71 replies to this topic

[cheapracer]

One day I will build a twin engined whatever - modern hatch and just throw a duplicate engine package into the rear. Because I want to and can.

I am intrested in peoples thoughts on control between the 2.

Some ideas

VW's tested idea - using superchargers, the rear engine blower feeds the front engine and vice versa, if you get overspeed with the rear engine it increases the power to the front and vice versa. (remember the series of Jettas they did then later the Scirroco? interesting reading - http://www.driversfo...istory/bimotor/ )

1 idea of mine (cheap) run the same EFI injector electrics off the rear engine's system, if the front engine overspeeds it will simply lean out and splutter, traction control-ish, leaving you with an underpowered rear drive car.

[bobqzzi]

Two separate ecus with a common throttle and transmissions with the same gear ratios. No need for any further integration

[murpia]

Might be worth considering a system to bias part-throttle torque more towards one end of the car than the other, mimicking the effect of a torque-biasing centre differential in a conventional 4WD system. Something like snail-cams with different profiles in the throttle mechanisms.

Regards, Ian

[Powersteer]

Could put a smaller engine at the front but with the same gear ratio to the bigger engine to get split power. Don't know about torque curve though.

[Moon Tricky]

I've lately entertained myself with the idea of essentially making an engine that's two engines combined in one package, with a clutch connecting the two to a common output shaft. One is a small "pony motor" that you use for normal driving, the other idles, or maybe even stops completely, until you put your foot down.

[phantom II]

Someone here posted a link to a hilarious twin engine mini video.

[Moon Tricky]

Originally posted by phantom II
Someone here posted a link to a hilarious twin engine mini video.

I saw a good one with a Fiat 126 that steered from both ends.

EDIT:

[cheapracer]

Originally posted by Moon Tricky
I've lately entertained myself with the idea of essentially making an engine that's two engines combined in one package, with a clutch connecting the two to a common output shaft. One is a small "pony motor" that you use for normal driving, the other idles, or maybe even stops completely, until you put your foot down.

Thats been done (hasnt everything), but sorry, dont ask me, cant remember.

[cheapracer]

Originally posted by phantom II
Someone here posted a link to a hilarious twin engine mini video.

The "John Cooper" built one of them.

[Joe Bosworth]

Let's see if anyone can remember a twin engine installation that has been successful at anything other land speed record attempts where the engines were close coupled and driving a single axle.

Regards

[kikiturbo2]

Tiger did a twin hayabusa engined Caterham seven which had a special motec ECU that could manage the individual engine power..

[WPT]

Posted by Joe Bosworth
Let's see if anyone can remember a twin engine installation that has been successful at anything other land speed record attempts where the engines were close coupled and driving a single axle.

Back in the mid 80's think Car&Driver built a twin engine CRX. WPT

[J. Edlund]

Let a main control unit handle two separate ECU's (of torque demand type); torque demands can be sent to each ECU from the main unit over a CAN bus (or optional means of communication). If you for example want more torque on the rear wheels during accelerations you simply increase the torque demand to the rear engine. Can be done using most modern stock ECU's, but it's probably a bit difficult to get them to work together as desired.

[britishtrident]

Originally posted by kikiturbo2
Tiger did a twin hayabusa engined Caterham seven which had a special motec ECU that could manage the individual engine power..

Tiger build "Sevenish" cars based on thier own chassis not the Caterham/Lotus Seven --- Caterhams lawyers have a long history of issuing writs against those who unfringe Caterhams trade marks and IP.
Tiger did however build a BEC (Bike Engined Car) with two engines on one of their own chassis which was shall we say not noted for reliability.

Going back 60 years no twin engine car that I know (other than land speed record cars) has ever succeeded in being reliable, most famous historical examples are the Fiat Bimotore Gp car and the Twin Mokes (one of which which nearly cost John cooper his life).

If the engines are coupled to the same transmission all sorts of nasty torsional vibration issues arise, if they aren't and one engine drives the front wheels and one the rear the problems of control and synchronisation are much too difficult to solve.

[britishtrident]

Originally posted by J. Edlund
Let a main control unit handle two separate ECU's (of torque demand type); torque demands can be sent to each ECU from the main unit over a CAN bus (or optional means of communication). If you for example want more torque on the rear wheels during accelerations you simply increase the torque demand to the rear engine. Can be done using most modern stock ECU's, but it's probably a bit difficult to get them to work together as desired.

Problem is even slight variations in torque or rpm could send the car into the barriers, the nightmare scenario being if one engine cuts out or worse seizes. One way round it might be to couple the drive to the two axles with a viscous coupling this would also effective damp any nasty torsional vibrations that might arise.

[Bill Sherwood]

I would suggest that rather than trying to be clever, try to make the car reliable and predictable.
You should try to control the under/oversteer with suspension dynamics and with the large masses at the effective ends of the car it will have a very high polar moment of yaw, and so be quite tame I would think.

[McGuire]

In my experience a fwd conversion to twin-engine AWD is pretty easy to build with automatic transmissions. (Years back under contract I built the "Multiple Citation," a Chevy Citation twin-AWD that was featured in some car mags. Actually there were two cars.) Using automatic boxes sidesteps the shifter linkage and clutch hassles, while the torque converters desensitize the F/R torque couple. Nowadays with electronic throttle control, torque limiting and traction control you could do some interesting things with torque biasing.

The simplest way is to start with a transverse FWD vehicle that employs full cradle-mount aka front subframe construction, so you can take a modular approach. Simply fab up some sheet-metal boxes in the rear floorpan to accept the mounts for the second cradle in the rear, build/modify the rear strut mounts, wire and plumb it and there you are essentially. Pretty simple really. Not a race car or terribly useful as a road car but a cool toy you can build fairly cheap.

[McGuire]

Originally posted by Bill Sherwood
I would suggest that rather than trying to be clever, try to make the car reliable and predictable.
You should try to control the under/oversteer

Or in this case understeer/understeer.

[275 GTB-4]

Originally posted by cheapracer
The "John Cooper" built one of them.

Which nearly killed him in a high speed failure

[Bill Sherwood]

Originally posted by McGuire


Or in this case understeer/understeer.

That's a fair point and quite possibly correct, but I think if it's built properly it should be slightly tail-heavy and so have a good basis to start with.

[McGuire]

Originally posted by britishtrident



Problem is even slight variations in torque or rpm could send the car into the barriers, the nightmare scenario being if one engine cuts out or worse seizes.

Not really unless the car is remarkably overpowered. And with two complete drivetrains installed the power/weight ratio will probably not be anything outlandish.

Originally posted by britishtrident



One way round it might be to couple the drive to the two axles with a viscous coupling this would also effective damp any nasty torsional vibrations that might arise.

With current elecronics it would be laughably easy to to control the rpm of one engine to within +/- 20 rpm of the other, via the serial data ports of the OE PCMs.

[cheapracer]

Originally posted by 275 GTB-4


Which nearly killed him in a high speed failure

Nothing to do with being twin engined - suspension component faliure from memory, didnt a rear suspension link fall off?

[cheapracer]

Originally posted by McGuire



With current elecronics it would be laughably easy to to control the rpm of one engine to within +/- 20 rpm of the other, via the serial data ports of the OE PCMs.

Thats what I would have thought, but of course we may like the rear engine to have a bit more grunt on occasion too.

[imaginesix]

Originally posted by McGuire
With current elecronics it would be laughably easy to to control the rpm of one engine to within +/- 20 rpm of the other, via the serial data ports of the OE PCMs.

Is that good enough though? Isn't it output that you want to be able to control proportionately, and rpm isn't tied to output at all.

[McGuire]

Originally posted by cheapracer


Thats what I would have thought, but of course we may like the rear engine to have a bit more grunt on occasion too.

Well, ideally we would like to be able to change the F/R torque bias at will, and as I said, with electronic throttle control, torque tailoring and traction control that should be pretty easy. In fact, that capability makes a rather old idea (tandem AWD) newly interesting.

[Joe Bosworth]

When this thread started I was forced to ask myself why would anyone want to build a twin engine thing in the first place?

Now, if the answer is because of the technical challenge and as an intellectual exercise I have to say good luck and godspeed.

On the other hand, if it is to meet a pre-planned performance level, such as a lap time for a given distance and condition and it is to do so at the lowest possible cost then I have to say don't go down this twin engine path.

I belong to a mechanical school that places simplicity, low polar moment, minimum weight and minimum cost as objectives to be strived for. The best performance will be attained in real life when these design elements are optimised.

There is alsmost no level of engine output that one might desire for anything short of top speed record competition that can't be attained easier and cheaper using one engine rather than two. If one decides that their target performance requires X HP, say anything from 100 to 1200 plus, then that X is attainable with ease from a single engine that is all of light, compact and inexpensive to build.

I can not think of any twin engine combination that will end up being lighter than a single engine application of the same HP.

Given the packaging constraints I can't think of a single twin engine configuration that wouldn't end up with a far higher polar moment than a single engine.

Two sets of engines to build up in the first palce will be more expensive and complex than a any single engine.

If one believes they need 4wd to meet the performance objective then I believe that conventional 4wd drive technology is cheaper, simpler and lighter than any twin transmission, twin engine alternative. If it is believed that torque splitting is important then any level of torque splitting can be attained today with conventual 4wd packages.

I can go on but I don't believe there is any cogent arguement to build a twin engine thing to best attain a given performance objective except at the furthest boundaries of objectives.

Of course, as I started with, if the objective is as an intgellectual ecercise then carry on to your hearts delight and don't let a sceptic and realist like me get in your way.

Even the Wright Brothers didn't know where they were leading to!

Regards

[275 GTB-4]

Originally posted by Joe Bosworth
When this thread started I was forced to ask myself why would anyone want to build a twin engine thing in the first place?

Now, if the answer is because of the technical challenge and as an intellectual exercise I have to say good luck and godspeed.

On the other hand, if it is to meet a pre-planned performance level, such as a lap time for a given distance and condition and it is to do so at the lowest possible cost then I have to say don't go down this twin engine path.

I belong to a mechanical school that places simplicity, low polar moment, minimum weight and minimum cost as objectives to be strived for. The best performance will be attained in real life when these design elements are optimised.

There is alsmost no level of engine output that one might desire for anything short of top speed record competition that can't be attained easier and cheaper using one engine rather than two. If one decides that their target performance requires X HP, say anything from 100 to 1200 plus, then that X is attainable with ease from a single engine that is all of light, compact and inexpensive to build.

I can not think of any twin engine combination that will end up being lighter than a single engine application of the same HP.

Given the packaging constraints I can't think of a single twin engine configuration that wouldn't end up with a far higher polar moment than a single engine.

Two sets of engines to build up in the first palce will be more expensive and complex than a any single engine.

If one believes they need 4wd to meet the performance objective then I believe that conventional 4wd drive technology is cheaper, simpler and lighter than any twin transmission, twin engine alternative. If it is believed that torque splitting is important then any level of torque splitting can be attained today with conventual 4wd packages.

I can go on but I don't believe there is any cogent arguement to build a twin engine thing to best attain a given performance objective except at the furthest boundaries of objectives.

Of course, as I started with, if the objective is as an intgellectual ecercise then carry on to your hearts delight and don't let a sceptic and realist like me get in your way.

Even the Wright Brothers didn't know where they were leading to!

Regards

Totally agree....however....I would have given my right testicle to drive that twin Toronado

[McGuire]

Originally posted by imaginesix
Is that good enough though? Isn't it output that you want to be able to control proportionately, and rpm isn't tied to output at all.

If the two engines are roughly identical it is.

[McGuire]

Originally posted by Joe Bosworth
There is alsmost no level of engine output that one might desire for anything short of top speed record competition that can't be attained easier and cheaper using one engine rather than two. If one decides that their target performance requires X HP, say anything from 100 to 1200 plus, then that X is attainable with ease from a single engine that is all of light, compact and inexpensive to build.

Not if you can get the second engine and driveline out of the junkyard for pennies. Then you have doubled horsepower very cheaply with stock output curves. When you double the hp of a production engine you no longer have a road engine, usually.

As I said earlier, a tandem AWD is essentially a toy. It won't be a useful road car (horrible fuel economy, no cargo capacity) and it won't be a successful race car (not competitive against single-engine cars of equal weight/displacement) but that doesn't mean it can't be a whole lot of fun, not only to drive but for people to look at. Are we opposed to fun now too?

[Joe Bosworth]

Cheap and McG

To have a whole lot of fun and to have something for people to look at:

Take Cheap,s hot hatch that he was starting with. Turf out the front engine while keeping the suspension and steering bits. (Might want/need to re-spring.)

Where Cheap was going to cannabalise the back end anyway, mate a 5+ liter Chev kinda thing, put it in the back, build a perspex surround so people can see whats back there when they get curious.

Clean up and dress the engine.

If you want a bit more do the very mild stuff that gets 500 driveable HP.

To frighten the natives and cause them to wonder when they first see the thing put on a wide set of tires for the rear which will be needed anyway.

Far simpler, more outragous, maybe even can be made to half handle. Lets see, 2200 pounds, 500 + Hp, quarter mile in 9.5 secs and 140 mph. Might even get 24+ MPG if you keep your foot out of it.

Now you have street cred and fun!!!

We were dreaming weren't we??

[cheapracer]

Originally posted by Joe Bosworth


1 When this thread started I was forced to ask myself why would anyone want to build a twin engine thing in the first place?


2 Two sets of engines to build up in the first palce will be more expensive and complex than a any single engine.


Regards

1 Very simple, I am locked into an engine contract and whatever vehicles I build require that I use that engine therefore if I want something a little more special maybe I will use 2.

2 Is cheaper but agreed more complex. You are unaware of my purchase proce.

[J. Edlund]

Originally posted by britishtrident



Problem is even slight variations in torque or rpm could send the car into the barriers, the nightmare scenario being if one engine cuts out or worse seizes. One way round it might be to couple the drive to the two axles with a viscous coupling this would also effective damp any nasty torsional vibrations that might arise.

Have you seen what happens to single engine car if the engine or drivetrain seizes? Yep, pretty much the same that will happen to twin engine car if one engine goes out.

To control engine speed and torque output would be rather easy with modern torque demand control systems, and if we add an anti spin system to this, we would hardly have any problems. To reduce the risk o a spin even further, it would probably be possible to watch over the torque output from each of the engines and if one produces a certain amount of more torque than the other, reduce the output of the engine that produces the highest amount of torque. Modern ECU's already have a function for this.

Two stock engines complete with their drivetrains won't have any problems with torsion vibration, or any other kind of issue. If you on the other hand tries to make some sort of mechanical coupling between the two engines, I would expect that you run in to some sort of problem.

One of the engines also have no need to power equipment like alternator, A/C compressor, servo steering pump and so on. Removing this equipment from one engine is certainly an advantage, but might mess things up with the torque control.

[britishtrident]

Originally posted by cheapracer




1 Very simple, I am locked into an engine contract and whatever vehicles I build require that I use that engine therefore if I want something a little more special maybe I will use 2.

2 Is cheaper but agreed more complex. You are unaware of my purchase proce.

Joke right ?

[bobqzzi]

Many years ago I went a "street" night at New England Dragway- my first time. By far the coolest car there was a Caddy Eldorado, which was very stock looking excepting the piece of stove pipe sticking out the passenger window.

He rolled up to the line while the jacked up Nova in the other lane did a mighty burn-out with an ear splitting scream. When they took off the Caddy simply pulled away smartly while the Nova was much more dramatic. It was no contest. The Caddy ran a low 10, the Nova a high 10.

The owner later explained he had a 472 in the front and a 500 in the back. Pretty stock except for carbs and a very mild cam.


I was impressed.

I am building a twin engine 4wd car, as McGuire says, for fun. One of the benefits is that you don't need some sort of extraordinary transmission that can take 700lb/ft of torque- 2 stock ones will do fine.

Not very practical, but should be a hoot.

[Knot]

A good approach would be to use two Veyron engines.

Or at least two RR merlin engines.

[Powersteer]

How about a Mini with two Hayabusa turbocharged engines front and mid-rear bringing engine size from 1.3 to 2.6, twin turbo 4wd.

[cheapracer]

Originally posted by britishtrident


Joke right ?

I'm sorry, did I miss something?


Anyway, life is meant to be fun, happy to oblige even if I dont understand what I did :-)

[cheapracer]

Originally posted by Powersteer
How about a Mini with two Hayabusa turbocharged engines front and mid-rear bringing engine size from 1.3 to 2.6, twin turbo 4wd.

"Whatstheirnames" who do the mid engine Mini's in England have an Ultima thats not far off this. Go to their website.

I remember, its "Z Cars".

[McGuire]

I think you will get far more satisfactory and manageable results using two engines of middling displacement with very flat torque curves, coupled to automatic transmissions with torque converters. Then as long as both transmissions are in the same gear, there will be no F/R torque couple to speak of. That will more or less eliminate the "synchronization" issue. However, with two peaky, small-displacement engines with manual transmissions you are magnifying the the potential torque differential F/R at any given time.

[Bill Sherwood]

Originally posted by Powersteer
How about a Mini with two Hayabusa turbocharged engines front and mid-rear bringing engine size from 1.3 to 2.6, twin turbo 4wd.

Crazy talk ....




.... it needs one on each wheel.

[desmo]

With clever control of individual wheels for tractive and braking torques, might steered wheels be (theoretically at least) largely unnecessary at speed?

[cheapracer]

Originally posted by desmo
With clever control of individual wheels for tractive and braking torques, might steered wheels be (theoretically at least) largely unnecessary at speed?

Is that like when I drive I rest my hands behind my head and steer with my knees?

Actually I was thinking why not tie up some small oil pumps to each engine pumping to each other, this would contain them but allow a small amount of 'slip'.

My engines are 'Chinese' Toyota 4AGE's with manual 5 speeds, no options.

[imaginesix]

Couldn't you use that kind of setup as a differential as well? And wouldn't the advantages and disadvantages be the same for interconnecting engines as they would be for interconnecting wheels on an axle? (Friction mostly, I think).

[cheapracer]

Originally posted by imaginesix
Couldn't you use that kind of setup as a differential as well? And wouldn't the advantages and disadvantages be the same for interconnecting engines as they would be for interconnecting wheels on an axle? (Friction mostly, I think).

Yes and it would be interesting and reasonably easy to do things like linking the left front to the right rear (and vice versa of course).

[britishtrident]

I suspect a lot of the twin engine advocates here haven't given thought to how a car steered by the front wheels actually corners.

If you start out by drawing out the classic ackerman geometry drawing it is pretty obvious that even on a neutral steering vehicle both front wheels are cornering on longer radius arc than the rears, therefore the rotation speed of the front wheels is greater than the rears.

In the real world ackerman geometry dosen't hold 100% true because to generate a cornering force the tyres must run at slip angles but it still holds true that the front end of a neutral handling car always follows a longer radius of arc than the rears.

[britishtrident]

Originally posted by cheapracer


Is that like when I drive I rest my hands behind my head and steer with my knees?

Actually I was thinking why not tie up some small oil pumps to each engine pumping to each other, this would contain them but allow a small amount of 'slip'.

My engines are 'Chinese' Toyota 4AGE's with manual 5 speeds, no options.

Power losses for linking the wheels hydraulically are large.

Various hydraulic transmission systems mainly using swash plate motors have been considered for vehicles many times, it is only really viable for very special purpose applications.

[cheapracer]

Originally posted by britishtrident



Power losses for linking the wheels hydraulically are large.

Various hydraulic transmission systems mainly using swash plate motors have been considered for vehicles many times, it is only really viable for very special purpose applications.

I mean a form of recirculating liquid system that would contain differences of speed. The point is to get a power loss/drain on the engine that overspeeds and transfer that loss to the other.

[cheapracer]

Originally posted by britishtrident
I suspect a lot of the twin engine advocates here haven't given thought to how a car steered by the front wheels actually corners.

Dont you just turn the steering wheel?

[britishtrident]

Originally posted by cheapracer



I mean a form of recirculating liquid system that would contain differences of speed. The point is to get a power loss/drain on the engine that overspeeds and transfer that loss to the other.

Any hydraulic power transmission system involves pumping a fluid at high velocity through small bore pipe work, even running in a straight line and the system isn't actually transfering any power the fluid is still being pumped around the system --- the fluid friction losses are large.

[vrbanana]

I’m in the middle of building a twin engine track car at the moment, it’s not finished yet but once it is I’ll add details on handling and the drivability.

Details of the build here

http://www.clubgti.c...ad.php?t=116662