Imperial Home Page -> Repair -> Fuel System -> Fuel Injection Conversion
Thanks to IML member Brad Hogg we now have a do it yourself guide to installing a Holley Projection system.... What follows is a sort of diary written by Brad as he progressed through this project. We have also included a few pictures and the Holley Projection Manual. Thanks to Brad for a great addition!
Saturday, August 05, 2000 11:12 AM
I took the plunge today and bought a carburetor gasket in preparation to install the Projection. I got the Thermoquad off ok without loosing any of the small parts. I then test fitted the Projection to the manifold and found that the bolt pattern on the Projection (Holley) is somewhat larger than that of the manifold (Carter). What to do now? Can I get an adaptor plate? Should I consider replacing the manifold? I expect not. What do you suggest? I will be calling my auto parts store to ask their advice also. I called the local speed shop, went down and he sold me ($55.00) an adapter plate that looks like exactly what I need.
Saturday, August 05, 2000 4:38 PM
I finished the installation at 6:15PM (just as the rain began) That took about 8 hours including three parts runs. There are a few items yet to be tended to:
ECU = Electronic Control Unit
TBFI = Throttle Body Fuel Injector
I wasn't able to connect the wire that is supposed to go to the starter solenoid as it began raining and I didn't want to get under the car at that point. That connection simply tells the ECU to pulse the injectors while you are cranking the engine. I need to connect the tranny kick-down as the TBFI unit is a little different and I was unable to connect it. The instructions suggest I buy a special Holley part for my Chrysler for this. I might. It works fine for now, I just have to do a manual shift down if I want to accelerate hard. I need to route the cable that goes to the ECU, whereby one can tune the unit while driving, through the firewall. The connector is about 1.5" long if anyone knows of a good place to go through the firewall on a 74 to 78 C body. I currently have it coming out under the rear of the hood and in through the passenger window. Not too bad except I can't use the wipers. (see above comments about rain). I want to get just a little more fuel line and do some house keeping in the rear near the fuel pump. I also need to find some rubber to mount that pump on as it is a noisy sucker. Other than that, the installation went pretty much as the instructions stated. I did end up using the carbon canister line that goes into the top of the fuel tank for the return fuel line. It is all steel and seems to work great. The canister is disconnected now. The stock mechanical fuel pump is just running free at the moment. Can anyone comment on if this might damage the pump? I know I can get a block off plate for the engine and just remove the pump all together. I might just do so. One note about the return fuel line. I noticed what looks like a check valve in the carbon canister line just under the front passenger door. I didn't want that obstructing the fuel flow so I just bypassed it with a short piece of fuel line. I'm glad I bought twice as much line as I thought I'd need as I still need a little more to finish the job properly.
The fuel pump block off plate is included in the kit and is not very hard to install except that the bolts that hold the fuel pump on are longer than those required to hold the plate on so you need to get shorter bolts.
August 9, 2000
I was just getting the Projection tuned when yesterday the diaphragm in the fuel regulator broke. It is now spewing fuel into the engine at a rate even higher than the Thermoquad was capable of. Scary huh? I have to order the repair kit from Holley (2 weeks) and pay another $40.00 for it. The saga continues...
The diaphragm did NOT break as Holley tech support suggested. The only problem was excessive pressure at the TBI.
August 11, 2000
Got the regulator diaphragm changed and it is no longer spewing fuel into the engine but there seems to be another problem. One or two of the injectors are not functioning properly. The rear-right injector looks to be delivering fuel like the rest at idle but when I open the throttle slowly, it soon stops delivering fuel until I have the engine at about 3000rpm and then it begins to deliver fuel in a random and uncontrolled manner. I have also noticed a similar situation with the front-left injector. I don't know if this is a mechanical problem with the injector or an electronic problem? I have read in the troubleshooting that I can swap two injectors to see if the problem follows the injector but I am not sure how to get the injectors out to swap them. Any ideas? Are all 4 injectors supposed to always deliver fuel or is there a situation where two may be shut off such as at idle?
This problem was also caused by excessive pressure at the TBI.
August 14, 2000
A call to Holley tech support this morning lead me to my mistake. The car is running very nicely now and just needs to be fine tuned for that silky smooth power that I am sure this thing can deliver.
TBI=Throttle Body Injector
EFI=Electronic Fuel Injection
The problem was that the TBI was getting way too much fuel (fuel pressure) and was flooding the engine. This was caused by a poor return line to the gas tank. The fact that that line was poor is strictly my fault. (against my better judgment) I used the vacuum canister line that goes from the engine room to the fuel storage for the fuel return line from the TBI. Being not so knowledgeable, I didn't know (nor did I consider) what was at the point where that line enters the fuel tank. Either there is some check valve or it is simply plugged. The electric gas pump, that sits at the rear of the car, could not pump fuel back to the tank after the TBI portioned off whatever it wanted to use at any given moment. Therefore, the TBI was flooding the engine and causing all kinds of grief. I actually 'burned' 89 litres of gas in about 250 km of city/highway driving. I mean, this is ridiculous even for THAT car! I have always thought that the toughest part of this project was getting that return line back into the tank. Well, this is how I solved that problem. There is a short piece of gas line that runs from the tank to the gas pump. This line is not under pressure as the gas pump simply draws gas out of the tank and then pumps it forward to the engine room for processing. The return gas line is not really under pressure either but there is a considerable flow to it. I filled a 3 gallon jerry can full from this line while the car was idling and that only took about 2 minutes. The fact that there will never be more fuel flowing back from the TBI than the gas pump it drawing out of the tank got me to thinking that a 'T' in the line between the tank and the pump was the answer. I now have the return line not delivering fuel back to the tank but to the tank outlet before the pump. This arrangement seems to work very well. I can hear the gas pump working much better now, and I can also see the spray of the injectors is lighter and more uniform. The car starts, runs, and drives way better than it ever did. I am anxious to tune it and calculate fuel economy. I did buy enough gas line today to replace the vacuum canister line as return line. I will probably go ahead and replace it but who knows, if it works, don't fix it. I was informed by the guy at the UAP/NAPA store, where I'd purchased nearly everything I used for the project, that the 3/8" gas line that I bought was not rated for the pressure that an EFI system produces. He can sell me the correct (180 psi) line for about $65.00/10' length. I'll probably need 2. This, I did not expect and will have to save up again for it. The line I bought is rated at 50 PSI. I have yet to look into how much pressure this thing actually produces.
The Holley Projection only produces 15 PSI so the 50 PSI rated line should be fine. I was, however, advised to replace all my gear type hose clamps with the special EFI type clamp. I will be doing this.
August 17, 2000
I am still getting rather poor fuel economy. It seems to have gone from 9 to 12 city to about 8. I haven't done any conclusive highway tests yet. The car is running much better now that I have a good fuel return line. I did do something today that seemed interesting. I wanted to take the gas lines off the TBI, so in order to release the pressure in they system, I simply disconnected the (power to the electric) gas pump while the engine was running so it would burn off the pressurized gas and then just run to a stop when it was all equalized. After I disconnected the pump, the car seemed to run better for the next few seconds. This made me think that maybe I still have a bit of a restriction in the return line and therefore presenting too much pressure at the TBI. I'm going to be doing a bit more testing to find out if this is in fact the case.
No problem found.
August 23, 2000
Well folks, I did a short test of the mileage on my Holley Projection 4 system installed on my 400cid RB in my 78 Newport 4-door. The results of this particular test are very good.
19.5 mpg (15 mpg US) probably 60% highway and 40% city.
I can live with that. I was previously getting about 15 to 18 mpg on the highway. This test was a rather short distance so it might be a little skewed. I traveled 104 km (64 miles) on 15 litres (3.3 Gal CDN). More results to come.
August 24, 2000
Today, I filled the ark for the second time since the real mileage tests began. On this fill, I traveled 163 km (101 Miles) on 32.4 litres (7.1 Imperial Gallons) of gas. That works out to about 14.2 MPIG (Miles Per Imperial Gallon), about 11.5 MPUSG (Miles Per US Gallon). This distance was travelling about 80% city and 20% highway. On average so far, I have travelled 267 km (165 Miles) on 47.4 litres (10.4 Imperial Gallons) for an average fuel consumption of 16.85 MPIG (13.5 MPUSG). This is up about 2 to 3 MPIG. I have only yet rough tuned the system and have not checked the fuel pressure. I intend to test the fuel pressure first and then set out to fine tune the EFI.
August 27, 2000
Well...here are the latest things I have learned about my Holley Projection 4. I took my car over to my cousin's shop and had his mechanic check the fuel pressure going into the TBI. The pressure was at 20 PSI. The manual states that the pressure should be 15 PSI. I pulled out my little hex wrench and adjusted it down to the recommended level. The car immediately began to run poorly as the ECU was tuned for the higher pressure. I attempted at that point to tune the ECU so at least the car would run. The car was at that point surging and flooding at idle and not running well at all when under load. Being unsuccessful, I decided to boost the pressure back up a little. Not having a pressure gauge of my own, I simply turned the regulator adjustment IN one turn. That should have brought the pressure back up toward 20 PSI where it was before I changed it in the first place. This did not really help make the car run better. I was at a loss. I had been noticing that the car ran much better when cold or when freshly warmed up. It had problems when it was fully warm or hot. I suspected a cooling problem and set out to discover what might cause this. I have no fan shroud on the engine at the moment so it does begin to heat up in traffic. This, as it turned out, was not the problem. (it is A problem but not THEE problem) Because I read the entire installation manual before beginning this project, I remembered reading some comments on the mounting of the ECU. I quote from page 11, "Find a suitable mounting location for the ECU in the interior of the vehicle such as under the dash, or under the seat. DO NOT mount the ECU where there is a possibility of exposure to extreme heat or moisture such as the engine compartment." I did have the ECU simply sitting on the front seat beside me so I could tune the unit as I drive. What I determined is, because we have been experiencing very hot weather here the past week or more, the ECU was in fact becoming very hot. It does produce a little heat itself but yesterday, at one point when the car wouldn't run, the ECU was so hot I couldn't touch it. I allowed the unit to cool down for a few hours and the car ran fine. I will now have to investigate getting a large heat sink to mount to the unit and possibly add a small cooling fan. This seems to be a very important point, keeping that thing cool. The entire installation manual can be seen below.
August 27, 2000
I was just out on a cool Manitoba Sunday afternoon tuning the Projection4. This is not a difficult task, but you need to do it in the right order. There are 6 settings on the front of the ECU. There are 2 settings to be done on the Projection TBI itself. Those are the idle setting and the fast idle setting. The former is simply a set-screw like any carburetor would have. The latter is an electric solenoid that is activated when you turn the ignition on while the engine temperature warrants a fast idle. To set the fast idle, one simply uses a 1/2" wrench to turn the solenoid's core in or out. I was thinking, it might be nice to connect a small toggle switch to that thing so I could activate a fast idle situation at any time I might want it, like when boosting another car etc. I digress. The 6 settings on the ECU are: Choke, Acceleration, Idle, Midrange, Power1, Power2. These are simply electronic potentiometers that turn about a 3/4 turn. One can lean the mixture by turning them CCW or richen it by turning CW. The instructions state that they should all start in the centre of their travel. Choke - Simulates a choke plate creating a rich mixture. This I have yet to tune. Acceleration - Simulates an accelerator pump. I have yet to see any real difference between a high and low setting on this one. Idle - Simply sets the mixture at idle. It is affected by the setting on the Midrange control. Midrange - This is the first setting to tune. Start by revving the warm engine to 3000 rpm and tuning it CCW just until the rpm begins to drop. Then set the idle. Repeat. Power 1 - Sets the mixture between 1500 and 5000 rpm. Power 2 - Sets the mixture between 1500 and 3500 rpm. You can see there is some overlap between the two Power settings. I did not see much difference in a high or low Power 2 setting. I will have to tune this more scientifically. After all this was done, I did a few 0 to 60 runs. The boat consistently made 0 to 60 mph in 18 seconds. I regret not taking a 0 to 60 benchmark while the Thermoquad was installed. I am interested to know if anyone else has any 0 to 60 mph results for their C bodies. The fun continues...
August 27, 2000
Suggestion from IML member Roger Cooper:
Why don't you temporally mount it on a piece of alum. plat about 12inches by 12 inches about 1\4 thick while your setting it up.
Reply from Brad:
Yes. I actually just had a friend suggest nearly the identical approach. I replied to him... The only mounting points provided are four plastic tabs one on each corner. I just have to be careful to keep it out of the sun until I can get a proper heat sink for it. I looked inside and there are four power transistors on each side, each screwed to the aluminum case for cooling. I think this is a good place to start
Question from Mike:
I have been thinking of putting a Holley Projection unit on one of my Imperials. More because they are already EFI and I like fuel injection, but would like a car that will be more dependable. Which Projection 4's are you using? Are you using the regular preprogrammed one, or the digital one that requires a laptop computer to fine tune and set up the correct performance curve desired? I like the digital one myself but it is worth a bit more. I think I might go with a Quadrajet from a late 80's Dodge Truck.
Reply from Brad:
I am using what they would call the Analog unit. It has a control box with 6 potentiometers in the face that you can adjust for different power bands etc. This is NOT the unit that is programmed via a computer. That unit would probably be an improvement except that 1) I don't have a laptop computer and 2) clicking away on a computer while you drive can't be as easy as turning a pot. I hope you can try this out. I think it would be an improvement on the stock EFI system.
Question from John:
Has your mileage improved?
Reply from Brad:
Slightly. Maybe 1 to 2%. (1 to 2 mpg) This alone does not make this project worthwhile. The PJ is not a progressive 4bbl so the potential for high fuel consumption is very real if one has a heavy foot.
September 22, 2000
Well...I have made a couple more improvements with the Projection. If you remember from a previous post, I was having problems with the thing flooding the engine. That was caused by two factors. 1) A small hole in the body of the PJ that the tech support guy told me was for manufacture only and should be plugged. I plugged it with some lead. 2) There are 4 o-rings that fit between the PJ body and the part that houses the injectors themselves. I'd taken that off as to not have it damaged when I needed to get a small part of the PJ body welded. When I put that part back on, the aluminum is rather sharp and it cut some bits off the o-rings. I got new ones but rather simply sealed the part with Permatex. The second problem allowed gas to run down the PJ and into the engine after I turned it off. This caused very hard starting. The engine starts in about 1/2 revolution now :) These two fixes have made this thing run nearly perfectly. The only issue at the moment is getting the idle mixture set just right and the 'choke' setting so it will idle nicely when cold. Because there are fairly short periods of time to test the 'choke', this is a slow process. After it is all just so, I'll do all the housekeeping. This includes a number of items. Route the ECU cables through the firewall and mount the ECU in the cab. Mount one or more heat sinks to the ECU to keep the thing cool in summer. Consider a small fan as well. Replace most of the hose clamps with those special FI clamps for safety. Consider buying the Morosso adaptor so I can run the return gas line back into the fuel filler tube rather than the 'T' just before the gas pump. (even though the 'T' seems to work fine) Re-mount and insulate the gas pump cause that sucker's noisy! Clean up the throttle linkage a little and adjust the kickdown points a little. (they shifted too low then I made them shift higher and now they are a bit too high) Paint a red racing stripe on the car. Get a sponsor. Hit the race track. Hmmmm. Got carried away a little.
April 28, 2001
Well folks! It's been a long time since I've toyed with the fuel injection on my car. I've been messing with it again for about a week now. I continue to learn and to fine tune my installation. This is what has happened this week.
I started driving the car to work every day on about April 20th. It was running ok but I was not pushing it at all, driving very smoothly and not accelerating hard at all. The commute is about 15 minutes so it was hardly a gruelling test. On April 26th, I got a little more daring. On the way home from work, I started doing some short WOT bursts and it performed acceptably. After about three of these however, the car refused to run. It was obviously flooding so in just a matter of 15 minutes or so, it started again and ran poorly but ran. I pulled it into the lower gears to keep the revs up as it seemed to like that. On April 27th, I checked the setting of the TPS (please see my DIOYS page, below, for acronym definitions) and it was WAY out. It was effectively telling the ECU that the throttle plates were way farther open than they really were. I corrected this and pretty much moved the TPS to the lowest setting to lean out the overall power band as much as possible. This helped the engine run much better but it still wasn't consistent. There was some component (or many of them) that was not performing consistently because I could make it run quite nicely and then after a little while, sometimes minutes, sometimes hours, it would run crappy again. After discussions with various people, my head was full of possibilities.
One suggestion was to change the power supply for the electric fuel pump (located at the rear, near the tank) so that it was powered by a relay rather than directly from the ECU. This would allow the pump to get stronger, cleaner +12V to run on. I did that and it seems to have helped. The pump now runs at a much more consistent speed.
Another suggestion was to plumb in the fuel return line directly into the tank rather than into the fuel line just between the tank and the pump. This was actually something I've wanted to do from the start. I was afraid that the return line might be feeding air into the gas pump and then eventually up to the TBI. Today, I took out the fuel tank s ender and drilled a hole in it and soldered a short piece of steel fuel line into it. I pointed it upwards upon the advice of a lister so it would have little likelihood to siphon. This was not really very hard to do. The hardest part was getting the sender out without dropping the tank. You can almost not do that due to the proximity of the frame rail. I got all that back together, it doesn't leak, and it seems to work just fine. I went for a short ride and the car performed admirably. Power is still at an acceptable level but I'm sure there are things to do yet.
Another suggestion is that the fuel tank may not be venting properly, thus causing a high or low pressure situation, affecting the fuel delivery. So far, I can quite certainly say that the thing is running plenty rich. The black soot at the tailpipe is a sure fire clue. To test this theory, I've remove the gas cap for a couple days to see if that helped. I have about 1/4 tank of fuel so I don't anticipate any spillage. I know in the past, the tank will indeed pressurize quite substantially if it is low. When I take the cap off with 1/4 tank or less. The pressure blows out at me. I never thought this was a very good thing. I am told also that a new, vented, gas cap is in order. Cheap fix probably.
Another thing I need to test today is the engine temperature sensor. The ECU is connected to a temp sensor so it can adjust the fuel delivery for varying engine temps. This is also used to allow the ECU to energize a 'fast idle' solenoid when the engine is cold. I noticed the other day (when it was actually cold) that the solenoid was not being energized. I connected it directly to +12V and it worked fine so apparently the ECU is not turning it on for some reason. I don't know if the tem sensor is bad or the ECU had gone south. I am hoping it is the sensor. Otherwise, the ECU will likely think the engine is ALWAYS cold and over fuel it constantly. The solenoid worked well last summer so am not sure why it doesn't work now.
May 19, 2001
Yesterday I finished a 2 day project of removing the Projection from my car. This decision was a long time coming and was hastened by a recent unfortunate situation. I came to realize that 900 cfm was not meant for a 400 smog motor. If it were installed on a higher compression 440 for example, it would perform well. The manual for the Projection clearly states that it is for an engine that is at least 300 cid and 350 HP. My 400 qualifies in the cid category but not even close in the HP. I've come to realize lately that even though I had gotten it to run better than ever, it will never run just right. The decision to remove it was hastened by an unfortunate incident Wednesday evening. I went to the store, and when I went to come home, as I drove across the parking lot, the car simply died. It would crank but no fire at all. I soon found the problem. I noticed that there was no fuel being delivered by the injectors. That was because the relay I had installed for the fuel pump, shorted out against a metal object and stopped the pump. I hoped that was all it had done. I am not confident that it is. I got the car running again but it was not running well at all. I think it went into "limp home" mode. It would start up and then surge over and over at idle and was impossible to throttle up. I got it home (slowly) and decided that since, at its best, the PJ had never run as well as my Thermoquad, that it was time this experiment came to an end. In one more week, I'll be relying on this car for my daily driver again and I needed it to be much more reliable than it has been. The good news is, with the TQ now reinstalled, the car runs beautifully. I've taken 2 seconds off my 0 - 60 time so while this is NOT a race car, it shows me that it is performing just fine as compared to the 18 second PJ. The Thermoquad always got better fuel economy than the PJ too. I am looking forward to this as gas has just gone up to over $0.75/litre here and promises to go even higher :( I also notice a couple things about the way the TQ had been set up and have corrected those with the help of my NYB parts car. For one, the choke rod was broken at the bottom so, no choke. A mechanic had disconnected the rod that actuates the vacuum secondaries. I replaced it with the (slightly longer) one from the NYB and it works great! With all the heat riser stuff connected and the new, dual snorkel air cleaner, the engine seems quite happy. The demise of this project was simply a combination of two things. A mismatched application and an unfortunate mistake on my part. I have got to say though, I learned a ton of stuff through this project. At this point, I am thinking that if the ECU is not fried, it might make a nice addition to that 68 440 my brother still has.
(click on small images to view full size versions)
The Holley PRO-Jection 4 Manual
(click on small images to view full size versions)
|Cover||Table of Contents||Page 1|
|Page 2||Page 3||Page 4|
|Page 5||Page 6||Page 7|
|Page 8||Page 9||Page 10|
|Page 11||Page 12||Page 13|
|Page 14||Page 15||Page 16|
|Page 17||Page 18||Page 19|
This page was last updated October 2, 2003. Send us your feedback, and come join the Imperial Mailing List - Online Car Club