• If you like antique engines, vintage tractors or old iron machinery, register and join us. When you register on Smokstak, please give complete answers and fill in all blanks. IF YOU ARE ON WIRELESS OR SATELLITE, GIVE YOUR CITY AND STATE! NO ZIPCODES! All registrations are manually approved.

1800 rpm Onan with 3600 rpm generator

Graycenphil

Registered
Last Subscription Date
02/16/2012
My Onan MDJE generator head died a few months ago, and I think I will run another generator head off of it, with a belt drive. 3600 rpm units seem much more plentiful, so here's my question:

Will putting a small pulley on the Onan shaft, and another pullley twice the size on the generator work right? I think that will give me the right rpm; will the governor operate properly, keeping the volts and hertz where they should be?

Thanks.
 

Power

Registered
You got it reversed! Big one goes on engine, but it is a tapered shaft, don't know how you can mount pulley unless you machine engine shaft and forever destroy it.
 

Roland Hayes

Registered
Last Subscription Date
01/11/2016
The taper will probably be a standard 5 degree, its easy to bore a taper into a pully on a lathe and with a shallow 5 degree taper there is no need for a key.
 

Graycenphil

Registered
Last Subscription Date
02/16/2012
Thanks. That's why I ask questions here. Is it as simple as using one pulley twice the diameter as the other, or is there a more complicated calculation?

I had thought about using the generator shaft, which isn't tapered, and putting the pulley on the end of it, but I like the idea of milling the pulley better.
 

Onan Dan

Subscriber
Last Subscription Date
03/28/2020
Got into that last year tried to run a 3600 generator head off my 1800rpm onan bge engine. had to reconfigure the shaft from a tapered head to a 3/4inch shaft by 4 inches long on the onan engine. well long story short used the big pulley 6in dia on the onan eng and the 3 in pulley on the gen head just the opposite your doing. here whats happened one pulley or belt will not run the setup. you will need three or 4 row gang pulleys to stop the friction . any way I wasted my time and money . luck onan dan:bonk:
 

Roland Hayes

Registered
Last Subscription Date
01/11/2016
As you are using pulleys, there is no need for the engine to run at any particular exact speed, the alternator of course must in order to get the correct Hz frequency, my suggestion is to run the engine at 2200 rpm which is not that much higher than 1800 but at that speed the engine will be up into the fattest part of its torque curve and make more power, it might even be more fuel efficient at 2200 than at 1800. All you need to do to prevent belt slip is to use big diameter pulleys, use a 10" dia on the engine and a 6" dia on the alternator and they should be twin groove with cogged V belts.

This is a generator that I built a few years ago, on this one the engine could be adjusted to run at between 750 and 1200 rpm, the 3000 rpm Knylor alternator has permanent magnet excitation and was chosen because it had good voltage control over a wide rpm range, for low power such as over night security lighting to reduce noise and fuel consumption the engine speed would be reduced and for maximum output the speed increased. the correct UK 50Hz occurred at about 1000 engine rpm
 

Attachments

Last edited:

JerryLee

Registered
Last Subscription Date
02/01/2016
Is it as simple as using one pulley twice the diameter as the other, or is there a more complicated calculation?

Seems to me that the circumference is the key dimension. If you run the engine at 1800 RPM and you want the generator to be at 3600 RPM, the circumference must be half on the generator as it is on the engine. You can find an easy conversion formula on the internet. Just do a search on "circumference formula" and see what you get.

Try to find a wide poly-V belt. Any decent machine shop can make both pulleys for you without breaking the bank.
 

Jim Rankin

Registered
Age
58
The best type belt drive is a cogged belt so no slippage and less tension required, so less side load on the engine. Not sure what kind of main bearing your engine has, but belt driving something vs direct driving it puts a very different load on the crankshaft bearings. Even though the cogged belt will need less initial tension than a V-belt, it still pulls the shaft sideways because it must transmit torque through a 90* turn. Mount the pulley as close to the engine as possible to lessen the strain on the crankshaft.

A ball or roller bearing can stand the side load a lot better than a babbit or aluminum sleeve bearing, so the 2 bearing alternator is readily able to stand it, but the engine may not do so well long term if it doesn't also have that kind of bearing.

If you use a cogged belt, you simply count teeth and pick 2 pulleys that give you the ratio you need and a load capacity which exceeds your generator's capacity. An internal combustion engine as a driver requires a higher "duty" factor (multiplier really) than a similar capacity electric motor because of the pulses of torque as each cylinder fires.

If you use v belts or micro vee belts, you select the pulleys based on rpm, and load and end up with 2 that have a "pitch diameter" ratio that result in the correct speed. Pitch diameter is related to outside diameter of the pulleys, but it's always a little less than the OD. So you can get close by using OD, but look up the PD and you can get it closer.
 

len k

Subscriber
Last Subscription Date
12/12/2018
My recomendation is to go with a cog belt not v-belts. V-belts have ~1% slip and will need more tension to transmit the torque. This tension will cause a little more wear on the shaft support bearings. There are standard formulas for sizing pully diameter to hp transmitted, making diameters large enough for a given belt size. An industrial supply house or belt manufacture's website should have that info. I wouldn't worry too much about pully "pitch diameters", you will have to use standard sides and can tweek engine rpm to maintain gen rpm. Pullys are not that cheap, maybe another used genhead??

typo.....I wouldn't worry too much about pully "pitch diameters", you will have to use standard sizes
 

ArodaPowerCo

Subscriber
Last Subscription Date
09/22/2019
I would be very careful belt driving a generator head with an Onan twin. From personal experience, Onan opposed twins do not like to have side-loaded crankshafts. The bearings are the least of your worries. Theses engines have two throws on the crank so that both pistons are at TDC or BDC at the same time. This creates a weakness due to the crank being shaped like a "Z". Too much belt tension for too long and the crankshaft WILL break. I used to maintain a fleet of stump grinders, and I couldn't keep Onans on them for longer than 1000 hours. Kohler CH25 V-twins (only one crank throw) would in many cases exceed 4000 hours (with an overhaul or two) before they were no longer rebuildable.

To alleviate the need for lots of belt tension with your Onan, use very large diameter two-groove pulleys to increase the belt gripping surface area. Like 12" on the engine and 6" or so on the gen head. This will significantly reduce the tension needed to acheive satisfactory belt friction, and if you use cast iron pulleys, add a lot of weight. Lots of rotating mass will buffer motor starting loads better (like a well pump), and reduce governor droop.

Not trying to discourage your idea, just be warned that you need to be careful when belt driving with these engines. Properly set up, your idea will work just fine for a very long time.

Eric

Edit: For some reason was thinking you were running an opposed twin. Just re-read and saw "MDJB". Similar holds true for inline twins due to no bearing support between cylinders.
 

len k

Subscriber
Last Subscription Date
12/12/2018
Arodapowerco is right about using larger diameter pulleys. For a given hp/torque transmitted, larger diameter pulley have less tension due to drive torque.

But I would still recommend a cog belt instead of a V-belt. In addition to tension due to drive torque, v-belts need additional tension just to produce the friction needed to pull the belt without slipping. Cog belts can be installed with NO or little initial tension.

The original genset likely has gen and engine inline coupled with a tapered shaft. So there is not much side loading to the crank bearing , except for maybe ~1/2 the weight of the rotor (if all bearings are perfectly concentric on axis).

When using a pulley the belt load will be away from the bearing (overhanging load) and will have a mild leverage effect, increasing the radial load on bearing. So put the belt as close to the bearing as you can.

IF you limit total pull of the belts to ~1/2 weight of rotor it may be an approximately equivent load on bearing, but go with the least you can. There will be an additional bending monent on the shaft , kinda as if the belts were pulling down on a swimming pool diving board. This may be the reason the cranks were breaking, repetative fatigue bending stress.

Without the stiff rotor and it's extra end bearing the vibration natural frequencies and shapes/stresses of crank will be different. Does it make a difference, who knows. It's major task to model this stuff on the computer. In the old days it was trial and error

---------- Post added at 11:16 PM ---------- Previous post was at 11:09 PM ----------

When designing shafts for overhanging loads you typically increase shaft diameter a bit.
 

Power

Registered
From what you guys are saying, sounds like he should consider trying to fit a small bearing after the cog belt pulley to stabilize the shaft and reduce harmonics.
 

len k

Subscriber
Last Subscription Date
12/12/2018
Not practical. Very difficult to do in practice. The axial concentricity of the 3 instlled bearings would have to be extreamly good. If 3-rd bearing was off it would deflect the shaft making a bending stress in it. It's kinda common to avoid needing more than 2 bearing because of this.

Could add a 2nd shaft supported by 2 bearings and flexable coupled to crank, put belt on 2nd shaft, but that's getting expensive.

Without doing a stress analysis I'ld say use cog, put belt near bearing, make pulleys large diameter. Calulate total belt tension (add both sides) and if it's small compared to 1/2 rotor weight you may get away with it.

Disclaimer... I haven't had to do this analysis since college so I don't have a feel for it. But it's an engineer's guess, (which is better than some peoples calculations)

As you go to large diameter pulleys, belt velocity increases. There is a limit on belt max speed, I'm guessing you are won't be near that. But again that info should be available from a belt manufacture's website, for a given belt size. and pulley spacing.
 

dependable

Registered
As mentioned above, be careful about side loading motors that were not designed for that. Perhaps adding a pillow block bearing or two before pulley would take load off crankshaft and bearings.
 

Power

Registered
As mentioned above, be careful about side loading motors that were not designed for that. Perhaps adding a pillow block bearing or two before pulley would take load off crankshaft and bearings.
Len thinks this is not practical, tho.
I do not see much difference between this and the third bearing on the generator. I think with a dial indicator, some thought and time, I could get it right.
 

dependable

Registered
Here is another option if you have some skill & lots of time. This is a pto I made years ago to re power a wood chipper that had a Wisconson motor thats crankshaft had broken due to side load (I bought it that way). (Now days I buy my machines instead of make them.) I re powered with a Chevy automotive engine with even less lateral load capabilities than the Wisconson.

The lateral load is taken by the rear axle hub and bearings of a old 3/4 ton truck. Output shaft from motor is made out of 4 speed transmission welded to truck axle shaft.

The generator would not need to be that heavy duty, just sharing the idea. Agree that finding the right gen head would be a lot easier.
 

Attachments

J.B. Castagnos

Registered
Last Subscription Date
01/01/2006
If I were putting a third bearing I would machine a disc-flywheel to go on the crank, have a flex plate to mount the shaft on (automatic trans flywheel), bolted to the flywheel. Dial in the third bearing, the flex plate will handle the side load and it will forgive any misalignment or run out.
 

Graycenphil

Registered
Last Subscription Date
02/16/2012
Thanks, I appreciate all the advice. If the belt drive really won't work, I'm okay with that.

I have few hope of finding a generator head to put on this engine, and especially of finding one that is in good condition. There seems to be no shortage of ruined ones. But my motor is in great shape, and I would love to use it.
 

len k

Subscriber
Last Subscription Date
12/12/2018
Not saying it won't work, just saying there is a potential risk that it won't.

I think the v-belts helped cause Arodapowerco's crank failure because they likely put about 2-3 times more total pull on shaft, than a cog belt would. Would be interestig to know if the crack was at the bearing or in between engine bearings (Z section). If at bearing likely that it's overhanging load related, if between them it's likely a vibration shape/freg issue.

3-rd bearing,,,,,, Don't know what everyones machine shop abilities are. Dad didn't have much at his car garge so I "assume" other people don't. To make it work would want radial stiffness of bearing support to be something like at least 10x that of shaft. And not creep over time , say like maybe a bolted connection might, add loctite to bolts. Pining the various bearing support plates might not be a bad idea.

By making support stiff it takes almost all the load so shaft doesn't deflect. Also don't want alot of radial play in bearing. I'm guessing several thousandths shaft deflection will make or break the design, I haven't run the numbers.

I had an idea that may help with the dificult alignment issue. Attach a steel plate to engine, guessing 3/4 to 1" thick. Install a bearing in another plate, install it on shaft, then bolt and pin to first plate.

Not saying you must have the 3-rd bearing either. If use cog belt with large dia pulleys you may get shaft bending stress from overhanging load low enough that it doesn't matter much. I haven't run the numbers, so I don't have a feel for it. It is just a posability.
 
Top