Did some injector testing/cleaning today

The Bronze

New member
I purchased a new injector bench tester/cleaner last month and besides running a couple of sets of injectors through it to make sure it was working, I hadn't tested it's capabilities fully. Another forum member asked if I could clean and test his injectors, so I had him send them up this way. I learned a couple of valuable lessons today, some of which may help remedy a problem this member was having, others reinforced that you never trust anything until proven otherwise.

The injectors were pretty clean on the outside (pintle area), so I checked the resistance (all 12.3 ohms) and then threw them on the flow bench to see how they balanced out. The flow time, PW, and presssure I didn't document, but it was probably about 15 seconds at 40 psi. I was just looking to see if they were all functioning and to see if the spray pattern was okay. You can see that #2 is a bit higher and #5 is a bit lower.




I then ran them repeatedly through a auto run cycle to make sure they didn't fail after being operated for some time. I had other things to do, so they ran through about 4 or 5 cycles before putting them in the ultrasonic tank.



After cleaning them twice, I reinstalled them on the flow bench. My goal was to do a few things:

1) Determine actual flow at various operating pressures. Plus I wanted to see when the injector simply couldn't compete with the rising pressure above the pintle and the flow begin to drop.

2) Determine minimum injector pulse width (IPW) for injector operation. Many injectors apparently stop working when pulse widths drop 1-1.2ms. I wanted to see where 60# injectors stopped. Looking at various datamaster logs, I see 1.2-1.4 IPW show up, which are lower than or close to the "minimum + 15%" that should be the bottom of the range. That would allow the tuner from dropping it near those values, and prevent misfires at or near idle. I also wanted to test this at various pressures as well.

3) Do a leak check of the injectors. Pressure on, cycle the injectors and turn them off. Make sure they are not dripping.

4) Run the injectors through a few more automated cycles where the IPW varies to make sure there wasn't any hot failures.

The third item was easy enough to accomplish. The injectors shut off and no fluid dripped or even built up on the tip. This picture looks like the injectors are streaming, but that is just the reflection off the back of the cylinder.



The next test I did was to lower the injector PW (at 40psi fuel pressure) and see at what IPW the injectors stopped spraying. I found that #5 injector stopped at 1.6ms, while the others sprayed down to 1.5ms. That was higher than I expected, so I played with the pressure a bit and then pulled out the meter. The meter showed that when I commanded 1.7ms of IPW, the unit actually provided 1.6ms. When I commanded 1.6ms, the unit provided 1.2ms. Still, one injector is needing more saturation time to build enough field to lift the pintle and start the flow of fuel. You can see #5 shuts off while the others are still operating.



Since the unit was unable to control the IPW correctly, I went another step further and grabbed my scope to see what the actual command was. Now these tests are halfway worthess...



This is where I see that this fine device from China provides 12V to the injectors, but drops to about 8.5V when the injector is on. IPW matches my meter, which shows that at low IPW, the bench doesn't have very good control, but works okay at higher IPW commands.

So now I have an injector that is failing at low voltages, but would it work at normal operating voltages? How am I suppose to know with this piece of junk? :rant:

With that question unable to be answered (but the injector is suspect), I moved along to the injector pattern verification test and flow test. This portion worked out okay at normal pressures and IPW, but the fuel supply system could not handle 100% duty cycle with these larger injectors. I decided to make the math easy and did 3000rpms at 20ms for 15 seconds. 6000rpms=20ms between firing events, so 3000 would be 50% (roughly, since this machine is only reasonably accurate). The cylinders were getting pretty full at the 2x second mark, so I dropped it to 15 seconds and let them run (at 45psi).



These injectors flowed 650-660cc/min. That's pretty good, considering current available was low due to the low battery voltage. I was not able to turn up the fuel pressure and witness when these injectors stopped flowing while at high IPW, It wasn't really necessary at our idle pressures, so I skipped it for now.

So 4 hours today = Chinese injector tester "LAUNCH" brand is okay for general cleaning and flow verification, but not ideal for performance testing of aftermarket-sized injectors. I will be talking with my tool distributor on getting it replaced with a more robust unit.

Also, it looks like a forum member should be purchasing one injector, which should help smooth out his idle and help allow him to use BLM for tuning purposes, but will not cure his heavy black smoke issue.
 
Last edited:

Don W.

Stab it and steer it
Re: Did some injector testing/cleaning today

Facinating. I've seen various do it yourself "cleaning" sites on the internet. Combined with your Chinese machine you could go into the discount injector cleaning/flow checking/matching business. :rotf: Really it's nice to see this stuff. Great info. :tup:
 

fivetodrive

CRISPY
Re: Did some injector testing/cleaning today

Did you ever get a more robust machine? I have way too many injectors and was considering buying this machine for a budget tester and flow machine for junkyard injectors. 36lb and below...
 

JSM

Active member
Re: Did some injector testing/cleaning today

Old thread but very interesting.

Something to know about, in stock form the ECM has what is called quasi fueling strategy. The idea is at idle instead of a small IPW, fire the injectors twice as long, but only every other time to avoid the small IPW issues.

Theory works better than in normal operation though, and I generally turn it off as they seem to run better without probably due to more variable that need tuned to work properly.
 

Slyclone

Well-known member
Re: Did some injector testing/cleaning today

These were actually my injectors. I sent the 60# out to be checked because I had thought there was issues. Like I thought there was.

I sent out a set of stock injectors recently, to be cleaned and flowed. I told Kelly that it would be cool to see another testing. I planned on using the stock ones in my new project typhoon.

I got excited when I saw the post and thought he had some new material for us to read up. I haven't seen anything yet..
 
Top