[Geeto67]

so I was having a conversation with a buddy of mine that runs a Turbo honda cvic in NHRA and NOPI, and this particular bench racing session ended with a crude drawing on a napkin as to how to not only turbo a cb750 (pretty easy) but how to fuel inject it as well.

Here is the basic premise, he works for subaru so he can get me a stock WRX STI turbo pretty cheap (they are tiny). Run the basic Mr Turbo style pipe configuration but instead of carbs we use a log intake manifold and a honda civic throttle body. Run an o2 sensor in the exhaust, and use a mega squirt computer to control the whole thing.

so here are my questions:

1) is it really that important for the FI to control the timing? I think I can just put a dyna S in and leave it at that. Do turbo motors run more ignition retard than stock or can I keep the stock curve?

2) TPS and O2 sensor ought to cover all the bases - is there another sensor I should be thinking about?

3) how do I figure out what size turbo is too much for my engine? I can get huge garrett ball bearing dual scroll turbos but I know that this will give me a powerband like a 2 stroke.





Edited by - Geeto67 on Jan 31 2007 01:14:30 AM

 

[el kabong]

quote:
Here is the basic premise, he works for subaru so he can get me a stock WRX STI turbo pretty cheap (they are tiny).
so here are my questions:

1) is it really that important for the FI to control the timing? I think I can just put a dyna S in and leave it at that. Do turbo motors run more ignition retard than stock or can I keep the stock curve?

2) TPS and O2 sensor ought to cover all the bases - is there another sensor I should be thinking about?

3) how do I figure out what size turbo is too much for my engine? I can get huge garrett ball bearing dual scroll turbos but I know that this will give me a powerband like a 2 stroke.

Edited by - Geeto67 on Jan 31 2007 01:14:30 AM

the subaru turbo will be too large. even the GT15 turbo from an old 1.5 litre hyundai scoupe is too large. this chart suggests a GT12:
http://img299.imageshack.us/my.php?image=turbospecsob3.jpg

in an automobile, ignition retard is crucial *under boost* to avoid detonation. if you ran more retard all the time, the bike would be a slug in regular nonboost driving.

IMHO the recent stuff i've read about turboing a street m/c tells me it's not worth it. a few factory turbo bikes were available in the 1980s but were soon abandonned for bigger displacement.

 

[jbranson]

I remember my drag racing days back in the 70's...there was a saying. "There is no replacement for displacement"
JohnnyB

 

[speeds]

k so if u r set on turboing a bike, everywhere i have read carbutation is better although fuel injection isn't bad there is just more chance of a power gain with a carb because fuel injection might not be able to adjust that much for your fuel to air ratio, and i have also read in another fourm that a turbo off of a 1987- 88 ford thunderbird turbo coup works pretty well, but that was on newer bikes, i duno all i can say is tryal and error, lol i duno i am not a wiz with this kind of stuff by any means so if i am wrong feal free to correct me, that is just what i have read and comprehended

the lil guy

 

[Geeto67]

Turbo kits were plentiful in the 1970s for motorcycles, and a lot of people felt they were ideal for those bikes because they ran low compression from the factory (most are around 8.5:1, not like todays bikes which are 12:1).

The factory turbo bikes (gpz750 turbo, cx500 turbo, cx650 turbo, Z1R turbo, Suzuki M85 Turbo) were a response to the DIY turbo kits of the 70s.

These 70's setups are not like todays setups - turbos back then were treated like they were exhaust driven superchargers. Todays stuff takes advantage of 60 years of engineering to run amazing amounts of boost (my buddy's civic is around 40 psi) that traditional roots style superchargers could not do, and they do it throu blowoff valves, tuned intake and exhaust, piping, and vast computer controls.

I can build a setup copying the Mr. Turbo setups of the 70s to run 6 psi, not fool with the ignition timing, and retain the factory carbs, and probably make about 80hp with it. To me that seems like a waste of 30 years of turbo advancement.

I figure with FI and the proper controls I can run 12-14 PSI and make 100hp or more on a SOHC cb750 and still have a streetable bike. That sounds like a lot of fun.

SO I guess the next question is, how do I convert a points ignition bike to run computer controlled ignition timing? How about if I run a Dyna 2000 setup?

Branson - I am very, very familar with that saying. I'm just bored and big bore kits are...yawn...not the stuff to pipe dream about.

 

[revtune]

If you are going to go with fuel injection, you'll also need either a MAP sensor of MAF. MAP is Manifold Air Pressure, and MAF is Mass Air Flow. The MAF is the one you really want, and having one or both of those is critical because otherwise the engine wont know how to adjust the fuel because it has no clue how much air it is receiving.

Timing is important as well, but I believe that with the right tuning and the proper setup, you may be able to get away with a set timing. But, it will probably give you poor performance when you're not boosting. If you only plan on running a small amount of boost, you may be able to get away with the stock timing. Also, if you are planning on boosting, I'd look into dropping the engine's compression ratio a bit. Dependent upon how much you're looking for.

Cams are also a big factor. I don't know what cams are available for that engine, but NA cams with alot of overlap will give you less power.

All in all, I say do it.


Edited by - revtune on Jan 31 2007 2:15:38 PM

 

[fuzzybutt]

carpy has had turbo cb750s in his shop before, he might be able to steer you in the right direction maybe

some people are like slinkies, basically useless but they still make me smile when i push them down the stairs

 

[Geeto67]

quote:
If you are going to go with fuel injection, you'll also need either a MAP sensor of MAF. MAP is Manifold Air Pressure, and MAF is Mass Air Flow. The MAF is the one you really want, and having one or both of those is critical because otherwise the engine wont know how to adjust the fuel because it has no clue how much air it is receiving.

Timing is important as well, but I believe that with the right tuning and the proper setup, you may be able to get away with a set timing. But, it will probably give you poor performance when you're not boosting. If you only plan on running a small amount of boost, you may be able to get away with the stock timing. Also, if you are planning on boosting, I'd look into dropping the engine's compression ratio a bit. Dependent upon how much you're looking for.

Cams are also a big factor. I don't know what cams are available for that engine, but NA cams with alot of overlap will give you less power.

All in all, I say do it.


Edited by - revtune on Jan 31 2007 2:15:38 PM

I have a Mustang 5.0 MAF throttle body (which is way to big, but when you are boosting does that matter?) and a Honda civic B18 MAP throttle body to work with.

The Mustang FI uses a hall effects distributor which I have no idea how to work (I had an idea though - they made four cylinder mustangs maybe I can rob one and install it using the spare ARD mag kit I have). The Dyna 2000 has four different ignition curves and maybe if I can program the computer to switch between maps I can change the timing (I have no idea if this is even possible, just speculating).

I can get cams custom ground so that is something else to figure out.

Stock compression is 8.5:1, that seems pretty low to me for 12 psi on a pump gas motor. My fathers supercharged harley runs 8.5:1 on 8psi, and seems to do ok.

keep the info coming.

Edit: just got off the phone with some guys I know who use the megasquirt computer and they tell me that the computer can control timing and uses the stock points as a reference for crank position so that may solve the problem.

Edited by - Geeto67 on Jan 31 2007 7:33:16 PM

 

[ttilghman]

Geeto-

check a buddy of mine's site:

http://www.stabbarps-auto.com/welcome.html

he's this crazy swede who's turbo'ed and mega-squirted a few of the modern bonnevilles. i've tried to get him to chime in here from time to time, he's got the whole carbs conversion pretty much figured out....may be worth sending him note. he's done some pretty brilliant stuff for the 1/8mile, including beating a hayabusa for the season final a couple years ago.

-tt

 

[speeds]

k so u guys got me a lil curious about turbo chargers, lol don't have ne desire to put one on my bike or ne thing, just had a quick question, i understand how a turbo system works all except for one thing, and u can call me a dummy for this but how do you deliver oil to a fluid bearing turbo??

the lil guy

 

[UngaWunga]

Megasquirt controled two-stroke...

http://www.rzrd500.com/phpBB2/viewtopic.php?t=1899

I think he also used it to control the timing. Not sure if the powervalve was also controlled by it.

Maybe you should tackle one thing at a time. Fuel inject the bike first, then worry about the turbo. The fuel injection is a big enough project alone.

 

[chetriply]

Check out RB racing if you want some good turbo advice. They've been buildin land speed racers for a long time.

 

[chetriply]

Sorry, forgot to include the site address. www.rbracing-rsr.com

 

[Geeto67]

quote:
k so u guys got me a lil curious about turbo chargers, lol don't have ne desire to put one on my bike or ne thing, just had a quick question, i understand how a turbo system works all except for one thing, and u can call me a dummy for this but how do you deliver oil to a fluid bearing turbo??

the lil guy

On the cb750s, the old turbo kit used to have a take off and return line plate that bolted on just before the oil filter housing. These plates are still available since they are also used to mount oil coolers.

I thought about doing one then the other, but then I realized that the whole reason for EFI was to be able to take advantage of modern turbo technology (like boost controllers and such), and if I am going to have 1000 hours in figuring out the maps, I might as well spend that time with the turbo in place. This is a long term project (which is way underfunded at this point) so I would rather spend 1000 hours tuning than 2000hours.

So reading more about the megasquirt last night, it can control individual coil packs and use the points as a reference. This makes the varible timing a real thing.

 

[UngaWunga]

quote:
So reading more about the megasquirt last night, it can control individual coil packs and use the points as a reference. This makes the varible timing a real thing.

Sounds like you should ditch the points and go for a Pertronix or something... anything to get rid of that variable.

I mean, if you're going to go through the trouble of setting up an exact thing like fuel injection, why hamstring it with points being the ignition trigger?

 

[Geeto67]

quote:

quote:
So reading more about the megasquirt last night, it can control individual coil packs and use the points as a reference. This makes the varible timing a real thing.

Sounds like you should ditch the points and go for a Pertronix or something... anything to get rid of that variable.

I mean, if you're going to go through the trouble of setting up an exact thing like fuel injection, why hamstring it with points being the ignition trigger?

I thought about that - Dyna S is a points replacement system and I have a spare. Originally I thought of a dyna 2000 because of its adjustability but if the EFI comp is going to control the timing then all I need is a trigger.

 

[UngaWunga]

quote:
I thought about that - Dyna S is a points replacement system and I have a spare. Originally I thought of a dyna 2000 because of its adjustability but if the EFI comp is going to control the timing then all I need is a trigger.

which the Dyna S pickup can probably do for you...

 

[Geeto67]

here's a question - how crutial is a tempreature sensor to an EFI system. We are talking an aircooled engine here so there isn't going to be any good way to get a temp reading.

I could convert the engine to Water cooling but then I might as well be turboin' and injectin' the fzr600 motor I have under the desk.

BTW I think that is another neat project, fzr600 motor in a real old jap chassis. I saw one in a 73 kawi h1 frame once a long time ago and it was really really fast. Faster than a stock H1 and a stock fzr.

 

[polytoxic]

Temp readings are very important in fuel injection. The first and most important would be the cold start enrichment map. The map takes temp readings and enrichens fuel as you program it so that it will start and run while warming up just like a choke circuit does on a carb. You can get a temp reading from a thermo sensor on the cylinder head and program the cold start map in mega squirt to work properly. The most efficient way to do it would be to get a modern 600cc or 750cc fuel injected motorcycle throttle body complete with fuel injectors and throttle position sensor. Use a MAP sensor as required by megasquirt which is an oem GM sensor, get a tach signal and a crank signal. Also the closed loop feature on megasquirt allows for a learning mode with the newer software so you can actually do full rpm pulls and as you do that the o2 sensor samples the air/fuel mixture and will adjust the map in small increments, just repeat until it runs well.

As far as turbo choices, most car turbo's in fact will be too big. Some people have had luck with SAAB turbos in larger 4 valve motors, we are using a smaller modified Rajay turbo on a vt750 shadow and have 80 wheel hp up from 43 stock hp using one HD cv carb. We are currently building a BSA 500cc goldstar turbo land speed race bike with one turbo from a twin turbo Mitsubishi 3000gt VR4 and have managed around 6psi from it and we are using another HD carb. We are still in tuning r&d so we don't know what to expect hp wise.

 

[Geeto67]

quote:
Temp readings are very important in fuel injection. The first and most important would be the cold start enrichment map. The map takes temp readings and enrichens fuel as you program it so that it will start and run while warming up just like a choke circuit does on a carb. You can get a temp reading from a thermo sensor on the cylinder head and program the cold start map in mega squirt to work properly. The most efficient way to do it would be to get a modern 600cc or 750cc fuel injected motorcycle throttle body complete with fuel injectors and throttle position sensor. Use a MAP sensor as required by megasquirt which is an oem GM sensor, get a tach signal and a crank signal. Also the closed loop feature on megasquirt allows for a learning mode with the newer software so you can actually do full rpm pulls and as you do that the o2 sensor samples the air/fuel mixture and will adjust the map in small increments, just repeat until it runs well.

As far as turbo choices, most car turbo's in fact will be too big. Some people have had luck with SAAB turbos in larger 4 valve motors, we are using a smaller modified Rajay turbo on a vt750 shadow and have 80 wheel hp up from 43 stock hp using one HD cv carb. We are currently building a BSA 500cc goldstar turbo land speed race bike with one turbo from a twin turbo Mitsubishi 3000gt VR4 and have managed around 6psi from it and we are using another HD carb. We are still in tuning r&d so we don't know what to expect hp wise.

So how do I take a temp reading on an aircooled bike? What is the sensor I use? There are plenty of aircooled EFI bikes and even cars (the last of the old skool beetles made in mexico)can I use one of those?. Can I use an EGT reading or will it be better to convert to liquid cooiling? Or can I take a reading from the oil temp?

Also, why a MAP GM sensor and not a MAF Ford sensor? Seems the MAF sensor would be the easier way to tune since I am not backing into the number.


Edited by - Geeto67 on Feb 02 2007 9:25:22 PM