Old school pontiac-Noise or crank signal issue? SOLVED!

[billr]

Yes, with the sensor disconnected you should be able to measure the internal pull-up between the signal (WHT) wire and the power (RED) wire. An external 1K resistor in parallel probably won't hurt anything, but it won't really help either. I wouldn't do it, as that would require the sensor to be sinking 24mA of current, more than the 20mA DIY shows for its max. That 10K resistor that is needed with some VRs is because a VR output increases with speed, some get quite high (~100V) and can "swamp" the MS input. That's probably not a perfect explanation, but point is: you don't need it with the Hall that has a discrete 0-5V output. Have you looked into cheap "scope" alternatives? Using an old sound-card is one, and I understand there are some pretty USB-based units that are reasonable. I am still unclear on the VR1 red and black wires, are those labels on a DIY cable, or pin names for the MicroSquirt? I'll take a peek at the MicroSquirt manual, maybe that will get me oriented better...

Edit: I don't know about the Micro, I only have experience with the MS3 V3.0, but I have found it to be very immune to "noise". Because my system has evolved over many years, stuff "scabbed on" as I went, I have broken just about every good wiring practice there is. I have wiring running intimately in-and-around ignition wires and coils. My battery is in the rear with only a short cable, the rest of grounding is through the frame. Things like that. I'm not recommending such a cavalier approach, but it is working fine and I don't think noise is really your root problem... unless the Micro is less immune to noise.

[opnwide]

Here is a link to the microsquirt v3 ignition sensor inputs. It talks about the internal circuit of the microsquirt and how it is made to directly work with hall effect crank triggers. It also talks about some differences between the microsquirt v3 and earlier microsquirts.

http://www.useasydocs.com/theory/vr.htm

[billr]

Okay, thanks for the link; it saved me some searching. However, I did do some additional searching and have come up with a contradiction that must be discussed! I may be getting myself confused, but it also may be a huge clue. You have previously stated that you are connecting the Hall signal to VR- and leaving VR+ unconnected, and the manual linked-to at the top of this page specifies that as correct. However, the link you provided in the recent post specifies connecting the Hall signal to VR+ and leaving VR- unconnected. I really don't know which is correct, but at least I now know what VR1 and VR2 and the +/- mean. You might want to try swapping the Hall signal from VR- to VR+.

[prof315]

Okay, thanks for the link; it saved me some searching. However, I did do some additional searching and have come up with a contradiction that must be discussed! I may be getting myself confused, but it also may be a huge clue. You have previously stated that you are connecting the Hall signal to VR- and leaving VR+ unconnected, and the manual linked-to at the top of this page specifies that as correct. However, the link you provided in the recent post specifies connecting the Hall signal to VR+ and leaving VR- unconnected. I really don't know which is correct, but at least I now know what VR1 and VR2 and the +/- mean. You might want to try swapping the Hall signal from VR- to VR+.

I agree with Bill on this. Every hall effect or optical sensor I have ever connected to the VR inputs on ANY MS product (all of my installs that use a hall effect for ckp/cmp in other words) I have hooked up to VR+.

[racingmini_mtl]

The Microsquirt V3 labeling for the VR inputs is done to try and keep some backwards compatibility with the older version. However, that is the opposite of how the VR conditioning chip is actually wired on the board. So when you use the wire labeled as VR- you actual use the chip VR+ input. That is why the MS2Extra doc recommends using the VR- wire.

Having said that, the chip can deal with either inputs when using a Hall or optical sensor as long as you leave the other input unconnected. The difference is the polarity of the signal. So if the correct documentation (the one on msextra not the "useasydocs" one) tell you to use VR-, use VR-. The code may expect a certain polarity to work correctly in certain wheel modes.

Jean

[LAV1000]

Did see the pictures on the sensor set up.
You need to make a strain relief on those wires leaving the sensor.
If you keep it this way wires are gone break just at the point where they leave the sensor.
Break (copper core) happens inside the isolation, so no visual indication.

This may be causing the issue right now.
Simple test.
Run engine at idle and wiggle those wires.

[opnwide]

Lav1000,

This is a brand new sensor, so I doubt there are any broken wires at this stage of the game. Once I get a solid crank signal, I will slowly refine my wiring harness for daily driving.

I did try the vr+ wire, and it didn't work so well. I had to adjust my timing then I gave it a try. Same breakup in crank signal as before. The trigger logs actually looks worse. So I put vr- back into play.

I am about to give up on this hall sensor and put a VR sensor in there. Any suggestions which to use with my 36-1 wheel? I was thinking of an old 5.0 mustang sensor with a built-in pigtail so I don't need to buy a plug.

toothlogger5.csv

[opnwide]

How can I test the hall sensor? I was going to check resistance to make sure that it had an internal pullup like what DIY Autotune was saying. Unexpected results:

From my power to my signal wire I am recording 210,000 ohms resistance! I thought it was an open circuit at first as I had my voltometer set on the 2000ohm scale. When I changed the scale, I was recording a tad over 200K ohmns.
The ground to signal wire was 14,000 ohms. and the power to ground wire was 205,000 ohms.

When I touch the sensor to a pair of vicegrips, the readings do not change. I thought it would.

thoughts?

[billr]

Easiest way is to apply 12V across the power/ground wires and see if the output toggles from 0 to 12V as you wave the steel in front of the sensor. That 200K does sound kind of high for a pull-up, but wait until tomorrow morning for Matt C. to comment.

Personally, I like VR for the crank, but Hall should not be giving you such grief. Either your brand-new sensor is defective, or there is some problem with your install that we are still missing. Point is, giving up on Hall is not necessary.

[opnwide]

Bill, when I hook up the sensor to a remote battery, the signal voltage does toggle as I wave it past a pair of vice grips. I'm not sure that proves anything in a real world application as the rpms increase.

UPDATE
I did spend half the night checking the runout on my toothed wheel. It started at .6mm and I got it down to .2mm, which I think is pretty good. Apparently, I never checked the runout on my 40yo pontiac pulley before I mounted the toothed wheel. Who would have thought that GM's quality control was a tad off in the mid '70s?

I also rerouted my VR1 and signal ground wire through a hole (that I had to drill) in my passenger side firewall. I kept both wires as far from sources of interference as possible. I kept the 12V power from the ECU's relay.

I also left the lower pulley on the concrete and tried to see how it would do without the alternator. Well, the voltage was dropping quick as the car was warming up, so I had to connect my Schumaker battery charger before I dipped below 12V. The charger was able to keep up with the ignition system no problem.

Ignition still cuts out around 2200rpm! I'm wondering if its my MSD 6A or something else, but the tooth logger keeps telling me that I have a problem with my signal, not a problem with my ignition system. Ignition system seemed fine on fuel only.

Again, any recommendations on sourcing a VR sensor? I'm tired of this Hall sensor.

greg

[billr]

Take a peek at Standard Motor Products #PC134. It's $21 from Rockauto and fits a lot of GM stuff from '87 to '07, should be available for a long time to come.

[krisr]

Can you bypass the msd and fire the coil directly? I.e do you have a coil driver in your ms? If it's a blaster coil it should fire with about 3.0ms of dwell but don't try fire a hvc coil or something like that!

[opnwide]

kris,
It doesn't look like the microsquirt can drive my coil. It only has logic circuits, no coil drivers.

[billr]

You can buy the parts to make a coil driver for less than $10.

[krisr]

Or see if you can snag a single ls1 coil from a wrecking yard or ebay. If you were in Sydney I've got 8 spares so could have lent you one. They have built in igniters.

[billr]

That's an interesting suggestion. Is that LS coil robust enough for continuous "single coil" use, or just a temporary way to get the MSD system out of the picture?

[opnwide]

Bill,
thanks for the recommendation for PC134. During my ebay search, I found a pc2 that looks like it should bolt into my existing hall sensor bracket with min modifications, so I ordered that one (cost was only $6!).

I have considered going to wasted spark, but I don't know what parts I need to make the microsquirt work with it.


Question: How sensitive is the WLED signal (that is going to my MSD 6A box) to noise? I currently have it in the loom with all my power wires. I'm wondering if the WLED is a low voltage signal (similar to sensor signals) that may be getting induced. I think I'm going to pull that wire out and see if it makes any difference.

[billr]

What is the OEM application for the PC2? I couldn't find that (the application), sometimes it is worth a few bucks more to design-in a part that you are confident will be available in the future.

[opnwide]

PC2 is supposed to be for 87-92 GM vehicles.
and a replacement for the pc134 for later 2000-2007 vehicles. and the pc178 as well. Not sure how directly they cross-over but they do look the same.

The pc2 seems to come in 2 versions- one with a plastic mounting tab similar to my hall sensor, and a second with a 90* connector and a metal tab. I obviously chose the former made by Standard Motor.

I hope this works. I stumbled across an old and long 2013 thread regarding wiring an LS5.3 to a V3microsquirt using a hall sensor, and they used a 1000Ohm resistor, hooking the signal wire to the VR1+ (not -). That goes against everything else I have read. http://www.msextra.com/forums/viewtopic ... x&start=60

[krisr]

Hmmm looking at your MSQ under Ignition settings, have you tried turning noise filtering on and tach interrupt masking off? Just curious if something funky is going on there especially as you are able to repeat the misfire.

On my MS3X with v3.0 board, I have noise filtering turned on for crank and cam signals and am running a VR magnetic sensor on a 36-1 tooth wheel. I dont have any issues at all and i've buzzed my 400 to 6800 on plenty of occasions.