Knock sensors for FE

[gregaba]

I would like to install knock sensors on my 428. Aluminum heads.
Anybody done this and how did you mount them and where?
Thanks
Greg

[My427stang]

I think that the juice isn't worth the squeeze.  A knock sensor is a fussy tool that needs to know what right is to ID what wrong is. 

I think staying within known spark/load/mixture tables will keep you exactly where you need to be in a modern EFI.  In fact, if you weren't trying to run uber lean and therefore requiring max timing, like a modern EPA/mileage constrained car, not sure it would ever do anything anyway

[gregaba]

Ok
Thanks
I was worried a little about detonation and thought I could use it to retard my timing in a bind.
Greg

[WConley]

As Ross said, It takes a lot of work to get a knock sensor right.  At Ford, we had to dyno test extensively to find the sweet spot.  You look at the signal on an oscilloscope for sensors in different places on the engine.

Engines make a lot of mechanical noise.  Changing anything on the engine (intake, cam, stroke, piston design, even accessories) changes what the sensor sees.  There is usually a spot on a head or the intake where the knock noise will really come through while the normal mechanical noise doesn't.  It's a black art.

[gregaba]

Well I don't have the resources that Ford has or the knowledge to set one up right so this will just be one of my dumb moments.
Greg

[cjshaker]

A knock sensor is a fussy tool that needs to know what right is to ID what wrong is. 

Ross is right. Manufacturers have sophisticated multi-million dollar equipment/software to determine where to place knock sensors, and then they have to be 'taught' what to listen for with more sophisticated software. Trying to do that in an old engine is next to impossible without the same research, computer aids and software to run it. If it were possible to use a knock sensor in an old engine to just trigger a light, then it would probably be set off by a multitude of noises. I'm also not aware of any way to 'filter' the noises that a sensor picks up, so adapting a stand-alone system to a non-computer controlled car would be near impossible.

Honestly, the best knock sensor for old engines is learning to read plugs. Looking for signs of 'peppering' on the plugs will indicate you are either running too lean, or have to much compression for the fuel, or that the timing curve is too fast or total too high for the load.

[gregaba]

I was thinking that with a sensor I could get by with running 100 instead of 110.
I guess I will just have to do it the old fashioned way.
Its just that I am a little lazy and all ways looking for the easy way.
Greg

[CV355]

As Ross said, It takes a lot of work to get a knock sensor right.  At Ford, we had to dyno test extensively to find the sweet spot.  You look at the signal on an oscilloscope for sensors in different places on the engine.

Engines make a lot of mechanical noise.  Changing anything on the engine (intake, cam, stroke, piston design, even accessories) changes what the sensor sees.  There is usually a spot on a head or the intake where the knock noise will really come through while the normal mechanical noise doesn't.  It's a black art.

Well said.  NVH analysis is insane.

[My427stang]

I was thinking that with a sensor I could get by with running 100 instead of 110.
I guess I will just have to do it the old fashioned way.
Its just that I am a little lazy and all ways looking for the easy way.
Greg

I am not sure it's going to be that much work, were you asking about tuning an EFI system?   Pretty much everything under the sun has been done, if you have the mixture right, we can tell you a timing curve, and you may not even need to run 100. A lot of us are making a lot of power without race gas.

[gregaba]

I had my machine work done in Texas [block blueprinted] and he told me I should run 110.
I had the block decked to 10.150, cam and mains aligned bored, crank assembly balanced. and lifters sleeved. Block was decked for MLS gaskets.
TFS heads version 3.
Autotec pistons -flat top with 5.000cc, bore of 4.165, comp. dist is 1.675 with the small ring pack.
Brent told me I would be around 11-4 compression ratio max.
I need to get the short block together to check how far down the bore the pistons are to figure it out for sure. 
I will be running all the FE Power front parts and the intake adapter with a TFS 351C single plane intake.
I am deciding between a megasquirt 3 or a Holly EFI. I haven't ordered one yet.
For the first 6 months I am going to run the Ford D cam and if I don't like it will change to a custom solid roller.
Going in a 63 Galaxie 2 door sedan that was a race car from birth. Haven't weighed it yet but there is no heater, sound deaden er etc in the car.
I would like to run pump premium in it but don't think it would work.
As far as the ignition goes I haven't decided yet. depends on what EFI system I go with.
Just a weekend driver I want to have some fun with.
Greg

[1968galaxie]

I would be very surprised that you need to run 110 octane fuel with your combination.
11.4 CR with a tight quench distance (0.050") and the reasonably decent TFS combustion chamber
along with a reasonably big camshaft should run quite well on 94 octane.
Not sure what pump fuel you have - perhaps some octane booster in a tank of fuel also.

Brent may have some comments on octane requirement here.
He has built a few TFS headed engines and would have a good idea regarding detonation with these heads.

Cool build. Are you going to dyno the build?

[gregaba]

I was thinking 110 would be a bit high for my combo.
The best I can do here is 91 octane so I contacted Rocket fuel and will be buying a 55 gallon drum of their fuel and mixing it with our sorry gas and see where I come out on it.
I was hoping the TFS heads would let me get by on at least 100 but if I can go lower safely then I will.
I will be assembling the short block in a couple of weeks and then when the heads come in I can finish it up.
It will only see a few thousand miles a year so the gas expense won't be to bad hopefully.
Greg

[chris401]

I would like to install knock sensors on my 428. Aluminum heads.
Anybody done this and how did you mount them and where?
Thanks
Greg

The typical place is just below deck and center of the head. Compared to what manufacturers do there is not an ideal place on the FE. With enough room for 3" header to block gap you could probably get away with taking a discrete plug of cast iron to the block between 2&3 and or 6&7 and tapping it for one. Just needs to be as close to the cylinder head as possible for it to read and have time to adjust.

As far as aftermarket systems I am clueless. Been thinking about adding one but have not done the research to know what fits my needs. Adjusting timing between 400 and 4000 feet above sea level gets bothersome after a while. Especially in the summer when ambient temperatures make it temperamental. Cruising is 12.6 up high vs 13.6 down low so I don't bother with jetting. Millions of regular folks and truckers made roadside adjustments in days past, now people think your broke down.

[My427stang]

So I am guessing that you are using a .050 Cometic, I get the same numbers as Brent

I'd say if running a common hyd roller cam with about 298 adv on the intake side, spread the centers to 112-114 and put it on 108, likely 114, the combination of the good chamber, tight quench, and not a ton of overlap will keep things happy on pump gas.   It's a big cam for a 428, but spreading them will tone it down to let the O2 sensor do it's thing and have a by product benefit of less "5th cycle" cylinder fill as it comes up on peaks

I doubt it would even have a funky timing curve or anything really that weird.  The rate of advance depends on the rest of the combo, but it won't need a lot of total or initial, and EFI allows you to crank with low initial and immediately kick in.  I'd prefer to see a 1/4 to 1/2 point less, and maybe a bit less cam depending on combo, but I don't think you'd have a problem if the cam choice matches your setup

I am not a "kill the compression with DCR guy" but the heads and flat top bought you a clean chamber which should work well.  Typically though I push decisions to be more (very simplified order)  cubes,  induction, cam for RPM range based on the first two,  then pick compression to support.  This is sort of backing into it, but sometimes we do what we have to do

ON EDIT: I was messaged that this was the D cam build, I forgot about your plans.  I am not by any means trying to revive the old argument we had, so please don't let me be the cause of a banter, but that much overlap, even though the DCR may drop, will fight the O2 sensor and create a situation that, if you can get to the peak RPM, will fill the cylinder to a point it likely will need to be spiked with some kind of octane booster.  How much depends on what you can do with the EFI for tuning and the conditions that you are experiencing at WOT.  As much as I would love to hear a D cam in a stout engine, an understatement is that it's really a tough cam for the parts you are using,  Additionally, as said earlier, but more so, I would expect the knock sensor to be a no-go with a solid lifter cam, they are noisy and slappy.  May not be the same frequency as detonation, but it's more noise the sensor would need to overcome or blank out

If you go this route, be sure to have the EFI tuner go open loop at any significant throttle setting, in fact, it may need it all the time, and pull timing out at RPM and load until you can sneak up on rate and total

[blykins]

I had my machine work done in Texas [block blueprinted] and he told me I should run 110.
I had the block decked to 10.150, cam and mains aligned bored, crank assembly balanced. and lifters sleeved. Block was decked for MLS gaskets.
TFS heads version 3.
Autotec pistons -flat top with 5.000cc, bore of 4.165, comp. dist is 1.675 with the small ring pack.
Brent told me I would be around 11-4 compression ratio max.
I need to get the short block together to check how far down the bore the pistons are to figure it out for sure. 
I will be running all the FE Power front parts and the intake adapter with a TFS 351C single plane intake.
I am deciding between a megasquirt 3 or a Holly EFI. I haven't ordered one yet.
For the first 6 months I am going to run the Ford D cam and if I don't like it will change to a custom solid roller.
Going in a 63 Galaxie 2 door sedan that was a race car from birth. Haven't weighed it yet but there is no heater, sound deaden er etc in the car.
I would like to run pump premium in it but don't think it would work.
As far as the ignition goes I haven't decided yet. depends on what EFI system I go with.
Just a weekend driver I want to have some fun with.
Greg

You don't need straight 110 but I wouldn't chance it on pump gas either, *especially* on 91 octane.   Mixing is the correct solution and you can mix something like 110 with 91 (50/50) and be ok.  I've had good luck with VP or Sunoco fuels.  You can try some VP C12 or Sunoco 110. 

If it were 11:1 and 93 octane was available, I'd go for that with the correct combination of parts.   

The "D" cam will be nostalgic for sure, but it will be down 20-30 hp to a modern camshaft that's smaller.  If I remember correctly, it has 100° of overlap (or maybe a hair more).   Jay's first camshaft in his cylinder head test mule engine had 102° and it was just pushing power out the exhaust.