Electric assist motor feasibility to improve MPG

[casm]

(Please note that this is not a thread to debate how much you love or hate hybrids; if that's what you're looking for, kindly take it over to the Off-Topic forum. Thanks.)

Having been watching what people have been doing to improve XJ fuel economy over the last few years, I've come to the conclusion that short of a diesel transplant there just aren't many other options that produce significant results while maintaining the vehicle's expected capabilities. This isn't to knock diesel by any means (I'm actually quite a fan of diesels) - but in North America, where suitable diesels aren't particularly plentiful to choose from and regulations may make transplanting them into an XJ problematic from a legal standpoint, it's not necessarily the most straightforward route to take.

This got me to thinking about what other approaches may be reasonable in relation to an XJ. 'Reasonable' in this context means 'not unduly compromising ability' - what we know it to be capable of now is what it should overwhelmingly remain capable of after the fact.

So, on to the basic idea: a big alternator and what would essentially be a wheel hub motor attached between the rear driveshaft and transfer case. This motor would have sufficient torque to move the XJ off the line and up to 30mph or so, at which point the driveline would be run solely by the gas engine. At no point does the gas engine switch off as in a hybrid; it provides the motive power to run the alternator powering the electric assist motor. Depending on the electric motor in use, one or two additional batteries may be necessary.

It's essentially the approach being taken by the Chevy Volt (which itself has been used by diesel-electric locomotives for decades), but on a backyard-engineering scale.

Thoughts? It seems at least plausible to my mind, but I'd be interested to see what holes can be poked in it.

 

[DieselSJ]

I thought diesel-electric locomotives just used the diesel engine to run a generator and all motion was supplied by the electric motors?

 

[bjoehandley]

They are unless it's the "Green Goat", which was a Switcher that has a "smaller" diesel engine to recharge it's batteries.

 

[yossarian19]

Yeah, the analogy to the diesel-electric & the Volt isn't quite right... you are building a backyard Honda Insight, I think.

You are going to run into tech challenges when you try and control this motor with the gas pedal without opening the throttle, finding a big enough alternator (and then mounting it and finding the right pulleys / belts) finding a motor beefy enough to move the Jeep... Then somehow the motor needs to spin the drive shaft but not the engine, right? because if the throttle opens, you are burning gas. If not, spinning the shaft is fighting the engine's compression...

What else...

You'll need the motor to disengage from the driveline and re engage smoothly,
the alternator will cause a good amount of drag on the engine, i'd think

I don't know, man. If you do it, be sure to document it. I like the idea, I just don't think it is going to be practical / cost effective on any reasonable timeline.

You can buy an Escape or a Highlander in a hybrid. Not solid axles or anything, but just fine for a camping / "where does this dirt road go?" sort of rig...

 

[bjoehandley]

Now that I think of it, GM and Chrysler use a hybrid system like that the F/S GM SUV's and it had been offered on the Aspen and Durango in conjunction with the Hemi Motor. There were a couple motors built into the trans that would operate the trucks up to 30mph on electricity, then it would kick on the gas after that if you were to exceed that speed. They had this Hoover like howl to them from the motors that sounded kinda cool, but odd at the same time since I would normally hear a LS motor growling instead under the same circumstances ......

 

[in4aride]

only thing is see is the "big alternator" is going to require extra rpms out of the motor to turn it over enough to move the jeep. the extra rpm requires more fuel eliminating the idea of better gas milage.

ive had pretty good luck with stock jeeps to get just at or barely over 20mpg.
i atribute it to low rpms@ normal driving speeds.... which for me is 70+
i imagine the best way to go is another 2hi selection where the gearing makes the motor rpm lower than stock. and 'on the fly' changability would be needed when going uphill or around town where you would want the lower gears...
make sense?

 

[CherBear]

EV drivetrain swap with an actual generator seems to be the easiest way. How ever that would be "compromising" according to your definition.
There is tonnes of electric motor tech out there, the problem would be blending it together with a gasoline motor. All of this hybrid tech is fairly new in the first place.

Really the only financially attractrive way i see this happening is you getting a hybrid drivetrain and stuffing it into an xj.

 

[yossarian19]

Yeah... "Financially attractive" and "hybrid drivetrain" rarely go together, at least in the after market...

 

[coldoath]

Hm...if you wanted eco friendly or a gas sipper an XJ is definitely NOT the vehicle of choice.

Also, by the time you end up spending all that money on fabrication and parts you're going to need to drive your XJ for 20 years before you recoup the cost of the install. lol

ANYTHING is "plausible". This idea is "plausible" however, it is not probable.

 

[casm]

Yeah, the analogy to the diesel-electric & the Volt isn't quite right... you are building a backyard Honda Insight, I think.

Yeah, that may be closer to the truth. FWIW, the idea in all of this is to avoid ending up with the layout typical of hybrids currently - packaging batteries into the vehicle and dragging around their weight isn't something I really want to get into. What actually gave me the idea of the assist motor on the rear driveline was looking into wheel hub motors: they make a lot of sense in vehicles that were designed to accept them, but trying to retrofit them into an existing vehicle isn't something I'm convinced is practical at the moment. But putting what is essentially one onto the rear driveline... You can see where this was going.

You are going to run into tech challenges when you try and control this motor with the gas pedal without opening the throttle, finding a big enough alternator (and then mounting it and finding the right pulleys / belts) finding a motor beefy enough to move the Jeep... Then somehow the motor needs to spin the drive shaft but not the engine, right? because if the throttle opens, you are burning gas. If not, spinning the shaft is fighting the engine's compression...

Agreed on all of the above. What I'd like to find is a motor that is internally-clutched or similar, running fully-disengaged when no power is applied; this would more or less solve the motor selection equation, at least as far as which one is transmitting power to the driveline at any given moment.

You'll need the motor to disengage from the driveline and re engage smoothly,
the alternator will cause a good amount of drag on the engine, i'd think

Finding a suitable alternator is proving tricky; outputs in the ranges necessary aren't common in automotive applications. One idea was to run the alternator for the assist motor on a clutched pulley similar to an A/C compressor; keep the stock alternator in place for running the 'normal' systems. Again, not entirely certain this is practical.

I don't know, man. If you do it, be sure to document it. I like the idea, I just don't think it is going to be practical / cost effective on any reasonable timeline.

I have to admit that I'm in agreement with you regarding the practicality of it. Something I'd like to see happen is that someone develops a kit for vehicles like ours that could be used to do a retrofit, but I'm not holding my breath on that one.

You can buy an Escape or a Highlander in a hybrid. Not solid axles or anything, but just fine for a camping / "where does this dirt road go?" sort of rig...

Understood, and I'm in agreement with you on that - part of what got me thinking about this was that I was toying with the idea of a Tracker/Vitara/Sidekick for exactly that kind of trail use, but with 25mpg economy. But I don't really want another vehicle, and that got the left-field idea in my head that it may be possible to come up with something for an XJ for around the cost of a used Suzuki.

Now that I think of it, GM and Chrysler use a hybrid system like that the F/S GM SUV's and it had been offered on the Aspen and Durango in conjunction with the Hemi Motor. There were a couple motors built into the trans that would operate the trucks up to 30mph on electricity, then it would kick on the gas after that if you were to exceed that speed. They had this Hoover like howl to them from the motors that sounded kinda cool, but odd at the same time since I would normally hear a LS motor growling instead under the same circumstances ......

Yep - there're a couple of (IIRC) Tahoe hybrids running around town that have that quality to them. It's kind of interesting to hear it, but totally not what you'd expect from a vehicle like that.

only thing is see is the "big alternator" is going to require extra rpms out of the motor to turn it over enough to move the jeep. the extra rpm requires more fuel eliminating the idea of better gas milage.

That's been exactly one of the things running through my head on this. However, if the drag of running that giant alternator during electric assist operation is less than what the drivetrain would have to absorb running gas-only to move the vehicle off from rest (where my off-the-cuff observations indicate the worst economy is), there may be enough of an offset to make it worthwhile. But it is understood that that's completely theoretical for right now.

ive had pretty good luck with stock jeeps to get just at or barely over 20mpg.
i atribute it to low rpms@ normal driving speeds.... which for me is 70+
i imagine the best way to go is another 2hi selection where the gearing makes the motor rpm lower than stock. and 'on the fly' changability would be needed when going uphill or around town where you would want the lower gears...
make sense?

Agreed on the above; I can get 19mpg even with all of the weight and aerodynamic disadvantages modifying my Jeep has introduced, and given what it has to work against, 19mpg isn't bad at all. Other off-the-cuff observations have me thinking that replacing the AW4 with a CVT may actually be more effective (and practical) than the electric assist route overall, but I'm not aware of a CVT with a divorced controller that would be anything even close to a drop-in.

EV drivetrain swap with an actual generator seems to be the easiest way. How ever that would be "compromising" according to your definition.

Understood, but I am pretty much proposing using the 4.0 as a generator for the assist motor under certain conditions. I see where you're going with the idea (and agree on the compromise part), but there's a bit more to this:

There is tonnes of electric motor tech out there, the problem would be blending it together with a gasoline motor. All of this hybrid tech is fairly new in the first place.

To be fair, the hybrid technologies we're seeing in production now can trace their roots back to the 1970s - and that's a large part of the reason I'm convinced that manufacturers are doing it the wrong way in this application for the most part. The basic concepts are overall fairly sound, but my gut feeling is that we're still a generation of design and production away from having something workable for vehicles that get used the way Jeeps tend to.

Really the only financially attractrive way i see this happening is you getting a hybrid drivetrain and stuffing it into an xj.

Not disagreeing - just not keen on that approach. The real killer for me is adding battery weight and everything that goes along with it from packaging on up; in a vehicle where you can design around that problem from the ground up it's not as big a deal, but from a retrofit standpoint it's very undesirable. Not to mention that there's something in me that hates the idea of adding weight: it's just countereffective to efficiency.

Hm...if you wanted eco friendly or a gas sipper an XJ is definitely NOT the vehicle of choice.

Yes, but neither of those were the points that I was addressing. To clarify: this is an exercise in what can be done to improve an existing system. I've got other cars that get better fuel economy than my XJ, and know that I could just use them when practical to do so, but the reality is that I'm behind the wheel of the Jeep most of the time.

Also, by the time you end up spending all that money on fabrication and parts you're going to need to drive your XJ for 20 years before you recoup the cost of the install. lol

Agreed that there would be a break-even point somewhere in the future from doing this, and conversion costs would have to be amortised out across running costs over a period of time. However, if an assist motor would take my average fuel economy from 15mpg to 20mpg, given how much I spend on fuel annually it would take about 3.5 years to recoup the cost of a $3000 installation.

For me, this is acceptable - I don't plan on retiring the XJ anytime soon, and $3000 is about what I would expect to spend on a serviceable older vehicle with better economy. However, by going the electric assist route, I would eliminate maintenance costs on that $3000 vehicle, insurance, and registration. If those are accounted for in the cost of that $3000 car, the break-even point is still around 4 years, given a roughly 23mpg economy average.

Obviously return on investment depends on a lot of factors; in my case, if I can keep to the $3000 figure, it makes sense. For someone else, it may not. The same basic set of caveats apply to buying anything really, though.

ANYTHING is "plausible". This idea is "plausible" however, it is not probable.

Possibly - I'm not in disagreement with you per se, but I'm not convinced that it's entirely non-probable. Of course, this is all theoretical at the moment, so it's a bit early to call it busted or confirmed just yet.

 

[CherBear]

I am still a little confused as to how this proposed system would work, this is what I imagine from your definition:

idle - 30mph

The gasoline motor supplies no power to wheels, but does power the alternator.

- Here is where i see a major problem. You are expecting your alternator to be making maximum power while the gasoline engine is at its lowest RPM. If you have to "rev" the engine to make more electric power, then how are you going to be saving any gas?
The electric motor can only be putting out the same amount of power as the alternator is taking in. Which in turn can only be as much as the engine puts out. ( side note, when i say " as much" I mean "less than" due to friction and other sources of loss)

So how are you expecting your electric motor to accelerate your vehicle when it only has the power of the gas engine at idle?

This is why hybrids have batteries, you have to be able to "bank" energy somewhere. Batteries can store energy gathered while coasting, idling ect. Therefore captureing energy that would have been wasted. You need batteries, otherwise I see no way for this to work.

 

[Milford Cubicle II]

Seems like it would be more feasible and practical to drop the 4.0 and go straight diesel-electric.

Right? :dunno:

 

[PhotographerMike]

Porsche racing has an interesting idea to capture "free" extra power, using a horizontally mounted, high speed, flywheel mounted where the passenger seat would be.

IIRC It is "charged" using braking energy (generators/motors in the front wheels) and adds power back for short bursts of acceleration.

In the Jeep (very theoretical), charge the flywheel off the T.C. (fixed location, using 90 deg. gear set) instead of lightly braking or shifting down to control speed. Then use the stored energy to help acceleration from a stop.

Another advantage of a rapidly spinning flywheel is that it would help keep the Jeep upright! (of course you might not be able to start up or down a hill)

 

[casm]

I am still a little confused as to how this proposed system would work, this is what I imagine from your definition:

idle - 30mph

The gasoline motor supplies no power to wheels, but does power the alternator.

Effectively, yes. Ideally this would be a driver-switchable setting, but that's the gist of it.

- Here is where i see a major problem. You are expecting your alternator to be making maximum power while the gasoline engine is at its lowest RPM. If you have to "rev" the engine to make more electric power, then how are you going to be saving any gas?

Ah, OK. I see why you're mentioning this; it's something I haven't really explained my thinking on. To clarify:

The electric motor can only be putting out the same amount of power as the alternator is taking in. Which in turn can only be as much as the engine puts out. ( side note, when i say " as much" I mean "less than" due to friction and other sources of loss)

So how are you expecting your electric motor to accelerate your vehicle when it only has the power of the gas engine at idle?

The idea was that the gas engine would run at the required RPM necessary to power the electric assist motor - so yes, the gas engine's RPM would rise and lower as necessary to run the secondary alternator. This will cause gasoline consumption to increase or decrease with RPMs, but what I'm looking at is if the electric assist motor can more efficiently overcome the initial push off of the line than the gas engine given that the drivetrain would largely be decoupled from the gas engine.

Note that my expectation in this isn't 'free' propulsion - just something that is an improvement over the existing system. How much of an improvement is definitely up for debate, though, and my gut feeling is that a good, modern diesel engine would probably outperform the system I'm proposing.

This is why hybrids have batteries, you have to be able to "bank" energy somewhere. Batteries can store energy gathered while coasting, idling ect. Therefore captureing energy that would have been wasted. You need batteries, otherwise I see no way for this to work.

True, but part of the reason why hybrids carry batteries is for fill-in power. In this instance, we're talking about more of a single-use system akin to, say, a portable generator: the better gensets in that regard typically measure demand against output and adjust their fuel consumption accordingly by altering RPMs; no batteries are required. This is closer to the type of installation I'm proposing.

Now, one possibility I'd be interested in would be using capacitors instead of batteries due to their weight and packaging advantages, but that opens up safe discharge, cost, and construction feasibility issues.

Seems like it would be more feasible and practical to drop the 4.0 and go straight diesel-electric.

Right? :dunno:

Actually... If I were going to do that, I'd just go straight diesel. That's not to say that I'm not in agreement with you, but given that I'm aiming more for an improvement to an existing system than ultimate efficiency, it's a little outside the scope of what I'm shooting for here. But there's no doubt in my mind that diesel would definitely be a preferable basis for all of this.

Having said that, there's probably a good chunk of diesel-electric technology that would transfer over to what I'm proposing here, so this is a good excuse to go back and do some re-reading on it.

Porsche racing has an interesting idea to capture "free" extra power, using a horizontally mounted, high speed, flywheel mounted where the passenger seat would be.

IIRC It is "charged" using braking energy (generators/motors in the front wheels) and adds power back for short bursts of acceleration.

Yep, a few manufacturers have played around with this idea at various times - the Pontiac FX85 and Chrysler Patriot spring to mind, and I have a vague recollection of having ridden on busses that used this kind of system under the lower floor when I was a kid. It's an interesting idea, but I'm not completely convinced that storing mechanical energy as mechanical energy is necessarily the way to go. However, as a mechanical store of electromechanical energy, it could work. Need to look into it a bit further.

In the Jeep (very theoretical), charge the flywheel off the T.C. (fixed location, using 90 deg. gear set) instead of lightly braking or shifting down to control speed. Then use the stored energy to help acceleration from a stop.

Don't see why it couldn't work. Packaging would be interesting, but that aside my only question would be whether it would deliver enough of an improvement to justify its installation.

Another advantage of a rapidly spinning flywheel is that it would help keep the Jeep upright! (of course you might not be able to start up or down a hill)

The world's first gyroscopically-controlled anti-rollover system, perfect for seriously off-camber trails... I love it

 

[CherBear]

Alright, i think i see your idea a little more clearly.
A few more questions....

IIRC one major advantage of a electric motor is that they have nearly 100% torque through out the entire rpm range. This makes them very efficient at acceleration.

Now gasoline engines only produce peak torque numbers through a limited rpm range. (less efficient) AND transmissions, especially autos, are very bad at conserving energy.

So it seems that your plan is to bypass most of the mechanical drive train, convert your gasoline into electric energy, then turn it into mechanical energy at the wheels.

All the while hoping this turns out to be more efficient than a standard drivetrain.

Sound about right at this point?

Now most of the EV conversions i have seen used 100hp-120hp electric motors. I think 100hp would be a good power range for this project, but could easily be scaled down.

If we look at this dyno sheet: (ignore the red lines, its for a muffler comparison)

You will see that in order for a 4.0 to make 100hp to give full power to the EM, you need to be at around 3250 rpm. Count in the loss from the EV drivetrain, we are probably looking between 3500 and 3750 rpm.

Now that is if you want FULL EM power to the wheels, so lower rpms would create lower voltage therefore low power.

Just looking at these numbers it doesnt sound like you are going to be getting much a difference in mpg. Let me remind you that i included a lot of speculation in that post, so take that with a grain of salt.

My suggestion still stands that you tear out the 4.0, install an EV drivetrain sans batteries and put in a generator.

 

[kastein]

I am still a little confused as to how this proposed system would work, this is what I imagine from your definition:

idle - 30mph

The gasoline motor supplies no power to wheels, but does power the alternator.

- Here is where i see a major problem. You are expecting your alternator to be making maximum power while the gasoline engine is at its lowest RPM. If you have to "rev" the engine to make more electric power, then how are you going to be saving any gas?
The electric motor can only be putting out the same amount of power as the alternator is taking in. Which in turn can only be as much as the engine puts out. ( side note, when i say " as much" I mean "less than" due to friction and other sources of loss)

So how are you expecting your electric motor to accelerate your vehicle when it only has the power of the gas engine at idle?

This is why hybrids have batteries, you have to be able to "bank" energy somewhere. Batteries can store energy gathered while coasting, idling ect. Therefore captureing energy that would have been wasted. You need batteries, otherwise I see no way for this to work.

The way I would do this -
Set the transmission up to be "in on" this. From 0 to 30mph, transmission stays in neutral, and the controller keeps the engine running at the speed at which it produces the most efficient output from the alternator (i.e. where motor efficiency * alternator efficiency is at a maximum.) Alternator feeds a modern switch-mode high current motor driver; ideally each phase from the alternator would feed directly into a power factor correction circuit (ON Semiconductor has a great whitepaper on these kinds of circuits, it is linked from the page for their NCP1650 active power factor correction chip) to produce ~400VDC at high efficiency, which would then be used directly by the motor controller PWM circuitry.

Basically, you can get away with a much smaller winding alternator and much cheaper rectifiers if you use what amounts to an extremely agile, self adjusting boost converter instead of a massive 6-diode rectifier pack, since you draw current from the coils for the entire period of the sine wave rather than only at peak, when the coil voltage plus the rectifier forward voltage drop is greater than the voltage on the filtering capacitor.

The controller would have to watch the VSS pulsetrain and determine when it should transition from powering the DC motor and leaving the AW4 in neutral to disconnecting the DC motor and putting the AW4 into gear. If you designed it carefully and integrated the whole thing with the cruise control solenoids and the AW4 TCU I think it could be done reasonably simply, aside from finding a place to graft the DC motor into the drivetrain... and a place to put it. If you were really careful about choosing the RPM the motor runs at while in DC-assist mode and what gear you switch over in (and what speed the transfer occurs at), you could probably even do it seamlessly - shift AW4 into whatever gear fits best, then disable the DC motor, instead of trying to switch the motor off, get the engine running at the right RPM, and shift the AW4 all at once without giving the driver whiplash.

 

[bigalpha]

Your plan focuses on using electric power only during acceleration. What about using electric power to maintain a constant speed, e.g. during cruising?

While the electric power would be very useful during acceleration (constant, 100% torque), would you be spending more time cruising?

 

[coldoath]

You had mentioned something about a budget of $3K, I would say good luck keeping that budget. Not only are you building an experimental vehicle, with which numerous things could go wrong, you are going to need to do some mean coding/altering of code to get the computer to known wtf to do and when to do it.

Sorry if it seams like I'm bashing you, which I might be but it just seems like a terrible waist of time and money. However, if you do decide to go through with it best of luck and provided you make a build thread I'd watch it like a hawk!:cheers:

 

[Dingo509]

A couple of ideas i've come up with while reading this; i'll explain each below;
1. Install a computer controller clutch in place of the torque converter
2. install a motor/generator between the transmission and transfer case
3. install a motor/generator in place of the vehicle starter.
4. Design and build a dual alternator system that is actually 2 alternators on in front of the other.

1. I have heard this was a popular mod to some circle track cars in the 50/60s using a 2 speed torqueflite transmission. think about it, it should work. you would have to do some work and some redesigning to keep the transmission pump turning, but other wise it is WAY more effecient than a standard torque converter.

2.Advanced adapters recently came out with a low range box for the new JK that installed a secondary low range between the tranny /xcase using that idea and possible moving the xcase back a few inches would allow you the location to install the electric motor for acceleration then use it as a generator for slowing and cruising. this would again require batteries to store the energy. but it is a good place to install one. I have seen similiar units under Smaller RVs converted to limos or "party buses"
although they are installed after the transmission right before the rear drive shaft.

3. Similiar to the idea above on a combination motor/generator except gear driven off of the flywheel. I know some aircraft use gear driven accessories on the motors so the technology exists.

4. This one is one that i have been thinking about for other reasons. but if you where to take 2 say 90 amp alternators and hook them end to end. then you would effectily have a 180 amp alternator that will still charge at low engine rpm. now granted bigger alternators are avialable but as the amperage goes up, so does the required minimum rpm. problems with this include cooling for the second rotor/stator and getting enough belt wrap.

 

[bigalpha]

You had mentioned something about a budget of $3K, I would say good luck keeping that budget. Not only are you building an experimental vehicle, with which numerous things could go wrong, you are going to need to do some mean coding/altering of code to get the computer to known wtf to do and when to do it.

Sorry if it seams like I'm bashing you, which I might be but it just seems like a terrible waist of time and money. However, if you do decide to go through with it best of luck and provided you make a build thread I'd watch it like a hawk!:cheers:

I can guarantee any code editing will not be a problem.