Picked up half a megawatt worth of Caterpillar power, think I'll make a PTO generator

[Sawyer Rob]

Most modern tractors have an over running clutch built in..

When it comes to "modern" tractors, only my tinker toy tractors have an over run clutches...

NONE of my modern "farm" tractors have a over run clutch, and none is needed!

SR

 

[strantor]

I think you are being pushed into over thinking the inertia load. I wouldn't worry about it until you have some experience with actually running it. If your engine wants to stall then push the pto switch off before it stalls and engage it again before the pto brake applys. With a bit of experience you'll have it figured out.

So a stalled engine is the worst that can happen? I'm worried about burning up the tractor's internal PTO clutch and/or stripping gears. Is that not something that can happen? I'm thinking about it in terms of an automobile clutch, which can easily be burnt up by too much abuse.

 

[ptsg]

The overrunning clutch is not expensive and save the PTO clutch pack on your tractor.

This new tractors are fitted with a clutch pack that will also act as a brake when in OFF position, since there's a spring that with activate this brake. It won't brake hard but will definitely help stopping the PTO shaft faster. And this brake can be burned quite easily with attachments that will have high inertia load, like wood chippers if it doesn't have a overrunning clutch.

I don't know about your tractor. But in my Branson, I can select 2 PTO speeds (540 and 540E) with a neutral position in between. Sometimes I just bring the RPM of the engine down to idle and gently push the PTO lever into neutral and let the mower coast to a stop and just turn the PTO switch off, this won't hurt anything since it's just free spinning the PTO shaft inside the gearbox without being connected to the clutch pack and engine.

I would definitely consider a way to help starting the generator without applying the full load to the tractor. Either a start motor as you mentioned or like they do in the big Krone balers. They have a hydraulic motor that will slowly increase the baler RPM up to certain RPM then they just turn the PTO on. This will also save the PTO clutch pack.

 

[strantor]

How about making a chain "step up box"?
45T #60 sprocket on the input shaft (ie: X1-1/2B 1-1/2" Bore Hub | Weld-On Sprocket Hubs | Sprockets | Power Transmission | www.surpluscenter.com and X1-1/2B 1-1/2" Bore Hub | Weld-On Sprocket Hubs | Sprockets | Power Transmission | www.surpluscenter.com )
15T #60 sprocket on the output shaft (ie: 15T 1-1/2 Bore 6P Sprocket | Finished Bore Sprockets | Sprockets | Power Transmission | www.surpluscenter.com )
A box of 60H chain (ie: #6-1 H 1' Box Of #6 Heavy Duty Roller Chain | Roller Chain | Roller Chain & Links | Power Transmission | www.surpluscenter.com )
Mount them in a box that you size so that one side matches the SAE #2 bell on the generator head and fill with enough oil that the chain runs through it and slings it around and off you go.

Aaron Z

I am crunching the numbers on the chain drive and realizing a bit of a challenge; it won't be as simple as we'd hoped.
The following are my notes; This post is mainly for me to reference later, but I welcome peer review if anyone is brave enough to follow me down this rabbit hole. Or any fresh new ideas that are better than chain drive.

...down the rabbit hole...
I want to be able to put out 100A @ 240V to run the house & shop, that's 24kw.
Assuming worst case 90% efficiency of the generator, that's 26.67kw mechanical, or 35.7 PTO HP. (my tractor doesn't put out that much HP yet)

Consulting the data table for #60 roller chain in this reference, 1,800RPM on a 15T sprocket is only good for 10.27HP.
If I follow the 1800RPM column for #60 chain down to where 36+ HP resides, I'm looking at a 35 tooth sprocket as the smallest sprocket I can put on the generator shaft. That would require a 116 tooth sprocket on the PTO. That's bonkers.

The table also has factors for multi-strand chain setups. 1.7 factor for dual chain, 2.5 for triple chain, 3.3 for quad chain. So if using multiple chains I can look up lower horsepower values in the chart, and use smaller sprockets:
Dual: 21.0HP = 25T on the rotor, 84T on the PTO
Triple: 14.3HP = 19T on the rotor, 64T on the PTO
Quad: 10.8HP = 16T on the rotor, 54T on the PTO

The largest #60 sprocket that surpluscenter sells is a 60T, so without scouring the whole of the internet it looks like a quad-chain setup would be best. A 54T sprocket cost $44. Four of those plus (4) 16T sprockets @$27 is $207 just in sprockets, and I won't be able to use their handy single-sprocket weld-on hubs anymore; I'll have to machine my own. Oh yeah, and 10ft of #60 quad-strand chain costs $450. I suppose I could go with 4 separate strands of #60 chain, that would only be just over $100, but I don't know if the HP rating factors still hold true for that kind of arrangement.

Another idea I had was to split the reduction into two stages, with two 1:1.826 (sqrt 3.333) steps:
540rpm(PTO):986rpm(intermediate) = 1:1.826
986rpm(intermediate): 1,800(Rotor) = 1:1.826

With 60T being the largest #60 sprocket surpluscenter has, my smaller sprockets would need to be 33T [60 / 1.826 = 32.859] . The #60 table in the horsepower reference doesn't have an entry for 33T, but following the data trend, 33T @ 1800RPM is good for about 34HP. Just shy of where I wanted to be, and stepping up to #80 chain isn't an option. However, if I step down to #50, now I have a 72T sprocket available. 72 / 1.826 = 39.43, and according to the #50 HP table, a 40T sprocket @ 1,800RPM is good for 38.67HP. And the total for two pairs of 72T/40T sprockets in #50 size cost $126, compared to $160 for two pairs of 60T/33T sprockets in #60 size.

NOTE: the location of #50 sprocket, 40T, 1800RPM falls squarely in the region of Type C lubrication (oil spray) which precludes the use of a simple oil bath. A small oil pump will be required. SAE30 oil, pump should not be a problem. I have some laying around that should work.

The two-stage reduction with #50 chain appears to be the best route forward, but there will be additional inefficiencies introduced by the new stage; more moving parts, more bearings, etc. a single chain drive, according to multiple sources on the internet is about 98% efficient. This would be compounded by the 2nd chain drive, 0.98*0.98 = 96% efficient chain drive, minus the inefficiency of the added ball bearings of the idler which are hard to estimate, minus the load of an electric oil pump (figure 100W). From the input to the output of the chain drive I estimate, eh... like 95% efficiency. Then the inefficiencies associated with the PTO drivetrain between tractor and chain drive might take off another 1% or so. So I'm thinking worst case 93% overall mechanical efficiency between PTO output shaft and rotor shaft. So an additional 2.5-3 PTO HP would be required to get 100A out of the generator.

 

[strantor]

Well after spending most of today doing math and pricing on drive components for PTO power, I might be making an abrupt change in direction. My surpluscenter shopping cart for the 2-stage chain drive, that's The sprockets, chain, shafting, bearings, adapters, etc required totalled nearly $600 and that's not including shipping, or the PTO shaft, slip clutch, overturning clutch, which was going to be nearly $400 on top. So at least $1000 to get this thing spinning off the PTO. So I started thinking more about my original plan to use an automotive differential, or some sort of automotive parts solution, and I asked my brother about a busted Honda Civic he has. I was thinking of using the transmission and clutch out of it if possible. He said the engine, trans, and clutch all work fine. It's just the rest of the car that is falling apart. Wait, the engine runs? Well I wanted this thing running on propane but gas works too. He only wants $300 for the car. It's a '97 civic HX, manual trans. I did the math and at 4,700 engine RPM in 5th gear it should put out 1800 RPM. I had him start it up and run it up to 5000 RPM and it really wasn't that loud. I'm thinking I can have a standalone honda gas-powered generator for less than a PTO power setup. I'll probably need a radiator and electric cooling fan out of a full size pickup, maybe bigger. But ither than that, is there any big problems I'm not considering?

 

[texasranger556]

Might work but sounds like a lot of work to get there, and probably several hundred more in nickels and dimes. Does it have cruise control or do you plan to build some sort of governor?

Are you planning to build this onto a trailer? It will be big, ugly, and heavy for sure

 

[4570Man]

Are you sure about the gear ratio? My truck transmission is 1-1 in fourth gear and 1-1.25 in fifth. That means the transmission is going to run faster than the engine in 5th.

 

[strantor]

Are you sure about the gear ratio? My truck transmission is 1-1 in fourth gear and 1-1.25 in fifth. That means the transmission is going to run faster than the engine in 5th.

Here are the gear ratios:
1st: 1:3.250
2nd: 1:1.782
3rd: 1:1.172
4th: 1:0.909
5th: 1:0.702
Final drive ("diff"): 1:3.720

My math is as follows: (4700÷3.72)÷0.702=1799.77 RPM

As far as I know, thats how the ratios are calculated but if I'm wrong please let me know. I suspect I may have something wrong here because I calculated with the 24"dia tires he has on it, at 80mph in 5th gear he should be around 2700 engine RPM but he says it's closer to 4000 engine RPM at that speed. Maybe the gear ratio data I found online is not accurate? It's a 1997 Honda civic HX, 1.6L. I will dig deeper and see if I can find some conflicting data on the gear ratios.

 

[strantor]

Might work but sounds like a lot of work to get there, and probably several hundred more in nickels and dimes. Does it have cruise control or do you plan to build some sort of governor?

Are you planning to build this onto a trailer? It will be big, ugly, and heavy for sure

It has cruise control I was planning to repurpose as the governor. Yes, it's going onto a trailer. I was planning the PTO build to be on a trailer to begin with, with room left on the forward end for an eventual engine.

 

[texasranger556]

Your math looks ok for engine speed vs axle RPM, but a 24" tire at 1800 RPM would travel 128.5 MPH so not sure where you came up with 80 MPH. Will a Civic go 128 MPH?