Redwood Chair
- Stock PSI Or Bust -
- Joined
- Mar 23, 2007
- Location
- Ocean Beach S.F.
Hello Guest, welcome to the initial stages of our new platform!
You can find some additional information about where we are in the process of migrating the board and setting up our new software here
Thank you for being a part of our community!
B20E...MSExtra
Have you attempted to mount a knock sensor?
flat tappet cam... good luck.
I'm running my B20 off ms2extra, full sequential with LS COP setup, and tried a few methods of knock detection. DET cans were the only way to tell. Even then, its hard to differentiate between cam noise, valve/lifter float, and knock.
flat tappet cam... good luck....I'm running my B20 off ms2extra, full sequential with LS COP setup
I had some success.
I am running a mostly stock B20E using a Bosch flat response sensor mounted on a stud on the left side of the block just below the head / block mating surface between cylinders 1 and 2. I used a Digilent analog discovery USB scope and the FFT in the Waveforms software to examine the frequency content of the signal. Yes, the B20 generates lots of noise; but, I was surprised to find that most of that noise occurs outside of the characteristic knock frequency of the B20. I will put a caveat on that comment because I did not test for noise in the characteristic frequency range beyond 4000 RPM.
The signal processing board I fabricated used a fairly common architecture. The knock sensor is connected up front to a differential input op amp to reject common mode noise. The output of the op amp is connected to a Max 7490 tuneable switched capacitor filter. The filter is configured as a band pass centered at the what I think he characteristic knock frequency is; but, I can't remember what the band pass roll off is. The output of the band pass filter feeds into a precision rectifier based upon a Max 4253 Op amp. At one point I was experimenting with using a switched integrator on the output of the precision rectifier; but, when I look at my prototype board it looks like I did not use that feature. I think I decided that it was not necessary with the count feature in the MSExtra software. The output of the precision rectifier appears to be connected through a final op amp which provides some adjustable output gain from the precision rectifier. The clock frequency for the switched capacitor is provided by a trinket overclocked board. The trinket allows me to adjust the clock frequency (center point of the band-pass) in software to experiment with different tuning for the knock frequency. The trinket also does double duty as I use its on board power supply to provide isolated power for all the op amps.
I believe I have the clock frequency set to give a tuned band pass somewhere around 3000 - 3300 hz with fairly sharp roll off. I am doing this from memory because I have the proto board in my hands; but, I am not sure where my design notes are.
I did a little testing of the board by cranking up the advance just in the ignition map cells around 3000 RPM 60 - 70 kPa such that I was able to detect audible detonation. The board generated a knock output under these circumstances and I did some initial fiddling with the knock set up in MSExtra to suppress knock. In my non exhaustive testing, the board did not generate extraneous knock outputs.
My MS board is set up with the knock sensor input and the knock sensor has been wired up to the terminals in the MS case; but, I have not implemented the knock sensing feature. My prototype board will not fit in the MS case slots, too narrow and too tall because of the use of breakout boards. The prototype board is also ..... clumsy. I used breakout boards to mount the the op amps and a proto board with a lot of surface wiring. I wanted to have a proper circuit board fabricated to allow direct mounting of all the SMDs with no green wiring required and design the board so that it would fit in the MS case slots.
I returned the ignition map to its pre detonation inducing condition and have not been super motivated to do any further work. The original thought was that a knock sensor might be a benefit in the event of some variable fuel quality. However, that has yet to present itself as a problem. With the current ignition map running 91 or 92 R+M/2 fuel, knock is pretty much a non audible event. If we get another COVID 19 lock down boredom might motivate me to go the next steps this winter - if I can find my original design notes and schematic.
ignition map running 91 or 92 R+M/2 fuel, knock is pretty much a non audible event. ...
But I can say that 22psi@7k rpm, theres a lot going on.
If this info is true below...then indirectly, this distributor is detecting knock
123 Ignition Explained
Spark Balancing - The 123 ignition analyzes engine performance on a per cylinder basis and adjusts the spark timing and energy levels of each cylinder in order to balance the power levels - optimizing engine operation. This spark balancing function, derived from Formula 1 engine management systems, has not been available for classic cars until now.
I'm not plugging 123, but they do list timing degrees for various engines with a given distributor...scroll down this page to see them....with a cross reference to a given Bosch distributor.
PS Pricing wise, I would shop around, in US and Europe
I don't think it is detecting knock, directly or indirectly.
123...had a vac reference.
If looking at distributor advance curves, they are usually based on distributor RPM and not crank RPM.
...detecting knock by looking at...
Joke?s on you, that never happens.... when I eventually finish grad school and start making some money.
...123...hook it up to my laptop?
So the 123 is pretty much like a pertronix setup but with more brains built in and I could hook it up to my laptop?