Agreed. I wonder if any such thing exists.

Not that I'm stuck on MSD (just happend to already have the 6AL in my car when I bought it, and only another $120 bought me the boost timing master), but here is MSD's most recent that gives even more programming options
MSD 6530 Digital Programmable 6AL-2 Ignition:eBay Motors (item 220506987834 end time Dec-07-09 04:30:55 PST) A good price, too!

I don't have the breakdown numbers regarding the contribution of the vacuum and centrifugal advance components to total advance. I was running around 26 total, but that is with no load. I am going to be a little more diligent in the future and try to obtain the advance data a bit better, and I will surely check it when it goes back on the dyno.

I agree with JFairman regarding the cost issue, but as I mulled this over coffee this morning, I decided that after two rebuilds, I ain't taking an more chances. Maybe if I had a regular 930, which is a more or less known quantity, maybe, but since I have basically a one of a kind deal, I will go J&S. I have no doubt the MSD will perform quite well, I was just looking for some feedback when I started this thread.

I don't think my piston failure was due to detonation, although some have told me it was. I don't want to do this rebuild thing again anytime soon.

As an aside, the 968 and the 986 in the garage, which reflect approimate generational jumps in Porsche engineering over CIS, are amazing with the sensors and the functionality of the CPU's to perform engine management. Static CR is way high, they never knock, and are good on gas if I don't beat the crap out of them. Not at all like the little turbo SC.

I'm intrigued with the possibility of the J&S, though their web site didn't give out as much info as I would have liked as to how to hook it up with CIS and standard non-ECU ignition applications (maybe I need to look closer). It would be nice to run as much advance as needed and not worry about individual cylinders detonating as a result of timing and/or fueling issues. If you go there, do please keep us posted!

Thanks John! Much better.

Now for the stupid questions:

Is there a particular model number of the Safeguard Porsche 930 applications that a person should use for ordering purposes?

I presume this will hook right up to my MSD6AL, since I'm not using the stock Porsche CDI unit any longer.

Regarding the knock sensor, does it (or do they) come with the kit? Isn't a sensor needed for each cylinder?

And, where and how do you mount the Knock Sensor?

I remember pics of a circuit board, is there a weather sealed case for it?
How long has it been in production, how many are in use on street cars, and how is reliability?

Seems you would have to have 6 sensitive discrete knock sensors and with a noisy air cooled motor with no water jacket to dampen and partially isolate mechanical noises what are the chances of timing errors?

If the electrical components are today's generic chinese computer grade I would be skeptical in the long run.
If they are hand selected/tested military or hospital grade or something similar and in a weather sealed case than I would have more confidence that it wouldn't go haywire or into a default mode and blow a motor.

..and of course, how much $

The "universal" version was redesigned last year, and I included two features especially for Porsche.

One is an "ac" output to simulate a mag pickup signal. MSD triggers on the rising edge of the mag pickup signal, but Bosch CDI triggers on the falling edge. I included an output polarity select mode switch so the unit could trigger either type.

The other is provision for a second coil driver, for use with dual distributor. One box can drive two coils, without need for MSD. The second driver can be added for $20 on request.

The unit is designed for Bosch or other "donut" type knock sensors, available from many sources for around $50. We can supply them with your order, including shielded cable with connector and pins.

These sensors have an 8 mm mounting hole in the center. I've seen posts about the factory knock spiders, so why not use them if they are available. Sensor mounting torque is 15lbs.

Note that only one sensor is required is do individual cylinder control. Mounting a sensor on each cylinder won't help to identify the knocking cylinder, as they will ALL hear it.

The knocking cylinder is the one that just fired. Software "knows" that cylinder won't fire for two more revolutions. When it's ready to fire again, software dials in the calculated amount of retard for THAT cylinder.

Sealed units are not available. Some of the components are rated for 70° C, so the unit must be mounted in the passenger compartment. Standard harness length is ten feet, but I can make them to any length you want for an additional charge.

I've been producing individual cylinder knock controllers since 1991. Thousands in use. Very low failure rate.

I want to say thanks for the product development and availability and your description of it.
Are any of them in edurance racing or drag racing competition use?

Also wondering does anyone use such a device for piston aircraft or are the twin ignitions used in aircraft tuned conservatively and don't need it?
Seems altitude and temperature changes while flying would require active ignition control.

Racers are very secretive. If any are using my product, they keep it to themselves.

I've trolled the web looking for high profile users, or just guys going fast, but haven't found much. I did turn up this link:

http://www.enginecombination.com/viewtopic.php?t=153

Link to new aircraft ignition using combustion pressure sensors for knock control:
General Aviation Modifications, Inc.

Note they use the term "safeguarding the engine from detonation".

I just read six pages of discussion on this topic, sceptics and believers alike. The general consensus is that knock sensing on air cooled 930 turbo engines does work. The caveat is that it may sense false positives and retard timing when not necessary, leaving HP on the table.

Anyway, good reading, form your own opinions.
http://forums.pelicanparts.com/911-930-turbo-super-charging-forum/405586-knock-sensor.html

Here is some John De Armond stuff from 1992:
http://yarchive.net/car/knock_sensors.html

Edit:
BTW, John De Armond is like Early_S_Man. He is a pretty sharp electronics guy. He has a couple of websites which may or may not be of interest.

Pat:

I've run across that link several times, and at one time tried to get in touch with him. I was trying to contact him to run a test on it.

From what I was able to gather, he's an engineer that retired from the TVA, where he was involved in nuclear instrumentation. He seems to know a lot about many subjects.

Here's another one from JDA:
Ignition (John De Armond; Dave Williams; Jon Lusky)

Lots of great information about ignition theory.

I didn't run across these links until I got on line in '99. I have to admit I was a little hacked off to see he was talking about one upping me in '92, but I guess it never came to anything.

John, this is a guy that built a Pyrex cylinder head to observe spark. Hands on testing is just so cool. Read some of his powerplant stories if you have time to kill, they are pretty funny. One of my buddies used to build nuke plants (I have stories the general public should not know!).

What is the knock frequency range, around 6 kHz? I suppose it is chamber volume dependant?

Bosch wide band knock sensors respond to all knock frequencies.

Well, I got my J&S in the mail testerday. I also picked up a used MSD retard unit on Ebay for calibration purposes). John pre-wired the harness for the dual knock sensors. I have determined that I cannot use a factory bridge unit for the knock sensors, as my old SC heads have no provision for mounting the bridges.

I am considering simply putting in a long 6mm bolt on the shroud on both sides of the engine, and bushing the knock sensors from 8mm to 6mm to fit. At least this way the sensors will actually be mounted directly to the cylinder head, and I can get a bit of symmetry from bank to bank.

Please keep us informed on this install. I am very interested in knowing if you get any false positives.

Knock Sensors Installed
 

Well, I did manage to get the sensors installed today. I ended up drilling out the 6mm holes which hold the shroud on #3 and 4, and tapped them to 8 mm. I put in 45mm studs, with two washers on the bottom. The sensors are pretty secure, and should get a good signal there. I suppose it's not ideal, but we'll see how it works out.

Since I had the shroud on and off a few times, I put the 14mm plug rings I was using for cylinder head temperature sending right on the fins of the head on #1 and 3. JB Weld with a spring paper clip. The bits of wood shim are there til the JB Weld dries.

http://forums.pelicanparts.com/uploa...1259450759.jpg
http://forums.pelicanparts.com/uploa...1259450774.jpg
http://forums.pelicanparts.com/uploa...1259450791.jpg
http://forums.pelicanparts.com/uploa...1259450803.jpg

Just a thought, if you have a choice I am wondering if it might be better to be opposite the cam chain end of the motor. You are mostly going to be sampling two cylinders and it might be best to try to be a isolated from the cam chains as possible. I has been said that Porsche redesigned the chain tensioner on the 3.6 to reduce noise that might interfere with knock sensing.

A second thought might be to sample the center two cylinders hoping any detonation in the adjoining cylinders might be transfered through the cam towers. I do not know if the center cylinders might run hotter and be more issue prone. I think the one near the oil cooler might be the hot spot but not sure.

I am thinking such a set up without the bridge may or may not neccessarly be helpful if one cylinder is going lean like if one injector is clogged. More so if the motor is running over the edge.

Not an expert, just thinking out loud.

Hope it all works out no mater.