Knock sensors revisited

[5DIGITS]

BTW, it costs about the same to get the correctly made piston (hey vendors!) as it does improperly designed pistons.
Who will be the first to market a proper piston? Nobody You can get this stuff done yourself but you shouldn't have to for a platform that still has some market.
I also bank ZERO $ into obsessing about pistons speeds because the baddest import engines start off with high piston speeds and they keep stroking them more and not looking back.
They do what is necessary to prevent knock be it modified pistons, whatever.
If you build the shortblock and design the pistons exactly the same and ignore pistons speeds or heat, that is your fault, not the engine's.
Now, spending all that time modifying a cylinder head and you plan on putting an off the shelf improperly designed piston in, again, your fault when things don't work.
Less spark advance is not a bad thing. A sign of a good turbo setup is needing less spark advance.
Bad tuning practices carried over from older tech/poorly designed cylinder heads need to die.
Just because its hard to tune something using backyard techniques does not mean that engine combo is junk.
I don't like spending money so I would probably avoid the hard to tune cylinder heads so that backyard techniques will be good enough.

True - When effort is put towards the right hardware and its complimented with accurate tuning, everything will work together regardless if its new or old backyard technology or custom machined from top to bottom.

[Force Fed Mopar]

Would going to a flat top piston help anything?

[Force Fed Mopar]

To put things into perspective, the desire to run a 2.5L engine to 6500RPM is much like running a 2.0L to 8300RPM and may be something to consider during the next up-shift.

Just as a note, I always shifted mine before 6k, as it didn't really pull any more past that. Usually shifted around 5k-5500.

[wheming]

So are there any off the shelf forged pistons that are better than the rest? Which forged piston has this shelf dimension the broadest?
And this should be a bit less of an issue with the G head?

[5DIGITS]

So are there any off the shelf forged pistons that are better than the rest?

There are many high quality units on the market but consider the performance levels that are being attempted, to determine whether the investment is justified.

Which forged piston has this shelf dimension the broadest?

As previously mentioned, many manufacturers will support customer supplied dimensions.
This comes back to understanding the output expectation and how to best compliment the other components being used.
At that point, the decision can be made regarding forged components or cast and if the upgraded parts should be custom or off the shelf.

And this should be a bit less of an issue with the G head?

Yes.
The G-Head is a bit more forgiving when considering spark advance and elevated cylinder pressures.

[Force Fed Mopar]

Update: Got the Lebaron DD'ing again, and have been working on re-tuning nearly every day for a month now. I have come to the conclusion that my valvetrain is setting off the knock light in most cases. Most often I'd see it after lifting to shift in normal driving. It'll come on in the higher revs sometimes under power, but it's not really regular, and boost level doesn't seem to affect it much, only rpm and maybe engine temp. Even out of boost, just running it up to 3500ish in first gear while taking off easy from a light would set it off every time. This is all on completely stock timing and boost levels.

Revving the engine in neutral under the hood, I can hear the valves pretty clearly. Not sure why, as I am running 2.4 lifters w/ the hole in the top w/ a pretty good shim under them. They are used lifters, so I suppose they could just be worn out. The valve springs are also pretty old, they are from an '88 Daytona So they may be allowing some slap also? Cam is a stock turbo roller. Checked all brackets on the engine, everything seems to be tight. Engine does have a lot of vibration, shakes the hood and headlight doors a lot at idle (poly mounts), not so bad when revved up but still a little bit. Oil pressure is good.

So I've been raising the threshold a little at a time trying to eliminate the engine noise, but I'm worrying now that in doing so, it may not pick up on actual knock? How much louder is spark knock than a noisy valvetrain?

[shackwrrr]

Update: Got the Lebaron DD'ing again, and have been working on re-tuning nearly every day for a month now. I have come to the conclusion that my valvetrain is setting off the knock light in most cases. Most often I'd see it after lifting to shift in normal driving. It'll come on in the higher revs sometimes under power, but it's not really regular, and boost level doesn't seem to affect it much, only rpm and maybe engine temp. Even out of boost, just running it up to 3500ish in first gear while taking off easy from a light would set it off every time. This is all on completely stock timing and boost levels.

Revving the engine in neutral under the hood, I can hear the valves pretty clearly. Not sure why, as I am running 2.4 lifters w/ the hole in the top w/ a pretty good shim under them. They are used lifters, so I suppose they could just be worn out. The valve springs are also pretty old, they are from an '88 Daytona So they may be allowing some slap also? Cam is a stock turbo roller. Checked all brackets on the engine, everything seems to be tight. Engine does have a lot of vibration, shakes the hood and headlight doors a lot at idle (poly mounts), not so bad when revved up but still a little bit. Oil pressure is good.

So I've been raising the threshold a little at a time trying to eliminate the engine noise, but I'm worrying now that in doing so, it may not pick up on actual knock? How much louder is spark knock than a noisy valvetrain?

Hey, I have the other half of my set of 26995 springs. my van hasn't even hinted at setting the knock light since Ive installed them. I have just the springs all you need for retainers are stock gm 3100/3400 retainers.

I bought the springs for 180ish I would sell the other half for 95 shipped if you are interested. I was able to do them on the van with an air fitting and an on car spring compressor.

[Force Fed Mopar]

I actually have a set of those also, just need to get retainers. I think that will be my next real mod on the car. Summit sells a 8mm beehive retainer for Jeep 4.0's that I think should have the right keeper angle for the stock keepers, so I'm gonna order one and see. If so, I'll order a set and install. If not, guess I'll rock some GM retainers like everyone else

[zin]

Summit sells a 8mm beehive retainer for Jeep 4.0's that I think should have the right keeper angle for the stock keepers, so I'm gonna order one and see.

I'm very keen on seeing if they do mate up properly, not knowing is the main reason I haven't gone this route... I just don't like the idea of the keepers and retainers not having matching angles... I know the other way works, but I get nervous in regards to the possibility of valves dropping, I've seen what can happen, and I don't what to be "that guy".

Mike

[Force Fed Mopar]

I posted this in another but I'll update it here also. I finally got brave enough to just set the knock threshold to 5v so it wouldn't come on and started tweaking up the timing. So far so good and it feels stronger then it ever has at 15 and 18 psi. Also running it slightly rich seems to make better power also, 10.8-11.0. Could be that my sensor isn't reading right anymore either, it's a few years old now.

[wheming]

I posted this in another but I'll update it here also. I finally got brave enough to just set the knock threshold to 5v so it wouldn't come on and started tweaking up the timing. So far so good and it feels stronger then it ever has at 15 and 18 psi. Also running it slightly rich seems to make better power also, 10.8-11.0. Could be that my sensor isn't reading right anymore either, it's a few years old now.

How are you monitoring knock then, just audibly? Or do you have the det can?

[Force Fed Mopar]

Just listening by ear, no det can yet. I know, dangerous...

[wheming]

I'm not judging...
I ask because I've experienced the similar occurence with Slug after rebuild (which included forged pistons) and using a turbonator cal with the CEL on knock retard.

Now, with MeanMini's rebuild it is observed also at low boost levels.
I'm inclined to believe it is noise or a sensitivity issue, as others theorize. I think it would be nice to have this feature and have it be a true indicator. I'm sure I'm not the first to wonder if we can use a better knock sensor from a different application?

Having at least a det can setup we could hear with our ears, but I'd like the computer to be able to hear "real" knock and then respond with retard and CEL flash.

While it may be agreed to not be an optimum way to tune, we do need to be able to recognize/observe/measure real knock to make sure we maintain a safety margin when trying to optimize a tune.

[ShelGame]

Any knock sensor is really just a microphone. As such, it picks up all noise.

In our system, the noise is used to charge a capacitor. The ecu then reads the voltage stored in the cap. This voltage is then compared to the knock threshold and rpm to determine if its knock, or just normal engine noise.

When you make major changes to the engine, it may naturally make more noise. You will then need to retune the knock threshold vs. rpm table to tell the ecu what voltage levels are OK and what is generated by knock. That's where the det can comes in handy.

If you modify the engine (even by just raising the boost) the stock knock threshold will no longer be valid.

Getting a 'better' knock sensor won't automatically fix everything. You still have to tell the difference between OK noise and knock.

[5DIGITS]

I'm not judging...
I ask because I've experienced the similar occurence with Slug after rebuild (which included forged pistons) and using a turbonator cal with the CEL on knock retard.

Now, with MeanMini's rebuild it is observed also at low boost levels.
I'm inclined to believe it is noise or a sensitivity issue, as others theorize. I think it would be nice to have this feature and have it be a true indicator. I'm sure I'm not the first to wonder if we can use a better knock sensor from a different application?

Having at least a det can setup we could hear with our ears, but I'd like the computer to be able to hear "real" knock and then respond with retard and CEL flash.

While it may be agreed to not be an optimum way to tune, we do need to be able to recognize/observe/measure real knock to make sure we maintain a safety margin when trying to optimize a tune.

What would help these cars is a knock sensor the utilizes 'sensor ground' rather than vehicle ground, to aid in noise isolation.
With the current single wire sensor, injector coil inductive collapse and alternator noise can be induced into the signal based on 'chassis' ground conditions.
If the production sensor is used, discard the passenger side clip-on ground strap and drill the opposing sides of the mount for bolt on wire eyelets.
Furthermore, add an eyelet equipped strap from the injector harness ground to the frame side of the passenger side mount.
This will improve the knock sensor accuracy especially if the sheet metal bulk head ground bolt is stripped or the passenger side mount strap has deteriorated over time.

[zin]

Ken, am I understanding you, in that you're suggesting a better ground to lower the induced voltage in the single wire of the knock sensor?

I'm wondering a couple of things, first, that some folks frustration with knock not being well controlled could be related to 30year old capacitors not holding tolerance and an ECU that still expect such. Which begs the question, can that CAP be replaced?, or is there another "work-around" that would be better?

Second, if my understanding is correct, the sensor we use generates it's own voltage via piezoelectric effect, so adding a "clean ground" between the intake and the sensor wouldn't dramatically help things, but would adding a grounded sheath around that single wire for a distance have benefits worth the investment in time?

Mike

PS Rob, thanks for the explanation of the knock circuit, it helped make things "click" in me brain!

[cordes]

Would installation of a multi-wire modern knock sensor be possible with stock electronics?

[ShelGame]

What would help these cars is a knock sensor the utilizes 'sensor ground' rather than vehicle ground, to aid in noise isolation.
With the current single wire sensor, injector coil inductive collapse and alternator noise can be induced into the signal based on 'chassis' ground conditions.
If the production sensor is used, discard the passenger side clip-on ground strap and drill the opposing sides of the mount for bolt on wire eyelets.
Furthermore, add an eyelet equipped strap from the injector harness ground to the frame side of the passenger side mount.
This will improve the knock sensor accuracy especially if the sheet metal bulk head ground bolt is stripped or the passenger side mount strap has deteriorated over time.

Don't know why I didn't think of that myself. Makes me wonder now how many guys that seem to get the knock sensor going off based on 'valvetrain' or 'piston' noise actually just have a bad ground or 2. I can imagine that if your ground wire to the manifold/fuel rail is working well, but another ground to the engine is not, that you could indeed get some bad flyback voltage on the ground wire closest to the knock sensor. The spark plugs / coil would be the prime example as they are grounded thru the block.

- - - Updated - - -

Would installation of a multi-wire modern knock sensor be possible with stock electronics?

We'd have to find a 2-wire knock sensor with the same output characteristics as ours. But, if there is one, I imagine it could be an improvement. At least in reliability.


EDIT:
Perhaps a '95 Neon knock sensor would fit the bill?

http://www.ebay.com/itm/Ignition-Kno...e601cf&vxp=mtr

I think it must have similar (if not the same) output characteristics - the '95 FCC is basically an SBECII, specialized for 4-cylinder only duty.

[Force Fed Mopar]

What would help these cars is a knock sensor the utilizes 'sensor ground' rather than vehicle ground, to aid in noise isolation.
With the current single wire sensor, injector coil inductive collapse and alternator noise can be induced into the signal based on 'chassis' ground conditions.
If the production sensor is used, discard the passenger side clip-on ground strap and drill the opposing sides of the mount for bolt on wire eyelets.
Furthermore, add an eyelet equipped strap from the injector harness ground to the frame side of the passenger side mount.
This will improve the knock sensor accuracy especially if the sheet metal bulk head ground bolt is stripped or the passenger side mount strap has deteriorated over time.

Don't know why I didn't think of that myself. Makes me wonder now how many guys that seem to get the knock sensor going off based on 'valvetrain' or 'piston' noise actually just have a bad ground or 2. I can imagine that if your ground wire to the manifold/fuel rail is working well, but another ground to the engine is not, that you could indeed get some bad flyback voltage on the ground wire closest to the knock sensor. The spark plugs / coil would be the prime example as they are grounded thru the block.

Never thought of this myself either. I think I have a ground from the intake to the firewall, not sure about the motor mount. If it does it's just the clip on type. I'll have to add grounds as mentioned and see how it does.

[cordes]

Don't know why I didn't think of that myself. Makes me wonder now how many guys that seem to get the knock sensor going off based on 'valvetrain' or 'piston' noise actually just have a bad ground or 2. I can imagine that if your ground wire to the manifold/fuel rail is working well, but another ground to the engine is not, that you could indeed get some bad flyback voltage on the ground wire closest to the knock sensor. The spark plugs / coil would be the prime example as they are grounded thru the block.

- - - Updated - - -



We'd have to find a 2-wire knock sensor with the same output characteristics as ours. But, if there is one, I imagine it could be an improvement. At least in reliability.


EDIT:
Perhaps a '95 Neon knock sensor would fit the bill?

http://www.ebay.com/itm/Ignition-Kno...e601cf&vxp=mtr

I think it must have similar (if not the same) output characteristics - the '95 FCC is basically an SBECII, specialized for 4-cylinder only duty.

That sensor could be the ticket. I seem to recall Bucar mentioning that the frequency tuning is based off of bore dia. That would bode really well for us in that aspect. That said, I would still be surprised if it only took a little extra wiring to use it.