240_to_miata

iTrader: (2)

OK so I have seen in a few threads that recently (last year or so) Brain has been saying he prefers the VR input over the 4g63 input.

What is the benefit?

I built my megasquirt about 5 years ago following the standard https://www.miataturbo.net/useful-sa...asquirt-13676/

I currently have the standard 4g63 mod we all did for years. I have always been plagued with the cruise studder when I loose sync. Would the VR help me?

What is needed other than swapping the TACHSELECT and TSEL jumpers?
I am not sure if I installed the other components required for the VR. I will have to sort thru the BOM in MegaSquirt - Electronic Fuel Injection Computer by Bowling and Grippo ©2010


Current setup:
96 miata
MS1 V3.0 board
Hi-Res code
.1uf cap mod installed

Thanks in advance.

duffbuster243

iTrader: (1)

I too was plagued by the cruise studder. I upgraded to MS2 (On the same board) last week and so far have not had this happen again. I only have about 30 min of driving time on it but I would usually have a couple of occurrences at that point.

Braineack

iTrader: (62)

WTF is cruise studder?

duffbuster243

iTrader: (1)

A sync loss that tends to happen during a steady cruise on the highway. feels like a miss-fire or studder. Shows up on my datalogs as a spike in rpm sometimes 10k+.

Braineack

iTrader: (62)

never heard of it. never experienced it. shouldnt be losing syncing on a simple CAS.


are you connected to the laptop when this happens? have you noticed it happens when your laptop goes to "sleep" mode?

duffbuster243

iTrader: (1)

I can't speak for the OP but I've had the issue both with and without a laptop connected.

Like I said though, I upgraded to MS2 and switched to Joe P's updated ignition output. It seems to have solved the issue for me as of right now. Maybe I just had a bad solder connection on the board itself which was fixed when switching things around for the MS2 upgrade.

Braineack

iTrader: (62)

possibly since you had to alter the cas inputs for the ms2.

240_to_miata

iTrader: (2)

It happens with or without a laptop connection. Many people with MS1 have had this. The cap helps. I'll find threads as examples when I am not on mobile

Braineack

iTrader: (62)

the cap is nessecary with the high res code.

but this "cruise studder" is not a common thing; this is the first I've heard of the term.

...and I say this not only as someone who's built probably 100 plus msI units begining in 2007, but as a client.

240_to_miata

iTrader: (2)

Ok anyway ...Back to the original question... Is VR. better? Why?

Braineack

iTrader: (62)

doesnt fix the issue that it's more than likely the second trigger that's the issue, not the cmk signal through the opto circuit.

karter74

iTrader: (4)

My fathers car suffers from the same studder condition you mention. It's quite random and can happen when cruising, or under WOT. Quite annoying.

I figured it was a poopy CAS, but I'm currently waiting on my Reverent built MS3 to come in. If it still happens, I think that will solidify something else being the issue.

Braineack

iTrader: (62)

do this instead:






you guys need to log this crap, then post it with the spot noted that you notice a "Studdah", then we can actually figure out exactly what's happening and how to solve it.

I'm doubting it's lost sync and more like TPS/MAP enrichments false triggering or somethign else that's silly and easily solvable and doesn't require it's own underground code name that I have never even heard in the last 5 years of solving people's MS issues.

240_to_miata

iTrader: (2)

Thanks ... I have a log somewhere in another thread..just a screen shot. I will try to find it now. I have switched computers too many times since then so I doubt I actually have the logs.

The issue has been brought up on and off but people seem to ignore their issue or replace their CAS or just upgrade from MS1. I want an enhanced MS2 but its not a purchase I will be making any time soon.

Fireindc

iTrader: (13)

I had a studder with 12krpm spikes for a bit on my ms1 when i first installed it. I got it on a datalog and made a thread here, one of the OG's here told me to add more grounds. So I ran another ground and viola! no more "cruise studder".

lol

240_to_miata

iTrader: (2)

Old Screenshot:



Back when I was parallel...
https://www.miataturbo.net/megasquir...-please-37671/
Others:
https://www.miataturbo.net/megasquir...-issues-43989/
https://www.miataturbo.net/megasquir...hen-hot-46772/
https://www.miataturbo.net/megasquir...rpm-dip-60289/

240_to_miata

iTrader: (2)

I did eventually add grounds direct from battery --> megasquirt ground location behind pass seat --> rear of engine --> front of engine.

It did help, but did not solve.

I only brought up this issue because I am opening up my megasquirt to add an input this winter while the car is still in storage. I figured I could kill 2 birds with one stone.

Joe Perez

iTrader: (8)

Used to happen to me with the MS1 and stock NA CAS on my '92. Just a very occasional dropout / sync loss while driving.

On that car, I solved it by drop-kicking the CAS into the trash and fabbing a crankwheel. That was the hard way.

These days, I solve it by building a proper input circuit.

No.

Both designs are horrible. The "VR" circuit is intended to be used with a VR sensor, and the other circuit is intended to be used with friggin' distributor points. And what's worse is that neither one properly addresses the "second" (cam) input, which someone years ago decided to just tie directly to the CPU and then forgot about.

Pretty much everything that has ever been written about building input circuits for Miata-style sensors onto the Rev 3.0 / 3.57 MS circuit boards is sub-optimal. If you want reliable triggering with no fuss, you need to use an IC-based circuit that's got some hysteresis in it. I first started with the MAX9926, which is intended for use with VR sensors but also works swimmingly well with open-collector sensors, and lately I've been using the CD40106, which is a much simpler device that's not compatible with VR sensors.

Either one is a good building block for putting together a "does not suck" input circuit. DIY has parted ways with the old designs, and now uses the MAX9926 in their MS3-Pro. The 40106 is easier to work with (physically) since it's available in a through-hole package rather than just a tiny little surface-mount.

There's more to it than just selecting a chip, of course. It's one part of a complete design that includes filtering, pullups, etc. Here's my version of a simple, suck-free input circuit for the MS1:

240_to_miata

iTrader: (2)

As usual, Thanks joe.

Dumb question time:

-CKP CMP, which one is second trigger? from the schematic I assume CMP is second.
- I pick up the first trigger from TACH SELECTright?
-IRQ-1 goes to TSEL right?

Joe Perez

iTrader: (8)

CMP is the second trigger. (Shorthand: CKP = CranKshaft Position, CMP = CaMshaft position.)

You can pick up CKP from TachSelect if you wish. That is where the TachIn pin from the DB-37 connector (pin 24) winds up.

You can run the line labeled "IRQ-1" in my schematic to the "Tsel" pin if you wish. That will get it into the CPU in the appropriate place.

If you build and connect the circuit as I have described, it is important that neither OptoIn nor VrIn be jumped to TachSelect, and that neither OptoOut nor VrOut nor VrOutInv be connected to Tsel. If you wish, you may omit all components on page 3 of the Rev 3.0 / 3.57 schematic.


The second input (CMP) can come from whichever DB-37 pad you elect to use. The common convention among MS1ers is to use pin 25, which is internally found at the IAC1A pad. This is a purely arbitrary choice.

When doing this, you will ignore all other instructions relevant to sensor capture, including the infamous (and nearly worthless) "cap mod."



By design, my circuit has a very long time-constant, which is part of the reason it's so stable. If you build this circuit, it is important that you properly calibrate the MS's "hardware latency" in Megatune / TunerStudio. Don't be surprised if it's well over 100 µs.


Also, the stability of C120 and C125 are highly critical. You should use as high-tolerance parts as you can find here.