Disturbing findings flowing and testing injectors
 

As many of you have seen I have been posting about dynoing a slightly modified 502 mpi in my shop. After seeing some erratic afrs on dyno I replaced #8 injector with another used one. The erratic afr readings on that cylinder went away. This disturbed me as I am not monitoring 6 of the 8 cylinders so IF one of the 2 I'm monitoring was going lean then coming back to normal WHATS going on in the ones I cant see?? So I made the decision to buy a injector flowing, cleaning testing machine with a ultrasonic cleaner also. As soon as I got it I promptly flowed the 13 MPI 502 injectors I have, saw a 22% difference from the best to worse, saw several that at higher rpms just dont come close to the average. Also flowed a set of 500efi injectors, they too had a good 15 % difference and one was stuck open. Flowed a set of siemans deka 80 lb ones I had from blower motor that only had 5 hours or so of total run time on dyno that were never in the boat , a good 8-10% difference from best to worst on 7 of them with one that outflows the rest of them by a good 20%. Bought a set of "replacement" 502 mpi injectors from a seller on ebay that claimed matched and flow tested within one percent, TOLD seller I was going to be testing them, he assured me they were good. They were the WORST by far with a good 40% difference from best to worst PLUS they started shutting down at 40 psi and by 3 bar/ 43.5 ALL shut down completely!!
After cleaning, reverse flowing then flowing 20+ times I was able to find 8 of the mpi stock injectors that were plus or minus 6% of average (published data on racetronix website shows the majority of NON matched injectors you buy are either 4% or 6% +/_). Sent e-bay injectors BACK-TOTAL JUNK. After flowing, cleaning the siemans dekas they were closer to 4% +/- except for the over achiever that still flows 10+% more than the rest, quite disturbed that they got so bad from 5 hours of run time and sitting around about as long as what a boat normally sets all winter. Think about that, IF you were tuning a motor using a afr meter in each collector you could have 1 cylindermaking you think you are too rich when you really have one thats too rich and 3others that are lean. The 500efi injectors came in fairly good after cleaning except one weak one that flows consistently low. On the advice of Racetronix I am no longer going to flow any used injectors until after they are cleaned as to not foul testing solution in machine.
Also bought a set of matched 63 lb siemans dekas that tested within 2% =/- on my machine (its a entry level machine, not a 10,000$ one by any means but not a cheap Chinese import) so at least gives me a good picture of whats going on.
A fellow osoer sent me 16 injectors from a 502 mpi that ran fine but suffered reversion, AFTER cleaning and flow testing/reverse flowing the ones I have done so far are within +/- 7%, originally one was stuck on until ran thru ultrasonic for 4 cleaning cycles, they too started out quite out of range. I am disturbed that they are so bad in general, there is no wonder mercruiser tunes their motors so pig rich, if they tuned them to the edge and a injector flowed 10-15% LOW no wonder there are so many horror stories on here of burned up motors!!
What my feelings now are is I would NEVER buy a set of injectors that weren't match flowed to start with, still flow them to see for sure they do match ("matched/flowed injectors" is a claim made by many injector suppliers) and they need to be cleaned and tested anytime a motor is being rebuilt or apart, if your tuning afrs on a injected motor its prob best to go a little richer than what you think would be good for a safety margin down the road, maybe their is some merit to filling a water separator up with injector cleaner once a season or so, and all of these 502 mpi injectors for sale on ebay are cheap Chinese knockoffs, coils so weak they cant open at 43.5 psi, WTF???. In past I had sent my blower motor injectors to Crockett everytime I had it apart, I would now consider that 100% mandatory and fool hardy NOT to! Thought I'd share, Smitty
the picture is of the supposed matched set tested at 40 psi that I bought on ebay!

After seeing afrs all over the place I also invested in a daytona 8-pack o2 sensor setup for my dyno so I could see whats really going on in all the cylinders!

Just bought 16 new injectors for these 502's to send to OSO member. OEM from Italy. His flow testing was bad as well. I suspect about 6 of them were ebay fake injectors as they all flow different as well. None of those flowed the same at all. Again DO NOT BUY INJECTORS from the net. You are asking for problems.

Even new injectors need to be tested. Simple tests first are for OHM's and you will see a difference in those as well. (New injectors) Test running Impedance and that will be different as well between new injectors. Funny Actually sad how they are not even meeting OEM spec differences between new injectors.

For a injector test machine - It is best to have a separate back flush system set up away from the path of the machine testing fluid. I do this always first before any flow testing of the injectors. The back flush really can help even new injectors plus nasty used injectors but not all the time. Keep in mind injectors are electrically and mechanically devices - with that said the electric portion of the injectors plays a huge role as well and no way a person can fix that other than sending it off to get remaned by professionals.

How are the bosch injectors out of the box?

Generally speaking, for the sake of preservation, a TBI system might be more reliable than a TPI system with individual injectors?

Well, as bad as i have seen these injectors varying you would think there would be efi motors washing rings and melting down all over the place BUT other than a occasional post about a burnt piston here and there and motors with blowby from washed pistons surprisingly i havent seen alot of threads about efi motors going down so i wouldnt say tbi injectors would be a better solution. Alot of boats with black transoms though

Talking to rcinjection tech they tell me the reason you dont see it in cars is they dont sit like boats do but after what im seeing im definately NOT going to tune motors no more without leaving a little extra fuel in the tunes to account for injector degradation.

When bought in shelf stock, non matched sets Bosch says this:
Static Flow Tolerance +/-%: 5%
BUT I haven't tested any new ones yet, you could see 10% from best to worst new if what they say is true

Hmmm, would we be better off going back to carb setups? :evilb:

A lot of us would probably toss and turn all night long if we truly knew what was really going on inside our engines. Nonetheless, excellent technical investigative work.

Once again Arctic, just GREAT and useful information. Thanks for sharing!

Depending on how and where you boat, the marine environment can be one of the best applications for a decently set up carb. For most of us, we boat at the same elevation and general temp range, this in itself negates two of the bigger drawbacks to carbs. For a sanely built marine, n/a app., I think carbs still make a lot of sense if you get them set up right to begin with.

I used to do injector development for Bosch. We garunteed to OEMs +- 3%. Densos with machined main bodies were the best at +-2%. The dynamic measurements were at a 2.5ms on time over a 10ms period.

Static flow rates are determined mostly by the orifice plate. Which could potentially erode over time. But boats get so few hours relative to cars.

The biggest issue I could see is that needle is harder than the seat and it usually wears a nice hard sealing surface for itself over time. This could also make the needle harder to lift off the seat and it would show up as higher variation in the dynamic flow readings.

I'm not surprised the Chinese ones are junk. There's so many places to go wrong in the manufacturing. Commenter above is correct, the strength of the magnetic circuit is important.

At Bosch our injectors were matched for flow at the dynamic rate during manufacturing by measuring flow as the needle spring force is set. We also experienced variation from machine to machine, making measuring anything less than +-1% impossible. We always tested statistically significant samples of 30, and used statistical analysis.

I'm surprised by how bad your findings are. I would suggest trying to control for the temperatures of your electrical components like drivers etc.

I'd suggest devising a test to determine your machines repeatability. Unfortunately this can be impossible to discern from a single parts test to test variability! At certain speeds I could watch an injector needle just spin around and around!

We test fuel systems mostly with viscor. It's essentially diesel fuel with an additive package to reduce flammability. I don't think it should be difficult to get.

I know some of OEM's say 7 % or less (some say 4 % or less as well in regards being in spec but alot of used injectors seem to be 10 % or even more off when flow testing. Also alot of times its not wise to mix new injectors with used injectors even from the exact same OEM. It might surprise you how much different the volume of flow comparing new vs old injectors can be.

Bosch has alot of equipment to test injectors plus all the R &D to do so - way more than anyone of our budgets can even think about.

Post #13 - I like the first sentence and totally agree.

With that said I sincerely hope the OP gets everything figured out. I have always been old school and like a carb/carbs on top of engines. It wasn't until recently I found myself researching a lot about good efi systems. Actually considering for 565 builds in the works. They have come a long ways over the years. The only thing that prevents me from doing so is when I read threads like this. I realize there are several good systems on the market working flawlessly however something is to be said about old school simplicity.

If you are worried about injectors have them checked out or buy from this company - one of the top rated injectors in the world - they teamed up with Bosch. Look at how they test injectors and so forth. I have sat in on anything they have for seminars in the past years. If you think you know alot about fuel injectors and fuel injection this place knows more about it than most ever could in a lifetime.

http://injectordynamics.com/company-overview/

That's pretty cool.

These other wanna be, 6 hand, know nothing purchasing from China toy factories ought to pay the ultimate price. When you have safety nets outside your factory cause so many are being worked to death 24/7 for food and housing I can imagine Quality is sacrificed. That's terrible what many are going through.

Thats the thing, i can tag the injectors that flow high or low, move them all over the machine, test 2 days later, a week later, they still flow low or high, pretty consistently. I might see a over achiever flow 10% high then maybe 1% one way or other, thats why i say in beginning my machine might vary 1% either way but if im seeing 10% thats big.

I recently became a dealer for racetronix who sells match flowed standard, bosch, siemans deka and a few other brands of injectors. The set of 63 lb i just bought matched within 2% either way on my machine , on different days, in different positions , was by far the best ones i tested.

THanks Bup.

Found this vid on them: https://www.youtube.com/watch?v=_UZ6wYTjUpo

I also added a pressure gauge at end of rail to ensure regulated pressure shown at console matched at furthest point from distribution, it did. I have switched the injectors thruout diff positions to see if they changed, also measured the tubes vs markings, tube 4 came up 1 ml short of all the others in a repeated measuring test injecting a 60 ml sample then draining, repeating 4times, at 120ml, was off the same 1 ml vs the others, i add that to all the readings. Yes, I'm measuring reading at the crown of meniscus, worked in boiler chem testing lab and o2n2plant lab before.

Steve alot guys are adding their own fuel pressure regulator in the mix as well.

I have posted this before I follow and have learned alot from Paul Yaw. I do not know him personally but sit in whatever he has at PRI and elsewhere. Last year they did not have a seminar. They do everything first class and wait to see their new electric fuel pump that is coming to market. I had the demo on it and a chance to run it as well in their test tank.

The last seminar I sat in was with Bosch and Mr. Yaw for 3 hours and for free. I will post a vid from him but you have to listen very closely what he says about fuel injectors and testing them. He is so right about the points he makes. Extremely very smart about fuel injectors / injection.

When testing them did you run them for a long time like maybe 20min? Just wondering the way our boat motors are loaded I wonder what happens to the flow as they beat up.


Btw

I just had a 500efi injector stick full on. The new one will be flow tested before it goes in.

A good vid for info.

Listen to it very closely - spray patterns, testing machines - old tech and 80's injectors. Spray patterns are the most important thing from a correctly working flowing injector(s).

https://www.youtube.com/watch?v=xFOTtSrFugM

I test run fuel injectors for 30 minutes straight and the GM Multecs are one of the worst. The longer the injectors run - heat can be be produced. The fuel injector Internal coil(s) can even short out causing very poor and or NO spray pattern at all. You can even test the injector(s) to see how well it flow volumes the first minute out of the gate compared to operating it for 30 mins. Sometimes the volume is way different.

EFI sounds life cool stuff...especially for someone that likes messing with hi-tech stuff compared to a carb app. For myself, I am glad that I have decided to go with a Holley carb---especially their Ultra HP Series with the black anodized anti-corrosion coating and some 30+ upgrades.

For guys like Articfriends/Smitty, EFI is something he loves doing and is good at. I just don't have the time to be messin' with all that kind of stuff.

I dont have temp controlled testing fluid in my machine so I havent ran injectors for more than 10 minutes flowing them BUT I have ran them thru (4) 15 minutes cleaning cycles triggering the coils in the ultasonic cleaner till they were good and hot, blew them out then got them on machine while still good and warm, havent seen the flow degrade on any so far from increased temperatures BUT havent tested any GM multecs yet as BUP mentioned.

There is another great video from Paul I'll try to find and post. They run all of their injectors for 30 minutes at 200 hz (24,000rpm) just to break them in before they do any flow testing on them.

WOW, they are pricey, I'm sure you get what you pay for though! After speaking with the owner of Racetronix I was impressed with how they buy each part number of injectors 240 at a time then match/flow them for same price as they are sold everywhere else so I bit the bullet and did a dealer buy in as I want the injectors I buy to start out as close as possible. was impressed with the first set I got from them last week, a second set of (6) 80lb siemans dekas just showed up today that are going in a 600 hp turbocharged 3800 in a 2004 Grand Prix GTP, will post findings once I flow them too, Smitty

Smitty,

If you would like to add another parameter to your test, get an LCR meter to measure the characteristics of the coil.
You may be able to correlate the results.

Jeff

Did you ask how they flow test them? A certain duty cycle?

They said they had a custom machine built thats labratory grade, temp controlled, tested at idle , mid than high rpm. I do my test at 750rpm 2ms, 2400 rpm 6 ms, 3000 rpm at 8 ms, 4500 at 9 ms then 5500 at 8 ms, this pretty much mirrors the pw time on my tune from 502 i have on my dyno

This is the video I was trying to find.

Pretty slick where you can call up with a serial number of an injector and they can send you a matched replacement.

I'm curious how these would flow on your machine.

I'd sure as heck would not want one of my engines blowing up because of a crappy injector! Thanks for your help Smitty.

I flowed and tested the (6)80 pound Sieman's Deka injectors today that came from racetronix as a match flowed set that are going in our 600-horsepower 3800 turbo project and they flowed plus or minus 1% even at high RPMs, impressed!!

I looked up the ID injectors. They are Bosch EV14s like I used to work on. I think improvements in controlling the tolerances of the parts that make up the magnetic circuit would have helped those. Like I said, the densos with turned main bodies were the cats meow. The EV14 had a deep drawn body with no other machining processes. The coil and coil housing tolerances could get out of wack and still be to print because some moron with an Indian name wasn't smart enough to do a stack up with a radius tolerance and an angle tolerance, I figured it out years after the release but no one cared. I think that could have affected the mag circuit strength and repeatability.

Flowed a set of stock injectors Saturday from a 220,000 mile 6.2 chevy LS that has sat for a year (customers motor thats going in a jeep), they flowed within 5% of each other with just 15 minutes of cleaning in ultrasound, flowed a set of injectors today from a 90,000 mile 2000 Camaro LS 5.7 that sat all winter that we are installing cam and intake on, they also flowed within 5% with 1 cleaning cycle in ultrasound, it seems these stock oem injectors flow MUCH better than any Mercruiser injector I have tested to date!