.

When I tried to connect the battery using the rectifier in a case, connecting the battery killed the running motor. I put the inductor between the battery and the pos dc output of the rectifier in a case and then the motor would start and run.

Light bulb: I tried a 12V light bulb, connecting it to the diode rectifier dc output. There was no visible glow from the filament. However, at this point I was getting cold and it was daylight, so maybe I did not see a glow or made a mistake connecting it up.

 

[duivendyk]

OK you have a floating input and a bonafide bridge rectifier.A battery that's been sittting around for months may be sickly or have expired.The voltage may be OK but it has as high resistance which means that it's not much of a load.If you increase the voltage above 14V to charge the thing but the current stays low, a few 10th of an Amp or so,it may have sulfated.If you have a variable dc supply or a trickle charger they can sometimes be rejuvenated by trickle charging them for a few days.Can you measure current ?.It would be nice to be able check the charge current.To check a battery :Use a 12 V tail light bulb as a load,the voltage should not drop more than 0.2-0.3V.Otherwise the battery is shot or discharged
It does not surprise me at all that your attempt at coming up with a homebrew inductor did not show any difference.To begin with the frequencies involved are quite low,50Hz corresponds to 6000rpm! (4 stroke engine).At that speed 1 milliHenry inductance is only
0.314 Ohms.I doubt that you have much more than that.The reason is that you have basically an air coil.That bolt is only a small fraction of the magnetic circuit ,the rest is air,with unity permeability.So the magnetic flux is still low as is the inductance,furthermore the shape factor is also bad,you need a sort of squarish coil shape, not a long sausage shape,to get the highest inductance for a certain dc resistance.I think you could more profitably spend you time finding out what dc load you can stand at the rectifier output.A 1k pot (2 watts or so rating) would be useful.See how low a resistance the engine will stand and also measure the dc out.There are better inductors to be had,but that is really premature at this juncture.

 

[loquin]

Plus, in the long run, that bride rectifier is a huge overkill - it's rated at 25 amps.

A 2 amp bridge rectifier would be fine, and, it drops less voltage across it. However, it would be best to locate 4 individual schottky diodes, as they only drop about .3 volts.

In a bridge diode circuit, you'll lose 2 times the forward voltage drop in a bridge rectifier. Your big diode bridge is rated at a 1.1 volt drop at 12A. At lower currents, silicon diodes usually drop about .6 - .7 volts. This means that if you have 14 volts available, you'll lose nearly 10 percent of the voltage (and therefore, 10% of the available power,) just by using silicon diodes. Schottky diodes would cut those diode losses in half.