Injector knock, is it something I’m doing??
 

I have a 1991 300D and 350SD.

The 300D has injector knock on injector/cylinder #1.

The injectors were refurbished. No change.

I swapped an injector from my quiet and smooth 350SD, and the 300D worked fine. So then I bought a “good used” injector. It knocked.

So then I took a different injector from my 350SD, installed it, and it too knocked. It doesn’t knock on my 350SD, but it is knocking in the 300D.

I always torque to 70nm. I bleed air. It’s plain to see when I crack the injector feed line that the knocking stops.

First I thought it was bad injectors. There could be two bad ones.

But now I have to wonder if somehow it’s me. It’s not like I haven’t swapped a lot of 60x and 61x injectors before. I have.

But could there be something I’m doing, or something with the hard line or return lines, that maybe I’m missing?

I do notice in the clear viton that there is some air in the return line. But it was also there with the quiet 603 injector I swapped….

Thanks for any guidance.

Given that you've tried knockless injectors and it still knocks, it's more than likely a poorly seated crush washer on the delivery valve for that cylinder, or a worn out delivery valve that no longer seals properly on that cylinder. Air will be present in the return line until the car is driven long enough to burp it out. The return lines are "leak lines" and very little fuel flows through them, it takes a while to get the air out when you've had them off.

It's also worth noting that brand new or refurbished injectors can knock or nail for up to 1000 miles after installation as they break in.

One knock less injector from my 350SD was perfect.

https://youtu.be/KNyLq5EADHc?si=LKmkbpE-WaFq0KYM

The “good used” and another injector subsequently knocked.

The one that didn’t knock was immediately silent. Even with the air still in the return line as can be seen in the video.

The refurbished ones were in for thousands of miles. There is no settling in or silencing.

Certainly a DV or something could be involved, but the noise is at the injector, not the IP. I can tell that definitely.

That’s why I’m kind of curious if it’s something else I’m doing when I install…

The DV's are there to do more than just act as a check valve. Their design is also supposed to displace some fluid in the delivery pipe to create a sharp and abrupt end to injection when they close.

If the DV leaks, has a worn seat, or the copper crush washer is weeping by, it will cause a timing problem in that cylinder (late injection) and can cause a dribble at the end of the injection event, causing a knock or nailing. The noise is always at the injector end, not at the IP itself.

Having been through DV issues with my first 300SDL, I'd be looking at at least redoing the crush washers and O-rings to see if it helps before diving deeper. Otherwise, the noise is more or less harmless, just annoying.

My '85 524TD has very loud injector knock when cold on Cylinder 4 with the lift sensor injector. Nothing I have done will silence it, it's just the nature of the beast with this particular engine. Once it warms up (maybe a minute or two of running), it's as quiet as the rest. Whatever, car runs fine and does 31+mpg when scooting 85+ on the highway, so I'm not losing sleep over it.

Thanks. That’s helpful. And with 23x k miles on the car, the DVs, and an overall IP removal and re-seal is probably prudent. I’ve done all the seals on my 603, so I know how to do the mildly annoying job.

But here’s why I’m still doubting myself. I went and swapped a few injectors back in. I took the injector I used last time when it was smooth and silent, and I got a knock. Then I swapped a few others and had varying levels of success. Amongst them, I then put that same injector that was quiet before, back in, and it was quiet again.

Air, I get… but the car starts perfect and smooth. And I bleed the lines for a bit before revving the engine.

So while I can see that DVs could be a culprit, what I don’t get is why/how the noise differs with different injectors, or the same injector installed multiple times. I guess maybe different injectors could have slightly different pressures that interact with a faulty DV in different ways… but it still seems strange…

Anyway, worth planning to do a DV job regardless, I guess, though it’s a balance between aint broke don’t fix it, and good pm practice, and I’m not sure which it is just yet.

But I’d like to be sound with knowing that I’ve exhausted all other options. When I get eight injectors back from Greazzer, then I guess I can make that call 100%.

Thanks as always for your insight. This board has really slowed down in the last few years, and your posts, and your history of posts, really is some of the best stuff on here…

A generous dose of a cetane improver may give some indication of whether there is a mechanical problem with the injection components, or a fuel problem.

Ok I’m still at my wits end over this.

Had greazzer look at the original injector, and rebuild an additional one with a matching nozzle.

Installed them both, got noise. I think the replacement was marginally better so I left that.

Kept investigating. I’m thinking more and more that #4 is also contributing to the noise.

Swapped #4, no dice. Swapped another newly refurbished injector from one of my 693 engines. More harshly metallic knocking sound!

There isn’t enough fuel return to move the air out of the lines. I’m not sure if this is normal or not.

I can definitely hear the knock stop on #4 when I loosen the fuel line. Not sure I noticed that when troubleshooting before. I do think #4 knocks more since I’ve started to mess with it.

So I’m at a loss for what it is. If somehow I’m not purging enough air?

Mercedes had lots of problems with knock, in so much that in 84-85 they moved the
injector 1mm closer to the dispersion ball. Too large of fuel droplets was thought too be
the cause. I would agree with this idea.
When a injector mists but sprays enough the timing will be late and it will be much quieter.
When cleaning the delivery valves I have notice debris around the valve seat holder, I do
feel this does cause problems with seat sealing .

Do check fuel pressure, low pressure and lower fuel level might possibly allow any air
bubbles to move to one or more cylinders at idle.

One thing I’ve noticed is that the fuel in the return lines never fills them, there’s always some empty space. Not sure that means there is air in the system, but it is noticable.

I was looking at some pics of the 1991 Mercedes 300D fuel injection pump. After the 3 flange nuts are loosened you can turn the adjustment to change your fuel injection pump timing.

If there is not some sort line already scribed from the block to the fuel injection pump flange (so you can return it back to where it was) you can make one. Simply try advancing your fuel injection pump timing a tiny bit and snug down the flange nut bolts and see what happens.

Fuel quality can also case nailing. I got an amount of bad fuel many years back that did that.

It is on my to do list to verify timing. Thing is it could be so many things. What I know is that one time I put in a used injector and it was quiet. But yet when I did it again, it knocked. This is what spurred the question of “is it me?

It has also been thousands of miles with the knock since I owned it. PO said it was a lifter. On certain occasions, or maybe almost always, it has been next to impossible to trace to a specific injector. This why I thought it was lifters. Only after putting in another remanufactured injector have I really found the ability for knocking to effectively be on or off on #4. But with a different, quiet injector. Which again points to something in doing or something else.

IP seems quiet with stethoscope.

And to follow up, here’s a video of what I’m seeing on injector #4. This is the second “quiet” injector from my 603, which were just refurbished.

When I first put the injector in, it was almost a metallic sound vs just a knock.

But I can crack the injector and hear a noticeable difference. This is consistent across at least four injectors I’ve tried:

https://youtu.be/IGO0fZeGzlQ?si=3lFjWiLK6H7q5PcX

The six 603 injectors I’ll hopefully get installed today. And re-validate their “quietness”.

Unless you have serious chain stretch from extreme lack of maintenance, IP timing and base engine timing are the least likely to be your issue here. 60x engines just don't eat chains like the 61x engines do and there's no offset keys to adjust cam timing either. I've serviced 603's with 300-400K on them and the original chain with 1-2˚ of stretch and left them well enough alone.

I would have hoped that my previous reply about the delivery valve leaking would have been heeded, but nobody ever listens to someone with experience, so why am I not surprised? There's a reason I rarely contribute to this forum anymore. The injector isn't the issue, it's delayed or improper combustion. I've dealt with many 60x engines with this noise and DV's generally improve or eliminate the issue. Newly rebuilt injectors shimmed to the correct pop pressure always seem to exacerbate a leaking DV compared to an older worn/weak injector. The older engines with the "vertical" injection prechambers are far noisier and always have a certain degree of a metallic "cackle" at idle. Some degree of knock is to be expected and has to be tolerated.

In this case, you have a combustion problem, likely a leaky DV or worn DV seat. Air in the return lines is normal. You don't normally see it with the black/grey rubber ones and it takes a LONG time for it to work its way out. The leakage through the return/leak ports in a well-lapped injector is very small, so the flow through those lines is minuscule.

Actually, I do need your advice. But with three kids and a full time job, I can only mess with this so often. I do appreciate your current and legacy posts, BTW.

I just received on Saturday, the spline socket and a set of DV washers and o-rings. So that is going to be another thing I work through. It’s excessively hot here so starting on that, amidst other stuff, just wasn’t going to happen. But I met up with greazzer in Delaware on Friday when he was passing through, so got some injectors, and was just excited that maybe there could be a difference. Remember that one time I put another injector in and it seemed quiet…

The TC wear is less than half a degree.

Anyway, thanks for your help, please don’t take anything as if you’re being ignored, I for one sure appreciate it!!

Replying again to let you know that I have taken your advice and did the DVs today.

Removing the intake is the biggest hassle. Not hard, just a hassle. DVs were easy. A mess, but easy. I used compressed air and cleaned as best as I could.

https://i.imgur.com/NZmlA0kl.jpeg

O-rings were hard as would expect.

https://i.imgur.com/ZTbUXksl.jpeg

And the old crush washers had a slight wear ring

https://i.imgur.com/wRUIOCZl.jpeg

https://i.imgur.com/bkdC2xUl.jpeg
https://i.imgur.com/9t2F67gl.jpeg

https://i.imgur.com/RmsW02Ml.jpeg

Torqued them as per the instructions.

https://i.imgur.com/ulkT4JGl.jpeg
https://i.imgur.com/J5bMV7zl.jpeg

https://i.imgur.com/SSlrMfQl.jpeg

It took some cranking to get the air out, but it worked out ok and fired up. It still has a knocking sound, and what I’m seeing is a lot of more frothy bubbles in the first cylinder return hoses. I think I do need to go back to the regular rubber ones to seal a bit better.

Anyway, was going to take it out until I saw that my turbo brace/stay was cracked and fell off. So now how to repair that…

https://i.imgur.com/BNCBvUKl.jpeg

I’ve put about 150 miles on it since I did the DVs.

My take is that the injectors are quieter, I can’t tell that there is as much knock.

I still get a rhythmic metallic tapping at a slower frequency. Maybe that’s the lift pump or something? I notice it’s most pronounced at hot idle and when driving in high gear at low rpm, 1100-1250 call it, where I hear it most pronounced.

Vacuum pump?

Not according to my mechanics stethoscope, but who knows at this point. It is the updated style, but I guess that may not mean much…

I drove it again today from cold. So the first cold and low speed drive since I had redone the DVs. In urban locations where there are structures on both sides to hear sound bouncing off, it sounds excactly the same as before. At idle it’s arguably quieter. But from just off idle to about 1500 it sounds like rocks in a box rattling around. After that it’s smooth and quiet.

It also seems smokier or smellier than before but that’s not necessarily true.

All injectors were redone by greazzer a few thousand miles ago, so I know they’re good, and all lifters are new INA.

I think you need to sit down and have a cup of coffee. All the sounds are perceived only.

Huh? No.

I have a lot of diesels as you can see from my signature. The knock that pushed me to work on lifters, injectors, and then DVs is for the 602 in my 91 300D 2.5.

I did have an issue of misfire/broken cam due to bad work from an MB shop, on one of my 1991 350SD cars. But that’s entirely separate.

Injector knock is only associated with my 1991 300D - the only cross between them is that the 603.97 and the 602.96 both use the same model injector in the angled design.

Edit - I see you edited (I had quoted your original which has changed). Perceived? It’s definitely noticable in terms of the difference between this 602, and my 601/603 engines. Let alone my 616/617.

Good luck with the fixed. If I were you, I would push them off the cliff. Out of sight, out of mind.

Huh??!? Why? If that’s the case why are we here? These are great driveable classics and a hobby for many of us. Getting some kinks out is part of the job.

My w123 cars are like new. My w126 with the broken cam was like new, it won’t take much to get it back - it’s already running great and the ax still holds charge The other w126 I’d drive across country tomorrow. The w201 is like new from original owner.

Perfect is the enemy of good enough, sure, but it doesn’t mean we shouldn’t stick to keep our cars close to new.

So, I can say that the injector sound is the same after replacing injectors and DVs. To be honest I can’t isolate the sound via stethoscope. I can isolate it via cracking fuel lines.

Definitely not a lifter either.

Any other thoughts?

Noticeable at idle. Louder and more obnoxious just off idle through 1200-ish RPM, then silent.

Mine does something like this, it seems to make the most noise at 1,200rpm, and sounds fine over 1,800. Mine also vibrates a lot at idle, which my mechanic assured me was just the engine mounts. If you figure it out great, but I don't think it's going to stop the car from going thousands or hundreds of thousands of more miles. Are you getting good power and economy?

I do think these engines make more noise/vibration at low rpm compared to om 602 or 606. All of the 602's I have seen are like this and 606 are smoother. But the 606 gets 33mpg, and the 602 is capable of 47mpg (I was surprised, but managed this on a 200mile trip recently).

Car runs absolutely beautiful, sounds fine, and is smooth.

5 cyl engines are less balanced than 6cyl, so there’s that. But a good running one is no issue at all.

To get 47 MPG you must have been driving 35-45 mph with no stops and no AC.

When I had my '91, cylinder #2 had a nailing injector. Swapping injectors did nothing. Turned out, in the IP, the delivery valve was almost a 1/4 inch too high, causing over fueling. I bought a used injector pump, but never got to install it.

I was going 55-63 most of the time. I couldn't believe it either, but I filled it up until I could see fuel in the filler neck both times and it used 3.1 gallons to go 142 miles. I had to check the distance on google because the odometer doesn't work. It was almost as good as the manual transmission TDI I have.

That’s interesting. How did thst happen? Was there a way to solve it through ip work?

May be worth swapping a good used IP at some point just for fun, and then I could re-seal mine.

It gets more interesting.

I’m on a 1000 mile road trip, and I’ve been doing a lot of highway. I’ve been running at
high concentrate cetane additive (Archoil).

After high speed runs it idles beautifully and quiet. No tapping at all.

After idling too long or sitting off for some time, the tapping is back again. Short slow driving keeps it there, these trips don’t get rid of it the noise. Only a hundred miles at high speed does it. It’s not like I haven’t driven like this plenty before but I think the difference is that I’m checking idle very soon after coming down from high speed use.

Maybe you should consider trying a water injection?

Not sure what that would have to do with anything? What is the purpose? I’m not familiar.

I tried another thing - foot off the interstate and didn’t turn off the engine. Perfect idle, quiet. Blipped the throttle, no knock.

Let it idle a bit, blipped the throttle. The knock came back around 1100rpm. Then it tapped as always.

The other thing that is odd is that the injectors won’t stay dry. The hard lines are torqued properly, and the return hoses were really hard to get on, but are indeed on.

I had removed the tygon lines because they seemed soft and easy to remove. But the wetness remains hundreds of miles later.

Cleans out the cylinders, rings and any carbon on the valves

Gotcha. None of those things are relevant to this issue though.

This is an issue from the fuel system.

Possible that the knocking is due to less than perfect mating of the sealing surfaces between the #1 injector and the #1 hard line (or a tiny hairline defect in the hard line itself).
The seal integrity could then vary with temperature and/or vibration.

Try swapping in a different hard line, or use a copper conical washer between the #1 injector and the current #1 hard line.

I’m pretty sure it’s #4 injector.

That’s the one that has the most noticable change when cracking lines.

You may be on to something, but it’s weird that after hard driving it will be perfect until I give it some pedal while idling, then it will come back again….

That would seem to be a seal, but what does the throttle have to do with it?

Maybe it would just take longer to introduce air or some sort of change with the flow at idle vs unloaded higher rpms??

The higher cylinder temperature when cruising at higher speeds allows for a more robust combustion event (as does higher cetane fuel).

Once the cylnder temperature comes down, like sitting overnight, the knock returns.

Blipping the throttle introduces more air incursion (think venturi effect).
This dampens the injection event and recreates the knock, even when the cylinder has not yet cooled.

That makes sense.

So the issue is air incursion then.

I had videos as case in point…

Sounding good after a long drive:

https://youtu.be/vu2ALi8Nve0?si=zHXafFU6nQ_zayRQ


And reversion not long after:

https://youtu.be/xrRdcK4W2D0?si=vih26sykG77psdRh

Thus is entirely repeatable.

Someone had been in the IP before I got the car.

There are little notches in the delivery valve that are supposed to line up with brass fingerlings (lack of a better word) and drop into place. You can see them from the top view of the IP with delivery valve removed.

I think that’s a different situation than air ingress though… I don’t recall that there is a ridge or other alignment spot to force the valves to align to anything.