cjernigan

iTrader: (13)

I'm in to help all I can to get a 99-00 board developed.

yertnamreg1218

Good new the computer is fixed. It was serious F'ed up though. They replace the lcd motherboard and hard drive...it cost me nothing!! So I pretty much have a new laptop.

The data was all recovered so I still have the schematics and layouts. I'm installing orcad now, so probably later tonight I will post the schematic with some explanations of what I'm doing.

99-00 is definitely next. I agree there is a huge untapped market, not to mention I think it would be cool if a bunch of un-organized shade tree mechanics made a product that beat DIY to the punch and was awesome. Anyway, I'm off to lab (embedded system in case any of you were wondering).

P.S. The guy that runs the board mill at school is retarded, so its taking longer than anticipated. I probably won't get the board made till the beginning of next week.

yertnamreg1218

I got the board back today. It looks good, but because we don't have plating capabilities, so I have to solder every component on both the top and bottom (for the connectors this is very difficult).

Braineack, we should talk about your board making capabilities. I should be able to have this debugged in the next week or two.

yertnamreg1218

This is the schematic. The pin numbers on the ECU connector do not match up with the actual connector numbers, so you guys can't really check that over (I didn't want to create a whole new part in schematic just for that connector).

Let me know what you think of it.

alden77

Hey 1218
Any new progress on this?' I am waffling between buying a DIY MS kit and soldering away and a PNP. If you get this adapter board up, I'll do the DIY. I'm in Indianapolis and I do know where Rose Hulman is, so I'll be a guinnea pig if possible.
Best Always
Alden

"The truth has always been free, and yet, it is priceless. Its absence is that which cannot be easily afforded."

yertnamreg1218

Sorry to dig up this old thread, but I feel its better to do that than start a new one.

Anyway, I finally got the motivation to debug the MS adapter board that I made back in April/Mayish. Several hours with a DMM later, the board is functional!!! I still need to check the extra circuits I added (A/C fix, EBC, Fans,etc.), but otherwise I'm thinking I'll get a few boards professionally made.

For those of you who may have forgotten, I've designed an adapter board that takes the DB37 from MS breaks out all the pins from it to headers and takes all the important signals and dumps them to the stock ECU connector. Basically its a PNP DIY if that makes any sense.

One more thing, I know everyone here for the most part uses LC-1's for O2. A friend of mine turned me on to the Just Another Wideband Controller (JAW). Its a Wideband Controller that you build yourself for alot less money ($45 + sensor cost). I got a chance to test mine last night and it worked really well. I'm not at all affiliated with him, but he does have a solid product that is cheap.

Lastly, Braineak I believe you said you could get boards cut for cheap. I'm thinking that by early next week I should be ready. Can you shoot me a PM with your email address so we can talk more in depth about this.

yertnamreg1218

I did a rough tune with Autotune and the car is drivable now. I'm super duper excited. I think I'm going to try to run with it at an AutoX Sunday in Lawrenceville, IL if anyone nearby wants to come check out the adapterboard.

Braineak, thank you for the information in the DIY MS thread. It has been a valuable tool in mine and everyone else's quest for MS glory.

soloracer

iTrader: (10)

94-95 version soon after????

Nitroracer

I just saw this thread pop up and read it through, thats awesome your new board has the car up and running. I'm still n/a powered but it would still be fun to give MS a go and this would make the build go a lot smoother.

yertnamreg1218

I finally have pictures to back up my talk of awesomeness and MS glory!!!



I fabricated an aluminum tray for everything to sit on. The board that I designed is the copper colored thing plugged in between the MS and the stock harness. The board above that is my wide band controller (Just Another Wideband). Seriously check that guy out, its so much cheaper than an LC-1 and mine is working great.

soloracer

iTrader: (10)

Nice work!

haledj

Sweet man, can't wait to see the PCB one.

Good work

yertnamreg1218

So as I've been debugging, I've been trying to think of other cool **** that I could add that would increase the utility of the board while not adding alot of cost.

I can add a 2 axis +/-1.7g accelerometer for next to nothing (cost wise). This would require running two wires inside the MS to the DB37. Adding this would in no way effect MS, but when logging this data would be logged in addition to everything else. Should be moderately useful (if not just cool) for autoX and other racing situations.

What do you guys think? Are there any other sensors or extra things I could add to make my board way awesomer than a MSPNP?

AbeFM

iTrader: (3)

There are some micro barometers, but I guess there's not a lot of point there. Are you out of pins already?

yertnamreg1218

When using MS-1 and extra, we have 2 analog inputs to play with. There are digital inputs, but those aren't as useful on a system such as an engine. It we do a 2 axis accelerometer that will use both channels leaving us with no inputs.

I'm thinking I should also include a basic knock sensor amplifier/detector circuit. It'd be way better than having to buy a KnocksenseMS, and would be give a less cluttered installation.

soloracer

iTrader: (10)

Quick! Make me one for a 95 w/ an integrated knock sensor before I get my parallel install done!

AbeFM

iTrader: (3)

On my long long list of theories and things-to-check-into, there are two chips in particular I'm interested in, two classes of chips:

1) Thermocouple readings. Ideal for EGT's.
2) Knock Sensor integrated chips

Of the later, there are chips all set up to do intelligent, tunable knock sensing, more than just a "there's some noise" which you then ignore and it stops doing you any good. I would really look into those. DIYEFI has some good info on some prospectives.

But it would be a really obvious thing to add. As would logging (or even monitoring) EGT's, esp for lean burn.

Joe Perez

iTrader: (8)

+e!

To be of value, a knock sensor interface needs to discriminate between noise that represents knock and the general noise that engine make all the time. Otherwise, you get false triggers all the time.

The KnockSenseMS is a partial answer, applying a notch filter to the analog signal to filter out-of-band noise. It still cannot discriminate between knock and in-band noise, so at sufficiently high RPMs it still false triggers.

There are indeed ASICs specifically for recognizing the character of knock. I've not done a lot of research, but check out the TPIC8101 for starters.

yertnamreg1218

I was looking for ASICs for this but couldn't find any. Thanks for the lead Joe.

I was reading on the delphi website that alot of OEM are going to a DSP based system that uses DFTs. I know some DSP stuff, but would rather not write a DSP knock conditioner.

I'll take a look at the various solutions out there an let ya'll know what I find.

AbeFM

iTrader: (3)

One of the best things I've seen, but probably beyond the scope of this board, is something that knows *where* the engine is, and listens only at the right time to pick up knock. So, noise outside of a certain window of crank rotation is ignored.

Also, you want to slew sensitivity with RPM - knock that can ruin a motor at 4k is quieter than engine opperating noises at 7k.