The post discusses a list of PWM controlled simple 3 phase motorcycle voltage regulator circuit which may be used for controlling the battery charging voltage in most two wheeler. The idea was requested by Mr. Junior.
hello my name is junior live in Brazil and work with manufacturing and recovery regulator rectifier motorcycle voltage and would appreciate a help u, I need a three-phase mosfet regulator circuit for motorcycles, entreda voltage 80-150 volts, correte Maximum 25A, maximum consumption of the system 300 watts,
I await return
The proposed 3 phase motorcycle voltage regulator circuit for motorcycle may be witnessed in the diagram below.
The schematic is rather easy to understand.
The 3 phase output from the alternator is sequentially applied across three power transistors which basically act like shunting devices for the alternator current.
As we all that while operating, an alternator winding could get subjected to huge reverse EMFs, to an extent which could get rip of the insulation cover of the winding destroying it permanently.
Regulating the alternator potential through the method of shunting or shorting to ground helps to keep the alternator potential under control without causing adverse effects in it.
The timing of the shunting period is crucial here and directly influences the magnitude of current that may finally reach the rectifier and the battery under charge.
A very simple way of controlling the shunting time period is by controlling the conduction of the three BJTs connected across the 3 winding of the alternator, as shown in the diagram.
Mosfets could also be used instead of the BJTs, but could be mush costlier than the BJTs.
The method is implemented by using a simple 555 IC PWM circuit.
The variable PWM output from pin3 of the IC is applied across the bases of the BJTs which in turn are forced to conduct in a controlled manner depending upon the PWM duty cycle.
The associated pot with the IC 555 circuit is appropriately adjust for obtaining the correct average RMS voltage for the battery in charge.
The method shown in the 3 phase motorcycle voltage regulator circuit using mosfets can be equally implemented for single alternators for getting identical results.
Peak voltage adjustment
A peak voltage regulation feature may be included in the above circuit as per the following diagram, in order to maintain a safe charging voltage level for the connected battery.
As can be seen, the ground line of the IC 555 is switched by the NPN BC547 whose base is controlled by the peak voltage from the alternator.
When the peak voltage exceeds 15 V, the BC547 conducts and activates the IC 555 PWM circuitry.
The MOSFET now conducts and begin shunting the excess voltage from the alternator to ground, at a rate determined by the PWM duty cycle.
The process prevents the alternator voltage exceeding above this threshold, thus ensuring that the battery is never over charged.
Motorcycle Battery Charging System
The second design presented below is a Rectifier plus Regulator for a 3-Phase charging system of Motorcycles. The rectifier is full-wave and the regulator is shunt-type regulator.
By: Abu Hafss
A motorcycle's charging system is different from that on cars. The voltage alternator or generator on cars are electro-magnet type which are quite easy to regulate. Whereas, the generators on motorcycles are permanent magnet type.
The voltage output of an alternator is directly proportional to the RPM i.e. at high RPM the alternator will produce high voltages more than 50V hence, a regulator becomes essential to protect the entire electrical system and the battery too.
Some small bikes and 3-wheelers which do not run at high speeds, only have 6 diodes (D6-D11) to perform full-wave rectification. They don't need regulation but those diodes are high ampere rated and dissipate a lot of heat during operation.
In bikes with proper regulated charging systems, normally shunt-type regulation is used. This is done by shorting out the alternator's windings for one cycle of the AC waveform. An SCR or sometimes a transistor is used as shunting device in each phase.
The network C1, R1, R2, ZD1, D1 and D2 forms the voltage detection circuit, and it is designed to trigger at about 14.4 volts. As soon as charging system passes this threshold voltage, T1 starts conducting.
This sends current to each gate of the three SCRs S1, S2 and S3, via current limiting resistors R3, R5 and R7. D3, D4 and D5 are important to isolate the gates from each other. R4, R6 and R8 help in draining any possible leakage from T1. S1, S2 & S3 should be heat-sinked and isolated from each other using mica insulator, if using common heat-sink.
For the rectifier, there are three options:
a) Six automotive diodes
b) One 3-phase rectifier
c) Two bridge rectifiers
All must be rated at least 15A and heat-sinked.
The automotive diodes are two types positive body or negative body hence, should be used accordingly. But they might be little difficult to contact to heat-sink.
Using Two Bridge Rectifiers
If using two bridge rectifiers, they may be used as shown.
Efficient Battery Charging through Motorcycle Shunt Regulation
The following email conversation between Mr Leoneard, an avid researchers/engineer and me, helps us to learn some very interesting facts regarding motorcycle shunt regulator drawbacks and limitations. It also helps us to know how to upgrade the concept simply into an effective yet cheap design.
You have an interesting circuit, but.....
My motorcycle has a 30 amp alternator, which I'm sure is RMS, and peaks at 43.2 Amps. Your 25 Amp circuit is not likely to hold up long at all.
In place of the rectifiers you suggest, an SQL50A is rated 50 Amps at 1,000 Volts. It is a 3-phase rectifier module, and should have no problem handling 45 amps peak. (I have two on hand.)
That also means the SCRs will have to handle that Amperage and three HS4040NAQ2 with RMS current of 40 Amps (non-repetitive surge to 520 Amps) should handle that quite well. Of course, they'll require a pretty healthy heatsink, and good air flow.
I'm thinking the control circuit should work pretty much as is.
I've replaced 3 regulators in the last three months and I'm about tried of throwing good money after bad. The last one lasted a total of ten seconds before it went bad too. I'm about to build my own and if I have to build it to power a battleship, so be it.
Another thing I've noticed, the laminations used in the alternator are considerably thicker than those used in electric motors. An 18-pole winding, and engine operating at highway speeds means much higher frequency, and far more eddy currents in the iron. What would be the effect on those eddy currents if using a series regulator that would allow the voltage to go as high as 70 Volts (RMS)? Would this increase the eddy currents to the point of overheating the iron, and risk damage to the windings of the alternator? If so, it would make sense not to allow the voltage to get above 14 Volts, but I still have 20 Amps coming from the alternator at 1500 RPM.
Thank you! Yes you must get rid of that high voltage which might put huge pressure on the alternator winding, the best way is to shunt it through heavy duty MOSFETs on heatsink
Actually, I'm not nearly as concerned about the effects of voltage on the windings. They appear to be coated with Poly-Armor Vinyl, which is also used in random wound stators operating at 480 Volts. I'm far more concerned about the heat from the eddy currents in the laminations, since they are so thick. Here in the States, with 60 htz line current, the thickness of motor laminations are a fraction of what they are in the alternator. At road speed, the frequency from the alternator can be 1.2 Khtz or higher. In other applications, that would call for a ferrite core to eliminate the eddy currents.
I'm trying to understand the role of eddy currents in this application. As RPM increases, so does the frequency, and the eddy currents as well. A parasitic load to level off the voltage generated? A means of leveling off current generated at high RPM? Just how much heat does that generate? Enough to burn out the winding at high RPM?
Located inside the engine, I can understand using engine oil to cool the assembly, however, with the centrifugal force of the flywheel, and the windings located inside that, I can't imagine any real amount of oil getting to them for cooling.
The highest voltage I've been able to read is 70 Volts RMS. That's not enough to arc through PAV coating on the wire, unless heat becomes excessive. However, in shunting the excess to ground, Is there a counter EMF that opposes the magnetic field from the rotating magnets? And if so, how effective is it?
Yes, increase in frequency will give rise to more eddy current in an iron based core, and an increase in heat.I have read that shunt control method is good for motor based generators, but this will also mean increased load on the alternator wheel and more fuel consumption by the vehicle.Is fan cooling an option? the current to the fan can be accessed from the alternator itself.
I'm afraid that a cooling fan is not an option for the alternator. That is mounted internal, inside the engine, and on my Vulcan, there are two aluminum covers over that.(Replacing the alternator winding means removing the engine from the motorcycle.) I do not see any way of reducing the eddy currents because they are induced by the magnets rotating inside the flywheel. However, I can reduce the current shunted to ground by raising the voltage of the shunt to 24 Volts, and following that with a series regulator set to 14 Volts. In testing the alternator, I do not see much effect from counter EMF in reducing short-circuit current. I can load the alternator to 30 Amps, and by shorting the leads, I still read 29 Amps.
However, if using the eddy currents as a parasitic load to level off the voltage and current at high RPM, it seems to be quite effective. Once the open circuit voltage reaches 70 Volts (RMS), it does not go higher even when engine RPM doubles. Shunting 20 Amps to ground (as done by factory regulators), increases the heat in the winding in addition to the eddy currents. By reducing the current through the windings, the heat generated by the windings should also be reduced. That won't reduce the eddy currents, but should reduce the overall heat generated by the alternator, hopefully preserving the winding insulation.
Considering the coating on the windings, I'm not nearly as concerned about the voltage generated. Having worked in electric motor rebuilding for years, I'm aware that HEAT is the worst enemy of the insulation. The quality of the insulation is reduced as operating temperature increases. At ambient temperature, PAV coating can hold 100 Volts "turn-to-turn". But raise that temperature by 100 C, and it may not.
I'm also curious. Electric motors use a steel alloy with 3% silicon to reduce the resistance to magnetic field reversal within the iron. Do they include that in their laminations or omit the silicon to further reduce the increase of voltage and current at high RPM? It does not add to the heat, but does reduce the efficiency of the iron, the higher the RPM. By increasing the resistance to magnetic field reversal in the core, the magnetic field may not penetrate as deeply into the core before it is required to reverse. So, the higher the RPM, the less penetration by the magnetic field. The eddy currents may further reduce that penetration.
Your analysis makes sense and appears very much technically sound. Being basically an electronics guy, my electrical knowledge is not very good, so suggesting motor internal working and modifications can be difficult for me. But, as you said in your last sentences by restricting the magnetic filed, the eddy current can be prevented from entering deep. I tried searching about this issue but couldn't find anything useful so far!
So, having worked with electric motors for 13 years, I have you at a slight disadvantage? Although, my studies have also been with electronics, and so was all my work until I found I could make more money working with motors. That also meant I did not keep up with integrated circuits, and MOSFETs were delicate little things that could quickly be blown out with the slightest static charge. So when it comes to electronics, you have me at a disadvantage. I was not able to keep up with new developments.
It's interesting that I have not been able to find much of my information in one place. Sort of as though none of the concepts are related to each other. Yet, when putting them all together, they begin to make sense. The higher the frequency, the less turns are required to get the same inductive reactance. So the higher the RPM, the less effective the magnetic field becomes. It's about the only way they can keep the output constant once the output reaches 70 volts.
But in looking at the pattern on an oscilloscope, I'm not impressed. A millisecond of charge time, followed by 6 to 8 millisecond of grounded output. Could this be why motorcycle batteries don't last long? Six months to a year, while automotive batteries go on for five years or more. This is why I'm opting to "clip" the voltage level to ground at a higher voltage, and that clipping being constant. Followed by a series regulator to maintain a constant charge rate according to what the battery, lights, and circuits require. Then by designing it to handle 50 Amps, I should never have to replace a regulator again.
I'm working with a 50 Amp rating, but I expect that by using a "clipper" the Amperage should be considerably lower than 20 Amps to ground. Perhaps as low as four Amps. Then the series regulator allows the (approximately) seven Amps for the battery, lights, and circuits for the engine. All well within the wattage rating of the components and not enough voltage to challenge the coating of the windings.
You wrote a very good article about shunt regulators, but 25 Amps is just too small for my application. Still, it is good inspiration.
Yes that's right, duty cycle of 1/6 will not charge a battery properly. But this can be easily solved through a bridge rectifier and a large filter capacitor, which will ensure that the battery gets enough DC for effective charging.I am glad liked my article. However the 25 Amp limit can be easily upgraded by increasing the MOSFET amp specs. Or may be by adding more devices in parallel.
At the same time, I'm trying to keep everything compact to fit into room available, so that large filter capacitor capacitor becomes a problem. It is also not needed if all three phases are clipped after the bridge rectifier. All ripple is clipped off, and the series regulator maintains 100% charge time.
Your circuit also maintains 100% charge time, however the current you shunt to ground will be much higher because you are clipping it at battery voltage.
As you can see in the waveforms, there should be no capacitor needed. But by clipping at a higher level, the current shunted to ground should be lower. Then, dropping the voltage across a series regulator should not hurt anything. There should be more than enough to keep the battery charged.
One note. Optimum charge voltage for a lead/acid battery is actually 13.7 volts. To hold it at 12 volts may not give the battery enough to start the engine. And my circuit is preliminary, and still subject to change.
The factory looks almost primitive, in the way it works. Their circuit charges the battery until it reaches the trigger level. then it shunts all current to ground until the battery drops below the trigger level. The result is a waveform with a short, harsh burst of charge that could be as high as 15 Amps. (I did not measure it) That followed by a longer line with a slight downward slope, and another burst.
I've seen automotive batteries last 5 to 10 years, or longer. As a kid on a farm, my father converted one of the old tractors from six volts to a twelve volt system, using an alternator from a car. Fifteen years later, that same battery was still starting the tractor. At the school I work with (Teaches motorcycle safety), all batteries need to be replaced within one year. WHY ? ? ? The only thing I've been able to come up with is the charging system. Most of the batteries I've worked with are only rated for a 2 Amp charge rate, Up to 70 volts, capable of 30 Amps, applied to the battery terminals for short bursts may be causing internal damage and shortening the life of the battery. Especially, in the batteries where you can not check the fluid levels. The only problem with the battery may be fluid level, but there is nothing you can do about it. If I'm able to check and maintain fluid levels, the battery life is extended considerably.
The leads coming from the alternator would be the metric equivalent of #16. According to the AWG table, that's good for 3.7 Amps as a transmission line, and 22 Amps in chassis wiring. On a 30 Amp alternator with a shunt regulator? The shunt level and the Amperage should be an inverse proportion, so by clipping the voltage in half, I should reduce the Amperage significantly. In looking at the rectified waveform, the highest concentration of EMF is in the lower half. Logic would suggest the current will be reduced to a fraction. I'll find out when I put it into use.
On a 1500cc engine, I don't expect to notice the reduced drag on the engine, but my fuel economy may improve. And, I remember, back when they first started putting solid-state regulators on automotive alternators, the magic number was 13.7 Volts. However, I was planning on setting my series regulator at about 14.2 Volts. Too high and the fluid evaporates more quickly. You were far more helpful than you know. Originally, I had six different circuits that I was considering and was going to breadboard each of them. Your article eliminated five of them, so I get to save considerable time and concentrate on just one. That saves me a good amount of work. That makes it very well worth the time to contact you.
You have my permission to experiment with my schematic and see what you come up with. On various forums, I'm reading where a number of people are talking about going to series regulators.Others caution against too high a voltage destroying the insulated coating on the wire. I suspect the happy medium may be a combination of both systems, but not shunting the full output to ground. The circuit is still simple, with few components, but not archaic.
Thank you very much for your time and attention. One of my sources for technical information is: OCW.MIT.EDU I've been doing engineering courses there for a few years now. You don't get any credit for doing them, but it's also completely free.