How to test injector return?

[BDaugherty]

OK. I've read TSB 14-003-06 and I've read all around the internet and I'm getting conflicting information. The TSB gives maximum specs for injector return flow, but it doesn't tell me how to perform the test. I know how to isolate the injectors and I have a cap for the rail. I know to perform the test for 1 minute, but I don't know at what rail pressure or engine speed. Would somebody please either copy/paste or direct me to the exact procedure.



I'm having a no start issue when the truck gets cold soaked below 50 degrees. I can crank until the batteries get weak, but it won't get over 800 psi in the fuel rail. If I disconnect the grid heaters and spray a little ether into the filter housing, it will make it spin a little faster and when it hits 1100-1200 psi, it fires and runs smooth with no smoke.



So far I have checked the following:



CP3 flowed 200mL during 30 seconds of cranking against minimum spec of 70.



Injectors are returning 50mL (all together) with the engine at idle for 1 minute. Rail psi is around 4700psi at this time.



The relief valve on the fuel rail is not leaking at idle speed. I have not tested it under other conditions.



The connector tubes moved a little when I torqued them, but I'm not sure if it was enough to cause problems. I parked outside and I will try to start it in the morning and see if there is any difference.



I also read that it was possible that the fan clutch could drag or be shorted and I suppose that could reduce cranking speed enough to have an effect on fuel flow to the rail.



I'm open to suggestions in addition to specifics on how to correctly test injector return.

 

[sag2]

I thought I posted a photo of the 9012 in the back of the filter housing and measuring the injector return off the return hose to the tank. But here goes again. If you have isolated the pump return from the head return and you only have 50ml in one minute your injector return is fine. But your symptoms would indicate otherwise.



The injector return is NOT measured off the 9012, it is measured off the return line at the bottom of the filter housing.

 

[BDaugherty]

I do not have the 9012 fitting so I had to do it the harder way. I blocked off the rail pressure relief valve with a solid bolt. Then I took the short section of hose off of the cp3 return. I capped the side coming from the filter housing and ran a hose from the cp3 return back to tank. I took the factory return loose from the back of the filter housing and ran it to my measuring container.

My main concern was whether I did the right thing by running the truck for 1 minute to clear out any air in the lines and then measuring the return flow for 1 minute. I did it twice and only had 48-50 mL each time. I did measure the CP3 return. It was right about 700mL per minute.

I was doing it wrong at first and that messed my "scientific process" up. I torqued all the connector tubes before I realized that I had crossed my hoses so I don't know how much of an improvement that may have made. The connector tubes did tighten down a decent amount, but I don't know just how loose they would have had to have been to have an effect.

 

[sag2]

If you are still set up to test, crack one or two of the easy tubes and you will be surprised how much difference it makes. Also does not hurt to run the test with the FCA unplugged. That was the method we used in 2003 before the scan tool was modified to raise the rail pressure. There is also quite a difference between 5k and 20k pressure when running the test. 20k tends to show the leak much more than at 5k.

Also run the test at operating temp. It will make a difference in the return also over a cold engine.

 

[BDaugherty]

If you are still set up to test, crack one or two of the easy tubes and you will be surprised how much difference it makes. Also does not hurt to run the test with the FCA unplugged. That was the method we used in 2003 before the scan tool was modified to raise the rail pressure. There is also quite a difference between 5k and 20k pressure when running the test. 20k tends to show the leak much more than at 5k.

Also run the test at operating temp. It will make a difference in the return also over a cold engine.

So you are saying I should retest a warm engine with the FCA unplugged, but use the same guidelines for flow specs.



Could you explain what you mean by cracking a couple injector lines?

 

[sag2]

Yes the test is supposed to be run at high pressure, same specs. Back off a connector tube a 1/4 turn and measure the return again, you will be surprised at how little the connector tube has to back off torque to have it not start.

 

[BDaugherty]

OK. I ran the test again and it came back kind of odd. I tested for 1 minute with the FCA unplugged. During that time, my gauge was reading 26,800 psi in the fuel rail. The relief valve did not leak at that pressure.



All 6 injectors returned 725mL. 700 with #1 capped. 695 with #2 capped. 690 with #3 capped. 690 with #4 capped. 690 with #5 capped. 680 with #6 capped. That gives me 25, 30, 35, 35, 35, and 45.



I'm certain that I had the return from the CP3 run to the tank and the rail relief was capped with a solid bolt. What really has me scratching my head is that the return numbers don't add up to the total. I'm sure my individual readings could be off a little, but there's no way they could be off more that 10mL or so for each reading. What am I missing? Also, I did torque each connecting tube before testing. They were all still tight from the last time I messed with it.



I guess I see a complete set of injectors in the immediate future for me.

 

[sag2]

I'm not sure I understand your numbers. If you really have the pump return isolated from the head return, you should still have less than 180ml for all six injectors in one minute. If you have more then that you have an injector problem if the tubes are in good condition and torqued properly.

You can always send them out to Dynomite Diesel and have them tested. Far less expensive than just buying six.

 

[BDaugherty]

I made plans to drive an hour down to Garmon's Diesel and have them tested there. Jeff said he could run them on the test bench while I waited and then I can make an informed decision on which ones to replace.



Since I torqued the connector tubes, it has been much better... but I haven't had any cold weather to test it with. It does seem to have a lot of injector rattle. I don't really know how much is normal or if it even hurts anything. Thanks for all the previous replies, help, and suggestions.

 

[CKelley1]

what you have is typical wear in the 03 mainly due to inadequate filtering. all 6 injectors will need to be replaced by now you should have the transfer pump in the tank. here is pix of several filters, the ugly ones are mine

#ad


#ad


#ad


#ad


#ad


#ad

 

[LNaugle]

In laymans terms... when everything is isolated properly... . injector return should not fill a normal shot glass in 1 minute. New injectors return just enough to have to lick the fuel out of the bottom of the shot glass! lol!

 

[BDaugherty]

I had my injectors tested and installed new stock nozzles onto my bodies. The cold start is MUCH improved. I can now start cold soaked at 30 degrees (coldest night we've had). I now have another possible injector related issue, but I will start another post. I just wanted this one to show that my cold start problem was for sure injector related.