Injectors Rebuilt Twice....Still Leak...I'm stumped

[Quantum TD]

If you are torquing the injectors into the head with a 27mm deep socket, the torque value should have no effect on leaking (causing or eliminating).

Having said that, go by the Bentley spec when you install into the head (70nM or 52 ft-lbs), and be sure to use new heat shields each time. Anti-seize is also a good idea.

I simply suggested that you remove your injectors, put the heads into a vice, and torque the bases (bodies) to the head up to the value of 58-60 ft-lbs.  I too had lapped the crap out of injector bodies. I was certain they were good, but I kept getting small leaks, and some large leaks when torqued to 52 ft-lbs.

In the end, I followed Andrew's suggestions, and torqued them a bit more (to like 55-60 ft-lbs). I haven't had a leak yet. The only problem with this trick, is that you need to calibrate the injectors for this extra torque. Increasing the torque on the bodies will increase the breaking (pop) pressure of the injector. I don't think it's more than about 10 psi, but it could affect performance (MPG), as it will effectively retard you injection timing a touch.

[burn_your_money]

How does increasing the torque on the injectors have any appreciable effect on the opening pressures?

[Vincent Waldon]

I think the key word here is:

appreciable

Changes in breaking pressure due to body torque would have to be caused by the shim end getting closer to the nozzle end, thus compressing the spring.

If you factor in the thread pitch, the small amount of additional twist you're going to get by upping the torque due to the two metal mating surfaces, etc, I suspect that the breaking pressure is technically increased but my own personal opinion would be that the additional error this torque causes is well within the "you won't notice it and neither will your engine" space. 

Somewhere there's a table of what shim height does to breaking pressure... or I suppose someone with a super-sensitive pop tester could do an experiment some day to confirm.... my experience with my pop tester is that it's tricky to get them within 10 psi on a good day... against 2200 psi this is a very small percentage.

So.... everyone is correct IMHO... it's just a matter of if you chose to worry about it in the overall scheme of things.

[rabbitman]

How does increasing the torque on the injectors have any appreciable effect on the opening pressures?

I think he means increasing the torque on the injector halves.

Found this on the other forum:

(hagar) another little "INJECTION" --of good old DIY ---know how. -- setting breaking pressure by hand --no tester nothing.---will do in a pinch.

This post is about hagars experience ( Saga ) with Rabbit VW diesels.---did I adjust injectors by hand ?------- yes many.----I double tested later in PRO tester --it works and that is gospel.

Both long and short spring Injectors use same pitch thread on barrels. --compress spring by 0,05mm and pressure increases 5 bar. (my Bentley page 23) --one turn on barrel is about 60 thou US inch.

My Amerikan friends ---please go Metric --it pay's. SEE ? apple pie.

(Fatmobile) Posted: Sat Sep 16, 2006 10:29 pm    Post subject: metric   

--------------------------------------------------------------------------------
 
Looks like .05mm = .001968504 inches. About 2 thousandths.
5 bar per .001968504 inches
so if you wanted 135 bar:
135/5=27
27x.001968504=.053149608
So smashing the spring .053 inches will get you 135 bar.

Each full turn equals .06"
(.06/.001968504)x5= 152.3999951 bar
(.06" (one turn) divided by .001968504 (distance moved to equal 5 bar) = 30.47999902)x 5 bar.
So one turn equals 152 bar.
9/10ths of a turn gets you about 135 bar.

That took me awhile so select, Ctrl C. I'd hate to lose it all when I go to post it.

[Doakster]

If you are torquing the injectors into the head with a 27mm deep socket, the torque value should have no effect on leaking (causing or eliminating).

Having said that, go by the Bentley spec when you install into the head (70nM or 52 ft-lbs), and be sure to use new heat shields each time. Anti-seize is also a good idea.

I simply suggested that you remove your injectors, put the heads into a vice, and torque the bases (bodies) to the head up to the value of 58-60 ft-lbs.  I too had lapped the crap out of injector bodies. I was certain they were good, but I kept getting small leaks, and some large leaks when torqued to 52 ft-lbs.

In the end, I followed Andrew's suggestions, and torqued them a bit more (to like 55-60 ft-lbs). I haven't had a leak yet. The only problem with this trick, is that you need to calibrate the injectors for this extra torque. Increasing the torque on the bodies will increase the breaking (pop) pressure of the injector. I don't think it's more than about 10 psi, but it could affect performance (MPG), as it will effectively retard you injection timing a touch.

I've pretty much followed the bentley specs every time I've installed them. I going to give the lower torque spec that apparently bosch is coming up with (31ft-lbs) a try but I'm doing it more to humor my shop.

I would slap them in a vise to torque the bodies up a little bit, but I don't have a test stand to verify the pop pressure. I'll mention this to the shop.

[Rabbit TD]

Alright, what is the torque suppose to be when you install the complete injector into the head?? Just got off the phone with them and they are telling me bosch recommends 31 or 35 i believe.

I was going to 51ft-lbs off my bentley manual, but I could be an idiot and not read it right. I assume the 51 ft-lbs is when you are resembling the injector halves themselve.

What they are getting at is that I'm tweeking the injector too much.

31 or 35 is way too low.


I'm going by memory but the injectors as a whole going into the head is 55 ft-lbs. I've heard 51 a lot as well so that might be right. Possibly a variance in the bentleys.


If they are only torquing the injectors to 31-35 at the shop then that'll be more prone to leaks. Do it to 58 ft-lbs

At the shop when they are assembling the injectors they are torquing the halves together at around 50-55ftlbs, i don't remember the exact number they stated, but they went off the bosch spec.

When I install them in the car, on the head I am torqueing them to 51ft-lbs

I just looked at my bently manual again. The specs are:

Injector halves: 51ftlbs
Injector into heads: 51ftlbs

The shop is saying that they contacted bosch directly and bosch stated they are to be torqued to 31ftlbs into the head. They are thinking that the 51ftlbs may have tweeked the halves a little bit to cause the leaking, they also agree that retorqueing to 31f lbs may not fix it.

I was simply going off my bently manual, but I'm going to try loosening and retorquing to 31ftlbs and see what happens.

If it leaks at 51 it's probably gonna be a lot worse at 31.  I think I would forget about that place altogether if I were you.  A little extra torque isn't going  to hurt a thing, loosening sure isn't going to make it seal better.  If a head gasket leaks at 80 foot pounds would you retorque it to 40 because somebody just said so when you have the specs.?  My next step on that problem would be 90 in that instance.  If they still leak at that point then just read the thread about lapping injectors in the search section and do it yourself {1 at a time} Remember to use new discs under the injectors everytime you put theem back in.

[burn_your_money]

Both long and short spring Injectors use same pitch thread on barrels. --compress spring by 0,05mm and pressure increases 5 bar. (my Bentley page 23) --one turn on barrel is about 60 thou US inch.

The logic behind this is flawed though. In order for the injector to seal, all the finely machined/lapped pieces inside the injector need to be compressed against each other. The torque value is how tightly they are being held together. You are not going to compress that metal enough to significantly change the opening pressure. You will strip the injector bodies long before you do.

[Doakster]

Just pulled the injectors again, retorqued in the heads to 31ft-lbs just to see what it would do, big surprise they still leak.

I'm going to call the shop tomorrow and discuss torquing the bodies to 55-60ft-lbs, I'll probably just do it anyway, unfortunately I don't have a vise where i am right now, I'll have to get creative. 

Another question, I notice that the nozzles actually have a spot where they touched the copper crush washers, I assume this is normal, they naturally touch at the smaller diameter hole in the washer

[Rabbit TD]

Just pulled the injectors again, retorqued in the heads to 31ft-lbs just to see what it would do, big surprise they still leak.

I'm going to call the shop tomorrow and discuss torquing the bodies to 55-60ft-lbs, I'll probably just do it anyway, unfortunately I don't have a vise where i am right now, I'll have to get creative. 

Another question, I notice that the nozzles actually have a spot where they touched the copper crush washers, I assume this is normal, they naturally touch at the smaller diameter hole in the washer

Copper crush washers?  Is that what's under the injectors in the head or are you just assuming that's what they are made out of.  If there are copper crush washers in there like off an older style spark plug or something then somebody put the wrong things in there somewhere along the line.

[Doakster]

I meant the heat shields

[burnt_servo]

i don't know if this will help anyone , but on injectors that i've had trouble sealing after rebuilding , i've lapped  the matting surfaces to each other using a bit of valve lapping compound .

[Vincent Waldon]

i don't know if this will help anyone , but on injectors that i've had trouble sealing after rebuilding , i've lapped  the matting surfaces to each other using a bit of valve lapping compound .

Thats a great idea...

A flat plate of glass and 2000-grit sandpaper will work as well.

http://vincewaldon.com/index.php?option=com_content&task=view&id=20&Itemid=28

Of course, the *real* answer is a very expensive injector lapping plate... I believe it's in the 4000-6000 grit range, so by comparison valve lapping compound and 2000 grid sandpaper are savage butchery... but if it works it works!!

[MizpahPAH]

When I did my first two sets of nozzles I wasn't using a vice,

I used a big Cresent wrench and the 27mm socket lying on thier sides

They leaked and leaked more times then not.

tore them down relapped several times still leaked between the halves,

Then I looked at the bottom of one of the nozzles and it was off center.

Now I use a vice and no more problems.

Also before you tear them down check the tourqe.
maybe the shop gopher has been using it as a breaker bar  

[rabbitman]

Both long and short spring Injectors use same pitch thread on barrels. --compress spring by 0,05mm and pressure increases 5 bar. (my Bentley page 23) --one turn on barrel is about 60 thou US inch.

The logic behind this is flawed though. In order for the injector to seal, all the finely machined/lapped pieces inside the injector need to be compressed against each other. The torque value is how tightly they are being held together. You are not going to compress that metal enough to significantly change the opening pressure. You will strip the injector bodies long before you do.

That bit hagar wrote is talking about before the lapped parts are even touching each other, like you start screwing the halves together and when you feel the spring start to compress then supposedly you can count the turns and gauge how hard the spring is compressed. After the lapped parts firmly touch I would "guess" that torquing it won't change the pop pressure much at all. I'm not an expert though that's just how I understood hagar's method.

I don't think bosch should care how hard they get torqued into the head, the head will strip long before the injector does. And I really doubt it will change the pop pressure.

[Doakster]

got of the phone with the shop, they are waiting back from their bosch rep to see if anyone has a set lying around, I know it will be long shot since they don't make them anymore. Plus the second time they did the rebuild they torqued the bodies together at 55ft-lbs to try and stop them from leaking, and that didn't work. 

We'll see what they come up with. Anyone have a set of spare TD injectors they want to part with??