Home 
Home Search search Menu menu Not logged in - Login | Register
> Jensen Healey & Jensen GT Tech > Ignition > Fuel Injection KIt for 907

 Moderated by: Greg Fletcher Page:    1  2  3  4  5  6  Next Page Last Page  
New Topic Reply Printer Friendly
Fuel Injection KIt for 907  Rating:  Rating
AuthorPost
 Posted: 11-29-2008 01:53 pm
  PM Quote Reply
1st Post
jgreen
Member
 

Joined: 01-28-2007
Location:  
Posts: 58
Status: 
Offline
HI Everyone,

I have some potentially exciting news for stromberg equipped 907's. I live about 20 miles form the developers house. It seems promising as he has developed ( in his own machine shop with all the computer driven machines) a kit using the proven delco fuel injection system in all gm cars. They are in current operation with tr-6's ( 6 cylinder) and mg ( single card). He is currently developing a kit  for 4 cylinder dual carbs. ( 175 srombergs)

I should be ready in the spring. If you are worried aout the stock look........ don't worry ....... the system fits inside the strombergs and uses all the original filters/manifolds/ etc. You really can't see it. Everyone would think you are carb equipped.  I have a Jensen GT w/ strombergs and he has offered to install the first one as soon as my gt is up and running.

The only modificationmust be done to the fuel tank. It needs a return line fitting for the extra fuel to return to the tank.

The kits come with everything  except fuel line, pretty amazing...... even a check engine light.

If you are interested, please email me and I will give you details and try to answer as many questions as possible. The developer i think is curious tro know how much interest is out there for a kit for the Jensen Healey like this. 

Regards

John

 

 

 

Everyone knows the convience of fuel injection so I won't so into it here.

Back To Top PM Quote Reply  

 Posted: 11-29-2008 06:54 pm
  PM Quote Reply
2nd Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
Does the Delco system handle the problem of TBI and siamese headers? I read about someone trying to get a MegaSquirt system tuned in for a Mini with a siamese intake header, and he was having all sort of problems, where on cylinder would go dead lean while the other was way rich. Airflow in those throttle bodies is not evenly distributed over two revolutions.

--
Joachim

Back To Top PM Quote Reply

 Posted: 11-30-2008 04:06 am
  PM Quote Reply
3rd Post
Jensen Healey
Super Moderator


Joined: 03-11-2005
Location: San Anselmo, California USA
Posts: 983
Status: 
Offline
Is this the Patton kit with JH maps?

http://www.pattonmachine.com/

Back To Top PM Quote Reply  

 Posted: 11-30-2008 02:09 pm
  PM Quote Reply
4th Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
It does sound like it. For those interested, there is an explanation of the siamese problem here:

http://www.starchak.ca/efi/siamese.htm

Edit: There may be a difference between a Mini head and the Stromberg manifold, but I think you still get the same problem. Knowing which tube the charge lands in is a bit of a guess, and probably not made easier by the propagation time from the valve to the throttle body, if fuel is delivered into the manifold between the suction pulses, it will always end in the inner cylinders. On the TR6, you don't have the same problem, since there is always 240º between the suction pulses (given 153624 firing order, which I believe it has).

I just get the feeling that this is a lot more difficult than it looks at first glance.

--
Joachim

Last edited on 11-30-2008 05:46 pm by subwoofer

Back To Top PM Quote Reply

 Posted: 11-30-2008 06:09 pm
  PM Quote Reply
5th Post
jgreen
Member
 

Joined: 01-28-2007
Location:  
Posts: 58
Status: 
Offline
Yes, it is the patton kit. You are proabaly right, it could be lots of problems if this issue isn't solved. I will talk to him next week and see it he is aware of it. There looks like a lot of promise if this does work well. I am not sure how much of a problem the lotus setup will be however.

FWIW, there is a detailed installation with photos on the site. He installs it on the tr-6. It's pretty much bolt on with some fine machining added in to fit the strombergs... very well done.

Back To Top PM Quote Reply  

 Posted: 12-03-2008 06:43 pm
  PM Quote Reply
6th Post
pc
Member
 

Joined: 03-16-2005
Location:  
Posts: 58
Status: 
Offline
The “siamesed port problem” pops up when trying to use timed, multi-port injection on cylinder heads with siamesed intake ports (which, by definition, aren’t multi-ports).

The 907 has individual intake ports so you can use timed multi-port injection.  Later generation Lotus engines do just that. 

But all that’s academic since the Patton Machine system is a throttle body injection setup, a.k.a. an electronic carburetor style of system.  As long as the injector provides a uniform “fog” to the throat of the throttle body and the intake manifold plenum the cylinder balance should be the same as with carbs.




 

In addition to the return line and you might need to install a larger supply line too.  I’m not sure if the stock line can handle the higher flow. 

You’ll also need to find a place to mount the ECU, run the harnessing, install the high pressure/flow pump, etc, etc. 

 

PC.

Last edited on 12-03-2008 10:15 pm by pc

Back To Top PM Quote Reply

 Posted: 12-13-2008 08:27 pm
  PM Quote Reply
7th Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
pc wrote:

But all that’s academic since the Patton Machine system is a throttle body injection setup, a.k.a. an electronic carburetor style of system.  As long as the injector provides a uniform “fog” to the throat of the throttle body and the intake manifold plenum the cylinder balance should be the same as with carbs.


I dare to disagree that it should be the same as with a carb, and will claim that it has all the same problems as the BMC/Rover heads, compounded by the long throat, but I will admit that it is based on logic, not actual experiments.

There is one significant difference between an actual carb and an "electronic carb" or TBI system, which is likely to bite in this particular configuration. In a carb, fuel delivery is a function of air mass flow, while in the case of EFI, fuel delivery is a function of time. The two are effectively one and the same if air flow is evenly/symmetrically distributed for the four-stroke cycle, but in the case of a dual carb 4 pot, it isn't.

The distance from the intake ports to the TB only makes the problem more difficult to solve, because of the pressure propagation time lag, causing the ideal moment for injection to move around with respect to the crank angle, RPM dependent.

Of course, I could be completely wrong, the length of the intake runners could smooth out the air flow speed fluctuations, causing the TBI system to behave closer to an actual carb, but the fact that intake runners are tuned to length for different RPM power peaks suggests to me that the air flow is not evened out, but pulsates.

Just some late night speculation on my part, does anyone else have any opinions?

--
Joachim

Back To Top PM Quote Reply  

 Posted: 12-17-2008 06:02 am
  PM Quote Reply
8th Post
pc
Member
 

Joined: 03-16-2005
Location:  
Posts: 58
Status: 
Offline
Fuel delivery in a carburetor is a function of air volume, i.e. Velocity x Area.  In the case of a constant area carb (Weber, Delorto, Holley, Mikuni,…) fuel flow varies with velocity and the venturi cross section is fixed.  In the case of a constant velocity carb (Zenith-Stromberg, SU, Hitachi,…) fuel flow varies with throat cross section and velocity is fixed.
 
Because carburetors deliver based on volume, changes in air density (mostly due to altitude and ambient temperature) cause changes in A/F ratio.  Factory tuned street engines are optimized for a “nominal” conditions so that they’ll operate reasonably at extremes.  Highly tuned race engine carbs are re-jetted for running on tracks at different altitude from where they were originally dyno’ed/tested/tuned.
 
 
EFI fuel delivery is a function of whatever the system designers builds it to be.
 
basic electronic control schemes commonly used include:
 
Alpha-N – fuel flow is based on rpm’s and throttle angle.  This scheme suffers from the same sensitivity to changes in air density as carburetors but is simple and not sensitive to low manifold vacuum.  It’s often used on motorcycles and racecars, which operate at wide open throttle (WOT) much of the time.
 
Speed-Density - fuel flow is based on calculated mass air flow derived from rpm’s, measured MAP (manifold absolute pressure) and air temperature combined with a lookup table of estimates of combustion chamber volumetric efficiency.
 
Mass Air Flow - fuel flow is based on measured mass air flow
 
Those are the basic schemes but real systems can include additional sensor inputs to “tweak” the controls to more precisely approximate actual fuel demand.
 
Real systems often combine modes.   They may for example operate as speed-density at part throttle and/or midrange rpm’s then transition to alpha-N at WOT and high rpm’s.
 
 
Typical throttle body injection systems are called “electronic carburetors” because they operate like carburetors, introducing fuel at the throttle plates, mixing the air and fuel in the intake manifold, fully wetting the intake tract and are not synchronized to valve operation/piston position.
 
The intent is to have a uniformly mixed air/fuel charge in the intake tract that is available any time the intake valve opens, making fuel intake to the combustion chambers independent of valve timing.  “Electronic carburetor” TBI systems can use any of the three control schemes.
 
Typical EC/TBI systems do fire injectors on a timed cycle but not synchronized with valve opening.  They’re timed to regulate fuel deliver.  Since injector solenoids are simply either on or off they can’t regulate fuel flow in a variable manner.  They are either not flowing anything or they’re flowing all they can.
 
The control computer cycles them on and off so the total fuel flow is the average of the fuel delivered between the on and off periods.  This is called “pulse width modulation.” The farther up the intake tract the fuel spray occurs the more time it has to vaporize and mix to a uniform air/fuel charge by the time it reaches the cylinder.  If it is well mixed in the intake plenum it will be uniform in the intake runners to each cylinder.
 
A timed, sequential port injection system can base flow volume on any of the schemes but synchronizes the “on” signal for the injectors with intake valve operation.  These systems will have the injectors positioned very close to the intake port and include crank position sensing, not just speed sensing, inputs to the control computer.
 
Since the Patton Machine installation notes mention neither MAF nor crank position sensors but does mention MAP and throttle position sensors it appears to be a speed-density/electronic carburetor system.
 
 
pc.

Last edited on 12-17-2008 06:04 am by pc

Back To Top PM Quote Reply

 Posted: 01-21-2009 04:53 pm
  PM Quote Reply
9th Post
Chris E
Member
 

Joined: 11-23-2006
Location: Fremont, California USA
Posts: 12
Status: 
Offline
 I've looked at doing this for a while, But the timing problems already mentioned coupled with the fact that GM TBI fires the injectors on every other ignition pulse seems to rule out a "dual sidedraft" solution.

 Although not that elegant in appearance, it should function well and it greatly simplifies sensor installation etc. see pic.

I'll post more when I set up the surge tank & wiring, and give it a try.

Chris

Attachment: P1010057.jpg (Downloaded 844 times)

Back To Top PM Quote Reply  

 Posted: 01-23-2009 06:41 pm
  PM Quote Reply
10th Post
Greg Fletcher
Administrator


Joined: 03-11-2005
Location: Lake Nacimiento, California USA
Posts: 430
Status: 
Offline
Interesting, but I'm not loving the idea of a Stromberg fuel rejection system all that much. Strombergs are the first thing I want to remove on the 907. That Stromberg intake manifold is not the greatest piece of engineering to come out of the UK and getting access to the distributor is almost impossible.

The Euro intake manifold is what the 907 was originally designed to use. You could put together a used Dellorto conversion for less than half what an injection system might cost.

Back To Top PM Quote Reply

 Posted: 01-23-2009 08:33 pm
  PM Quote Reply
11th Post
Chris E
Member
 

Joined: 11-23-2006
Location: Fremont, California USA
Posts: 12
Status: 
Offline
 What I'm looking for is a modest improvement in performance & driveability on a stock engine with significant gains in efficiency.

 Cost is also another factor. The  GM 1227747 is one of the most common ecm's around (I paid less than $100 for the ecm and throttle body on ebay). I have less than $600 into everything required (a lot of homemade parts + ebay).

Back To Top PM Quote Reply  

 Posted: 01-23-2009 08:41 pm
  PM Quote Reply
12th Post
Greg Fletcher
Administrator


Joined: 03-11-2005
Location: Lake Nacimiento, California USA
Posts: 430
Status: 
Offline
Sorry Chris, I was talking about the Patton kit for $1,500-$2,000. Yours does seem economical at that price. I'd like to hear how it turns out and how it feels on the road.

Back To Top PM Quote Reply

 Posted: 01-24-2009 08:13 am
  PM Quote Reply
13th Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
Greg Fletcher wrote:
Sorry Chris, I was talking about the Patton kit for $1,500-$2,000. Yours does seem economical at that price. I'd like to hear how it turns out and how it feels on the road.

If the Patton kit is that expensive, I believe there are better ways to go about it, unless stock appearance is vital.

A complete set of DCOE style throttle bodies and injectors (by KMS) costs $1000, 25% VAT inclusive, in Norway, add a MegaSquirt or a Haltech ECU, Euro intake manifold, and the rest of the needed parts scavenged from your local breaker's yard, you should come out with a better result for roughly the same price.

Chris: please let us know how the TBI experiment works out, the idea may be worth working on, but I think it needs a bit in the looks department. Possibly something along the lines of the TR6 PI intake manifold, connected to a Euro mainfold?

--
Joachim

Back To Top PM Quote Reply  

 Posted: 01-27-2009 07:42 pm
  PM Quote Reply
14th Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
Finally had some time to cook up a reply to this:

pc wrote:
Fuel delivery in a carburetor is a function of air volume, i.e. Velocity x Area.  In the case of a constant area carb (Weber, Delorto, Holley, Mikuni,…) fuel flow varies with velocity and the venturi cross section is fixed.  In the case of a constant velocity carb (Zenith-Stromberg, SU, Hitachi,…) fuel flow varies with throat cross section and velocity is fixed.

It is actually a function of air mass, since fluid density is a part of the Bernoulli equation (describing the pressure change for a given change in flow velocity), and the reduced pressure above the jet is the driving force of fuel delivery. The need for retuning due to altitude and atmospheric conditions is most likely caused by non-ideal behaviour of a carb and the combustion cycle.

EFI fuel delivery is a function of whatever the system designers builds it to be.
 
basic electronic control schemes commonly used include:
<snip different metering strategies>


I am aware that there is a lot of different strategies used to calculate the pulse width for the injectors, but is does not change the fact that fuel delivery in any EFI system is strictly based on time.

The controller delivers a certain amount of fuel at a specified time window, regardless of the momentaneous flow of air. Because of that, you can run into problems if air flow isn't evenly distributed over the full 720 degree cycle of the engine.

If the air is flowing slowly past the injector at the time if fires, the charge is denser than if the air is moving at a higher speed. Since the two cylinders running off the same TB are spaced 180/540 degrees apart, not 360/360, it would be reasonable to expect that one cylinder will get a richer charge than the other, unless the volume of air in the common part of the manifold is much larger than the volume of the individual intake runners. Since the bifurcation is quite close to the carb/TB, I don't think that is the case for the Stromberg equipped 907.

Just thinking out loud...

--
Joachim

Back To Top PM Quote Reply

 Posted: 01-28-2009 05:43 am
  PM Quote Reply
15th Post
Chris E
Member
 

Joined: 11-23-2006
Location: Fremont, California USA
Posts: 12
Status: 
Offline
 I originally had plans to build a couple of sidedraft throttle bodies which would look cool & bolt right on, However after looking into it a little more there are a couple of issues that are hard to get around. As Joachim mentioned the speed/density changes will make it difficult to get a consistant mixture with a constant stream of fuel. But it gets worse than that. GM TBI works with pulse width modulation with one pulse every OTHER ignition event. Therefore it is possible for most of the fuel to go to #2 while #1 gets almost none. see pic. For it to work, the injectors would need to be timed to the intake events. 

 I believe that for a TBI to work correctly on a 4 cylinder both injectors need to fire into a single plenum. (this is what they're designed to do) otherwise we need to go to a port injection system. This is what prompted my quick & dirty design.

 That's my opinion anyway. I'll post further as my experiment gets further along.

Chris

 

Attachment: TBI Timing.jpg (Downloaded 54 times)

Back To Top PM Quote Reply  

 Posted: 01-28-2009 07:28 am
  PM Quote Reply
16th Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
You may consider getting a manifold, fuel rail and TBs from an S4 Esprit, I guess Mike at lotusbits.com could source a set. No S4 parts listed, but I guess he may be able to find some parts. That is a proper port injected setup, and fits the engine like a glove.

Don't know about price, though.

--
Joachim

Back To Top PM Quote Reply

 Posted: 01-28-2009 10:03 pm
  PM Quote Reply
17th Post
pc
Member
 

Joined: 03-16-2005
Location:  
Posts: 58
Status: 
Offline
When I was batting around the ideal of doing EFI on mine that’s the conclusion I came to.

The more I thought about it the more it looked like getting used Esprit manifold/plenum/throttle body bits was going to be a lot easier than piecing together other stuff or fabricating custom parts and probably a lot cheaper.

 

pc.

Back To Top PM Quote Reply  

 Posted: 01-29-2009 12:31 pm
  PM Quote Reply
18th Post
Jensenman
Member
 

Joined: 04-14-2005
Location: Columbia, South Carolina USA
Posts: 156
Status: 
Offline
I have Dellortos on my car. I still have the Stromberg intake sitting in my shop, every so often I look at it and wonder just how hard it would be to weld injector bungs into the tubes (the Dell intake already has them rough cast, BTW) and then add a couple of air doors in the stock Stromberg carb position. A Toyota MR2 etc of the '80's has a 'batch fire' ECM that seems like a perfect match for this type setup (or even use a Megasquirt) and a Geo Metro throttle body should be a close match to the Stromberg throttle diameter. Or even use a pair of motorcycle TB's.

Back To Top PM Quote Reply

 Posted: 01-29-2009 12:35 pm
  PM Quote Reply
19th Post
subwoofer
Member
 

Joined: 04-01-2008
Location: Sandefjord, Norway
Posts: 617
Status: 
Offline
I have no idea if adding injectors to that manifold is feasible, but if you go that route, you could just rip out the innards of the Strombergs and use them for throttle bodies.

--
Joachim

Back To Top PM Quote Reply  

 Posted: 01-29-2009 04:50 pm
  PM Quote Reply
20th Post
Jensenman
Member
 

Joined: 04-14-2005
Location: Columbia, South Carolina USA
Posts: 156
Status: 
Offline
The Strombergs could be easily replaced by a pair of 65mm throttle bodies which is a pretty common size. The stock 175CD is about 45mm so that would be a pretty big airflow boost. I'd rather do that than gut the Strombergs because that way I could easily install a throttle position sensor, idle air control, etc.

For comparison, 2 45mm Dellortos have 9.868 sq inches of throttle opening. 2 65mm throttle bodies would have 10.286 sq inches of throttle opening. The TBs might even be a bit large since the 45mm figure on the Dells does not take into account the choke size.

Weld in injector bungs are readily available: http://www.spectrum5racing.com/bungs.htm

 Hmmm. I'm starting to talk myself into it. :-D Once the rotary is installed in the other J-H and that's running, this could be my next project.

Last edited on 01-29-2009 06:06 pm by Jensenman

Back To Top PM Quote Reply

Current time is 05:24 pm Page:    1  2  3  4  5  6  Next Page Last Page    
> Jensen Healey & Jensen GT Tech > Ignition > Fuel Injection KIt for 907 Top




UltraBB 1.172 Copyright © 2007-2011 Data 1 Systems