PD: Primary Drive Make your own Generator

oexing

Well Known and Active Forum User
VOC Member
Myself having no experience in stepper motors I´d think you might try one for generator use. You got one with four wires ? So you would get two phases from two sets of field coils. Basically a stepper motor is designed - well, for steps, so the feel you get when spinning the rotor is not so smooth compared to a nine pole ac brushless as in my tests. This is no factor for the bearings as they are designed for it - and the load onto the primary chain is negligible. The motor will draw half an hp at best and the chain has to handle 50 , 60, 100 hp any day. The "power" required to drive the dynamo/alternator from the middle rollers is simply nothing , at maybe one quarter or half of a hp really, no issue at all, never ! Sorry to repeat myself, the root of all troubles in the primary drive is the Elastic Shock Generator - definitely .
I guess, even with the old "high power" 6 V Miller 50 W dynamo owners will have seen worn middle rollers in the triplex chain - and you are not saying it was too much power that the Miller had asked from the chain ?? Really the single factor in all that drama is the mass/inertia from the armature/rotor in the generator that f***s up the chain rollers. The impulse for the shocks from inertia comes from the triplex chain when the ESA slams onto its stops because the shape of the lobes is completely wrong for the purpose and no progressive and smoother action is designed into them. So any shocks get into the triplex chain and along it into the rotating armature that wants to keep its speed against the harsh acting chain. It is easy to see that a lighter rotor will be less of an issue than a heavy type, the wattage not being ANY factor at all. No surprise for me that a geared Alton will never stand a decent mileage with its tiny gear teeth inside,lubrication just one aspect. But really any harshness from the ESA compromised triplex chain will be multiplied by the ratio in the Alton and observed in broken bits inside. So in one way a generator at the old magneto place is a better proposition than hoping for survival in the ESA primary environment.
The best way to overcome this dilemma is to throw out the ESA or weld it up. No ESA is a lot milder on the chain than with it. There have been lots of broken chains and engine cases in the event just because of this very flawed ESA, even though the chain without it would transmit a lot more power than most engines got. In my photos below you can see a duplex chain from a Horex 400 cc, 460 in my case - next to the triplex. The added cross sections of all links are a tad more in the duplex than in the triplex. But the rollers are metric at 6mm , 1/4 or 6,35 mm in the triplex. Yet there have never been broken Horex chains in a production run of more than 50 000 in 5 years - and that with no ESA there and SPLIT rollers in them. Just some poor rubber elements in the clutch sprocket included. There are photos of the Horex gear box contents next to the very strong Vincent gears, so what to worry about throwing out the horrible ESA , never a problem with these gears, I say. A big Vincent is just two 500 cc singles flying in close formation, after all.
Brian, don´t worry, be happy - with your belt drive and the Alton ! You will never have a problem with the alternator, non geared, i guess. Your drive has no longer the ESA so a smooth primary in your bike. But certainly any load on a dynamo chain sprocket or toothed belt gear must be prevented for a silent operation.
Don´t know much about Vespas but I guess the Hegeler sytem got its design from a starter motor planetary gear like in my photo. BUT this is a REDUCTION gear , not for speeding up a rotor. And in this role that will put quite a bit higher load onto timing gears as you can test by spinning said alternator via the planetary gear in its role as multiplier. Myself, I would not like this - nor would I like another gearbox on the alternator to care for . I got something similar on my twenties BMW in the Bosch type of magdyno - and I hate it. No great lubrication in there possible and a bit noisy too. The armature spins about five times faster than the input from the timing side, at just 30 W 6 V output.

Vic

P1070018.JPG


P1070023.JPG


P1070027.JPG


P1070037.JPG



P1030759.JPG
 

vibrac

Well Known and Active Forum User
VOC Member
well I only know after our closed road lap of the TT circuit in '99 my new triplex chain had dropped all its central run rollers in the bottom of the chain case of course I may have had one of those dreaded welded roller chains but that was before the trouble with them had been highlighted (at least to me)
 

timetraveller

Well Known and Active Forum User
VOC Member
Perhaps I can throw some more dirt into the already muddied waters of this discussion. I have used stepper motors in two different telescope drive systems, one weighing 4.5 tonnes and one built in York in 1860 with a 15 foot long tube and a large amount of cast iron and bronze, with great success, but the electronics were done by specialists so what knowledge I have is peripheral to the fundamentals of these motors. First, for those who have no knowledge at all, a telescope has to be driven to follow the stars across the sky. The drive system should show no 'granularity' or otherwise any star images taken by photograph or CCDs will be widened by this 'granularity of the drive. Typically the stepper motors used had 200 steps per 360 degrees of rotation, which is much too coarse so they are used in what is called 'micro-stepping' mode wherein each step is divided into a further 64 micro steps. This will give a step size on the motor of 0.028125 degrees. Much too coarse, so then a further mechanical gear reduction of 1:100, or more, is required to get the less than one arc second of granularity required. What might be relevant in the present context is that one has to ramp up and ramp down the speed of movement, or the motor will stall. In addition one cannot try to get very high revs as the torque reduces to the extent that the motor will cease to rotate at all. Used as a motor, there are severe limits as to the speed at which these motors can rotate and still respond the their input pulses.

I am a great admirer of people who put their money where their mouths are. Had John Emmanuel and Dave Dunfey not put their time and effort in to fork modifications we would not have had the improved steering heads. So I am very pleased that people are now trying to find a way of using motors as generators. I hope I am wrong but I have doubts about stepper motors being able to generate significant power and I still have a niggle in the back of my head from school days about the phase of the windings and stators having to be different when the system is used as a motor or a generator. We will see.
 

oexing

Well Known and Active Forum User
VOC Member
Sorry, Norman, to disagree regarding motors or generators, no essential difference. Don´t use starter motors for reference, these are an extremely specialized application for hundreds of amps and breaking down supply voltage when activated in cars or bikes.
You can test the good old car DC dynamos for function by hooking a battery to them: They will run as motors - but not overly strong at that. So you can use DC motors for dynamos, like for small wind generators . Same goes for brushless motors as alternators. You just have to hook a three phase motorcycle regulator on them to get DC for your battery. You mentioned your considerations with some Honda permanent magnet alternators, millions of this type are daily assembled on current models on about all makes. Just looking at Ducati alternators, they had both types, the magnetic rotor inside, the coils around it. And the other more current type with static coils inside and the magnets in a kind of flywheel around the coils. But the principle is the same : A rotating magnet produces the field wherein coils generate the voltage, no brushes required, simple, even suitable for in oil operation. An Alton does just the same, exactly like my China brushless, no difference at all: A small magnet rotor is driven, the voltage is generated in the coils around it. Same three phases are controlled by a motorbike regulator, millions are sold worldwide in all sorts of bikes. And in case of my 68 mm o.d. motors I guess they might output a bit more than the non-geared Alton as that is limited in motor size by the 1/2 " shaft offset to be fitted without machining the engine case. I honestly see more than 11 V at 100 W quartz bulb at around 2000 crank rpm - no discussion on that, I swear ! But then, maybe the separate cradle of the later type Vincent could be modified even easier for a China brushless. A saving of € 500.- on an Alton is worthwhile for me to fabricate my mod.
Vibrac, did you have the standard ESA in your bike - plus a dynamo ??? Don´t say your dynamo had more than 20 hp for destroying the central rollers ! I think we will agree the whole triplex chain will handle min. 60 hp, even 100 hp or more ? So one row of rollers should be allright for minmum 20 hp ??? So in your case we will agree that the power of the dynamo cannot possibly be the root of the problem ?? Now my conclusion it is still the shockwave from the bashing Elastic Shock Generator that reaches the dynamo sprocket with its spinning rotor/armature mass and the inertia of the rotor mass overloads the rollers - not the pure smooth load from the generated current/voltage . The split rollers like in the duplex Horex chain are no factor at all , never were in a production of more than 50 000 bikes.
As we all know by now, these shock waves are dramatic enough to even snap the triplex chain so no wonder some rollers get cracked from that rotor inertia load.

Vic

Ducati rotor inside type:
Ducati rotor

Ducati flywheel rotor outside:
Duc flywheel type
 

erik

Well Known and Active Forum User
VOC Member
If we compare the Generators of the same type mounted in twins or Singles ,we should see the influence of the driving gears on their reliebility !?On my comet I had no Problems with the Alton since 8000km. My twin is not on the road ,yet. Erik
 

timetraveller

Well Known and Active Forum User
VOC Member
No problem with a disagreement Vic. It is only by exchanging ideas and experiences that progress is made. I wish you well with your experiments. However, I do have one problem concerning your ideas about the Vincent ESA shortening the life of the chain. I am not going to defend the design of the ESA but it is worth noting that when John Emmanuel used a stroboscope to investigate the working of the ESA he found that it moved rapidly up to the maximum travel in one direction when under load and then immediately to maximum movement in the other direction when on the overrun. The larger number of springs used in the Australian design of ESA had a different effect, discovered by Dick Sherwin. He used that ESA on his oversized twin with a sidecar fitted. Later dismantling showed that only a small amount of the potential movement was being used despite the extra power from the engine and the extra load of the heavy sidecar. The solution was to remove some of the springs, after which, most of the potential movement was being used. It is entirely possible that the design of ESA which you like would be better but the Vincent design can work with a little tweaking.
 

Bill Thomas

Well Known and Active Forum User
VOC Member
Norman, I had mine, For the Comet, Off you about 15 years ago, It's been under the bench !, Is that the new or old type ?, I don't think I will be going far !. Cheers Bill.
 

timetraveller

Well Known and Active Forum User
VOC Member
Hi Bill, that is the old type. The main difference is that it uses a 30 amp Iskra alternator, which requires an external regulator. The new ones use a 40 amp Nippon Denso alternator which has an inbuilt regulator. The new one is about one centimetre smaller diameter and one centimetre longer compared with the earlier one.
There was nothing wrong with the Iskras. Indeed they have a slot for a key in the input shaft which the N,D. ones do not have. However, one of the main reasons for using the Iskras was that they were intended for use on Citroen 2CV and other small cars and at the time of development Dick Sherwin and I felt that there would be some merit in using an alternator which could be found in almost any scrap yard in France, and probably other European countries, if ever there was a break down. Things have, of course, moved on. 2 CVs are now rare and in addition the N.D. small alternators are used on many small tractors, kit cars and so on. It was Peter Holmes (wife Sue with the ruptured Achilles tendon) who first got me to use a ND alternator. He had access to one and wanted it fitted to his bike. It has proved reliable over many thousands of miles.
Wherever possible I have used parts from the old kit to mount the new alternators and this has led to some delays in supplying them.
In the context of extra wear on the primary chain drive and Vic's comments about destruction of the primary chain. It might be wondered what happens when one uses either a 30 amp or 40 amp alternator. There has never been a failure yet. There are several things to be aware of. First the chance of using the maximum output from these alternators is minimal. When starting the bike the ammeter will go fully over until the battery comes up to 14.3 volts. After a few seconds this reduces to zero after which the electrical load is minimal. Note that this is done by reducing the current sent to the field windings of the alternator so that power is not being generated rather than producing the power all the time and then bleeding it off to an earth. My design philosophy was, rather like the Vincent twin engine, more power than you can possibly use or need apart from rare occasions. I also deliberately designed the system to use a multi vee belt drive. These are used on everything from washing machines, lawn mowers and alternator drive systems on cars. They have the capability to allow some very small amounts of slip which can be used as a shock absorber between the Vincent twin primary drive and the alternator. Of course the bike does not look standard but beauty lies in the eye of the beholder and nothing of the original bike has to be modified to fit the system.
I have tried to make that a design feature of all of the parts I have supplied for Vincents. These include the modification for the lift mechanism for the twin clutch, the new steering heads, the hydraulic steering dampers for both the new steering heads and the original, although with the latter one does have to tap out the hole which originally takes the anchor pin for the original friction steering damper.
 

Bill Thomas

Well Known and Active Forum User
VOC Member
For me, When the middle chain rollers break up, Apart from when there was some bad chains supplied,
It's because the sprocket is too far in mesh, You must also allow for a bit of slop in the chain wheel bushes,
When ever there is a bunch of Vin's around, You can always hear one or two, That are too tight.
Cheers Bill.
 

stu spalding

Well Known and Active Forum User
VOC Member
With regard to armature weights I have been running a Fiat 500 dynamo for 30 years. The Fiat armature weighs 3lb 3oz compared to 2lb for the Miller. When I first looked into using a Fiat unit I was warned that the chain would not stand the extra weight but in all that time the only trouble I've had with centre row rollers was with the rubbish Eurochain, at least I think that's what it was called, that was being put about in the late '90s. I've had no trouble at all with Reynolds and Tsubaki chains. Cheers, Stu.
 
Top