Air in Mechanical Fuel injection System

Well, I had a Mercedes Diesel 300 that did this because the ex-wife ran it out of Diesel. They had to pretty much flush out the whole system because little pockets of air got it and sort of just oscillated back and forth rather than clearing themselves out. Plus it stirred up a lot of crud from the bottom of the tank.

I thought that the high pressure pump got fed by a low pressure supply pump and regulator and that excess fuel got recycled back to the tank on these old systems...maybe fuel sloshing around introduced some air in the primary loop?

 

Jim, if the system is Lucas and/or Tecalemit based (with slide throttles and a stepped mixture control against the cam at the back of the mechanical pump), you'll just need to get the car running and peer down into the intake trumpets to be sure each injector is firing and not dripping.

It's not uncommon to see air bubbles in the clear lines to the injectors and through the return bleed to the tank. Your system should have a fuel pressure gauge reading in the 80-100psi range (IIRC) and as long as you get that with the HP pump on, you should be ok. I have cracked the lines at the injectors to try and "purge" the lines, but that's a fire hazard!

I just finished with a Brabham/Lucas system on a Super Vee and that's what I've done with the Alfa V-12 F1 car in the past.

This is the system I'm talking about. http://www.lucasinjection.com/Notes%20on%20Installation%20and%20Calibration%20-%20Racing%20Cars%20LSK%20535.htm

 

As I recall this (it was 1981), I believe that they first replaced all the filters, including the big water separator. I believe that they then backwashed the main line back to the tank. What they did to the high-pressure injection pump I did not see. There was talk of replacing it because of grit, but it turned out OK. I think it was allowed to clear itself out by operating it.

But your car is a gasoline injection system without near that much pressure, right? Like for example, a 1972 mechanical injection Porsche? These had a sort of variable cam that mechanically adjusted fuel delivery by sliding to different spots on the cam profiles to send fuel at the right rate. I don't see how it could by itself get air into the system without also leaking all over the place...I just intuitavely thought about the tank feed situation and the primary fuel delivery pump as more likely air-introduction points.

 

Jim I have had that problem from a cracked pickup pipe in the tank. Fuel level gets below the area of the crack and sucks air. Winds up going to the pump as an emulsion and a mechanical system just wont work that way. The fact is if you have any air leakage from the immersed part of the pick up to the inlet of the pump that will happen and depending on how the car is plumbed it may not show up as a fuel leak.

 

Both good possibilities if the o-rings are fresh on the mechanical pump. Take a look at the Kinsler site. Lots of good information there. Good luck!

 

This is the logical choice if the system started full and didn't lose prime.

 

Your low pressure system should have a vent on it to purge out any air the low pressure system picks up. This should be a small orifice to let the air out but minimizing the amount of fuel returning to the tank. Any chance it is clogged and allowing the air to get into the high presssure pump?

Cheers Jim

 

Cavitation won't introduce air, that would be making something from nothing. Cavitation can produce bubbles of vapor, but those bubbles will collapse as soon as the pressure increases (often with impressive force), so those bubbles are not going to pumped anywhere, and they certainly won't be static in a fuel line.

 

Hope you're right: nice catch!

Still, this would also mean that the air-purge system is either not there or defective.

 

I second that!

 

One thing to remember is that a one way valve only works for a short period of time...probably not overnight. They need to have pressure against the high side - substantial pressure - in order to work at all. In near equilibrium, they are must about the same as nothing at all.

 

A higher residual fuel pressure will better help prevent the fuel from vaporizing if the lines are getting too warm, but you'll be limited in how much you can raise it by the opening pressure (~3.4 bar?) of the injectors (or whatever "release valve" is downstream of the ports on the metering unit) minus some reasonable margin. Better thermal insulation for wherever fuel is retained would help too if this is part of your problem.

Your mixture control unit looks particularly simple -- does it have any provision (like the "overfuel lever" on a road unit) to aid starting/restarting?

 

Can't agree that the intention on a TR's AC is to heat the fuel (as on a cold day you wouldn't even be using the AC). More the other way around -- to help the AC work a little better.

Although the water lines may be hotter than ambient, they are still cooler than a lot of other things so "protecting" them with the water lines isn't unreasonable.

But is your mixture control unit a "race only" arrangement -- i.e., designed around the assumptions that it wouldn't be restarted and that it would almost always be used for max power?

 

In that case, it never could get hot enough to make fuel temperature much of an issue, no?

 

I went through this once on a lightweight E-type. Turned out to be an air leak on the suction side. It was fixed by adding a low pressure pump and a FI tank. I would suggest a simple test (assuming you have access to the fuel pickup). Pull the fuel pickup out of the tank and connect a hand vacuum pump to it. Next seal the fuel discharge on the high pressure pump and the any venting on the FI tank and attempt to pull a vacuum on the system. If it does not hold a vacuum you have found your issue. Just isolate parts of the system until you narrow it down. If it does hold a vacuum then it is not an air leak into the suction side. If it passes the vacuum test then I would check the vent on the FI tank to see if it is fuctioning.

Cheers Jim