ECUs and Tuning Discuss Engine Management, Tuning, & Programming

Arduino as ECU?

Thread Tools
 
Search this Thread
 
Old 05-03-2018, 10:25 AM
  #401  
Senior Member
 
der_vierte's Avatar
 
Join Date: Mar 2017
Location: GER
Posts: 757
Total Cats: 113
Default

anyone got their speeduino from the last batch?

aussiepost is soooo slooooow they got my package on 26th April and it didn't even left the country....

edit:
went with "fast shipping" aswell
der_vierte is offline  
Old 05-03-2018, 07:17 PM
  #402  
Junior Member
 
noisymime's Avatar
 
Join Date: Jun 2015
Posts: 75
Total Cats: 11
Default

Originally Posted by der_vierte
anyone got their speeduino from the last batch?

aussiepost is soooo slooooow they got my package on 26th April and it didn't even left the country....

edit:
went with "fast shipping" aswell
Flick me a PM, I'll chase it up with them as that's not right at all. Even Aus Post normally aren't THAT bad.
noisymime is offline  
Old 05-03-2018, 09:26 PM
  #403  
All-round "Good Guy"
 
Lokiel's Avatar
 
Join Date: Dec 2009
Location: Brisbane, AUSTRALIA
Posts: 993
Total Cats: 245
Default

^
Australia Post sucks big-time!
More than once it's taken goods only a couple of days from being shipped in the US to arrive in Australia, only then to sit around for 3-4 days in an Australian warehouse before being delivered.
If your goods incur duty, add about another 10 days to that.
I was very proactive when I received my MS3-Pro, I tracked the package all the way and submitted all the forms electronically when it arrived without waiting to receive the hard-copies and it still took 5 days!

UPS or FedEx are MUCH better shipping options, even if they're more expensive.

USPS should be avoided at all costs since it's a crapshoot whether or not you'll receive your goods.
USPS lost 2 of my packages within the US (the last tracked positions were their arrivals in USPS facilities) and they simply washed their hands of them because they were destined for another country.
I can understand their position if the goods had arrived in Australia so were completely outside their control but both were lost in the US, inside USPS facilities so I'll never use them again.
Lokiel is offline  
Old 05-04-2018, 03:00 AM
  #404  
Senior Member
 
der_vierte's Avatar
 
Join Date: Mar 2017
Location: GER
Posts: 757
Total Cats: 113
Default

Originally Posted by noisymime
Flick me a PM, I'll chase it up with them as that's not right at all. Even Aus Post normally aren't THAT bad.
magically it arrived a few minutes ago in frankfurt/germany. i have to apologize to auspost
der_vierte is offline  
Old 05-08-2018, 03:01 PM
  #405  
Senior Member
iTrader: (1)
 
huesmann's Avatar
 
Join Date: Jul 2014
Location: Silver Spring, MD
Posts: 659
Total Cats: -16
Default

Mine arrived (East Coast USA) yesterday, ordered April 24. On to VTPS issues.
huesmann is offline  
Old 05-20-2018, 07:59 PM
  #406  
Newb
 
danielvolt's Avatar
 
Join Date: May 2018
Posts: 3
Total Cats: 0
Default

Originally Posted by noisymime
The NA8s with the 48-pin ECU will work fine with the current one.

The 64-pin requires a bit more work and unfortunately will have to cost a bit more. The size of the 64-pin connector means it has to go up to a PCB >100mm, which means it goes into the next price bracket. It also means a larger case, which will cost slightly more again.
I have started the work on these, but not sure when they'll be ready yet. I will need testers though if you're interested...?
Are you sure the NA8 with 48 pin ECU will work? My 94 1.8 is a 48 pin and if you confirm it works i will order one straight away.
danielvolt is offline  
Old 05-21-2018, 04:35 PM
  #407  
Senior Member
iTrader: (1)
 
poobs's Avatar
 
Join Date: Apr 2008
Location: Florida
Posts: 810
Total Cats: -6
Question Teach kid to tune

Hello Experts !!!

Well, Here's my story....

I had a MS years back and sold it.

My son is now older, loves cars and somewhat trying to learn from me.

I'm looking to the Speeduino PnP for my 91 as an inexpensive way to teach him what I know about aftermarket ECUs and tuning them.

Am I on the right track ????

Thank you very much in advance for your feed- back and information.
poobs is offline  
Old 05-22-2018, 01:17 AM
  #408  
Junior Member
 
noisymime's Avatar
 
Join Date: Jun 2015
Posts: 75
Total Cats: 11
Default

Originally Posted by danielvolt
Are you sure the NA8 with 48 pin ECU will work? My 94 1.8 is a 48 pin and if you confirm it works i will order one straight away.
Yep it should. The only variation I've seen so far that currently doesn't work out of the box is the '95-'98 Euro 1.6s, which require 1 jumper wire to be added (There's a post on it slightly higher in this thread).

I've seen 2x 1.8s with the 48-pin plugs up and running on these now.
noisymime is offline  
Old 05-23-2018, 01:22 PM
  #409  
Senior Member
iTrader: (1)
 
poobs's Avatar
 
Join Date: Apr 2008
Location: Florida
Posts: 810
Total Cats: -6
Thumbs up

Originally Posted by noisymime
Yep it should. The only variation I've seen so far that currently doesn't work out of the box is the '95-'98 Euro 1.6s, which require 1 jumper wire to be added (There's a post on it slightly higher in this thread).

I've seen 2x 1.8s with the 48-pin plugs up and running on these now.

Thank you very much !! I'm trying to make time to read this entire thread.
poobs is offline  
Old 05-24-2018, 03:36 PM
  #410  
Senior Member
 
themonkeyman's Avatar
 
Join Date: Nov 2016
Location: Los Angeles, CA
Posts: 552
Total Cats: 75
Default

So, very keen on getting/building one of these. What would be needed to run true sequential ignition and injection on a US market 1.6? I'm assuming a different (individual) coil setup and adding wires to break the injectors out of pairs? Also, it seems the current setup uses a PCB-mounted MAP sensor with a vac line running to it, is it possible to use a manifold-mounted MAP sensor (e.g. late-model GM maybe?) and just add wires/pins? Forgive any ignorance I may have, still new to the ECU game.

Also, will this run fine off the 1.6 CAS or is it recommended to add a crank reluctor? Will the arduino want/accept independent crank and cam position signals? thanks!
themonkeyman is offline  
Old 05-24-2018, 09:45 PM
  #411  
Junior Member
 
noisymime's Avatar
 
Join Date: Jun 2015
Posts: 75
Total Cats: 11
Default

Originally Posted by themonkeyman
So, very keen on getting/building one of these. What would be needed to run true sequential ignition and injection on a US market 1.6? I'm assuming a different (individual) coil setup and adding wires to break the injectors out of pairs? Also, it seems the current setup uses a PCB-mounted MAP sensor with a vac line running to it, is it possible to use a manifold-mounted MAP sensor (e.g. late-model GM maybe?) and just add wires/pins?
You certainly could, but why would you bother? Running a vacuum line is very simple, much simpler than mounting a new sensor and running the 3 wires for it IMHO. There's also more risk of noise on the signal lines with an under bonnet MAP. Was there a reason you want to go that way?

Also, will this run fine off the 1.6 CAS or is it recommended to add a crank reluctor? Will the arduino want/accept independent crank and cam position signals? thanks!
It runs fine with the stock CAS, but that CAS obviously has limitations. It's low resolution and it suffers from any wear in the valve train components etc, which are just inherent problems with that setup. Using the Yoshifab 12-1 trigger disk is one option that a few people have gone with and seen good results. Moving to something like a 36-1 wheel on the crank is the best option for accuracy, but is a considerable amount of work.
noisymime is offline  
Old 05-24-2018, 10:06 PM
  #412  
Newb
 
Mayjaplaya's Avatar
 
Join Date: May 2018
Location: Vancouver, BC
Posts: 2
Total Cats: 0
Default

Originally Posted by noisymime
You certainly could, but why would you bother? Running a vacuum line is very simple, much simpler than mounting a new sensor and running the 3 wires for it IMHO. There's also more risk of noise on the signal lines with an under bonnet MAP. Was there a reason you want to go that way?
Hmm, I read on https://speeduino.com/wiki/index.php/MX5_PNP that an IAT was needed if I were to delete the AFM. I plan to delete the AFM.
Mayjaplaya is offline  
Old 05-24-2018, 11:23 PM
  #413  
Junior Member
 
noisymime's Avatar
 
Join Date: Jun 2015
Posts: 75
Total Cats: 11
Default

Originally Posted by Mayjaplaya
Hmm, I read on https://speeduino.com/wiki/index.php/MX5_PNP that an IAT was needed if I were to delete the AFM. I plan to delete the AFM.
Adding the IAT is fairly simple, but doesn't require you to add a separate MAP sensor as well. Most people go with the standard GM IAT sensor as its cheap (I sell them for $22 with a wiring pigtail) and easy to install. You can just about get away without one if you're in a place where the temp doesn't vary too much and you're NA, but it's recommended to have one.
noisymime is offline  
Old 05-25-2018, 01:14 AM
  #414  
Newb
 
Mayjaplaya's Avatar
 
Join Date: May 2018
Location: Vancouver, BC
Posts: 2
Total Cats: 0
Default

Originally Posted by noisymime
Adding the IAT is fairly simple, but doesn't require you to add a separate MAP sensor as well. Most people go with the standard GM IAT sensor as its cheap (I sell them for $22 with a wiring pigtail) and easy to install. You can just about get away without one if you're in a place where the temp doesn't vary too much and you're NA, but it's recommended to have one.
I see. I misunderstood the other poster's question then.

I looked at the Speeduino shop (that's you, right?) but it said "This product is no longer in stock" for the GM IAT sensor so I bought an identical sensor with pigtail from diyautotune for $.49 more.
Mayjaplaya is offline  
Old 05-25-2018, 09:22 AM
  #415  
Senior Member
 
themonkeyman's Avatar
 
Join Date: Nov 2016
Location: Los Angeles, CA
Posts: 552
Total Cats: 75
Default

Originally Posted by noisymime
You certainly could, but why would you bother? Running a vacuum line is very simple, much simpler than mounting a new sensor and running the 3 wires for it IMHO. There's also more risk of noise on the signal lines with an under bonnet MAP. Was there a reason you want to go that way?
Ah, interesting. If anything I would've assumed it would get a cleaner/more sensitive MAP reading with the sensor directly in the manifold instead of reading via a meter+ of vac line. I feel like a shielded, twisted set of signal wires would yield a fairly clean signal, no?


Originally Posted by noisymime
It runs fine with the stock CAS, but that CAS obviously has limitations. It's low resolution and it suffers from any wear in the valve train components etc, which are just inherent problems with that setup. Using the Yoshifab 12-1 trigger disk is one option that a few people have gone with and seen good results. Moving to something like a 36-1 wheel on the crank is the best option for accuracy, but is a considerable amount of work.
Ok, good to know. Were I to run two seperate cam and crank signals, will the arduino read both? Is that just overkill? Of the two, I assume crank would be preferred to eliminate any timing belt slop?

Does the 48-pin connector have enough spare pins to run true sequential spark and fuel? I figure I can ditch the CEL and the two self diagnose pins, just would need one more.

Thank you for answering my (no-doubt) pedestrian questions. Very excited about this option.
themonkeyman is offline  
Old 05-25-2018, 01:21 PM
  #416  
Senior Member
iTrader: (1)
 
poobs's Avatar
 
Join Date: Apr 2008
Location: Florida
Posts: 810
Total Cats: -6
Default

How's the idle quality on this ECU ?
poobs is offline  
Old 05-26-2018, 07:49 AM
  #417  
Junior Member
 
lsdlsd88's Avatar
 
Join Date: Aug 2017
Posts: 77
Total Cats: -1
Default

Originally Posted by themonkeyman
Ah, interesting. If anything I would've assumed it would get a cleaner/more sensitive MAP reading with the sensor directly in the manifold instead of reading via a meter+ of vac line.
Yes, you get a more sensitive map reading and then you NEED to filter it in the software to hide the pulsation from intake valve opening events, adding filtering to a signal adds delay, and so on...

shielding low voltage signals from all the EM noise in the engine bay from coils, alternator, sometimes is really hard work (difficult to detect and can vary with ambient conditions), the less you depend on those signals the better it is.

found this also:
https://www.miataturbo.net/megasquir...ecu-box-89909/

Originally Posted by poobs
How's the idle quality on this ECU ?
I only used open loop idle control and no issues, I can't really feel any difference from the stock ecu.

Last edited by lsdlsd88; 05-26-2018 at 08:01 AM.
lsdlsd88 is offline  
Old 05-29-2018, 11:43 AM
  #418  
Newb
 
danielvolt's Avatar
 
Join Date: May 2018
Posts: 3
Total Cats: 0
Default

Originally Posted by noisymime
Yep it should. The only variation I've seen so far that currently doesn't work out of the box is the '95-'98 Euro 1.6s, which require 1 jumper wire to be added (There's a post on it slightly higher in this thread).

I've seen 2x 1.8s with the 48-pin plugs up and running on these now.
Sick!!! Do you also provide a 1.8 Base Tune? I have seen a few people that struggled to get their 1.8 started.

Going to order the ECU in a few weeks, so excited
danielvolt is offline  
Old 05-29-2018, 08:18 PM
  #419  
Junior Member
 
noisymime's Avatar
 
Join Date: Jun 2015
Posts: 75
Total Cats: 11
Default

Originally Posted by themonkeyman
Ok, good to know. Were I to run two seperate cam and crank signals, will the arduino read both? Is that just overkill? Of the two, I assume crank would be preferred to eliminate any timing belt slop?
In my opinion, the best setup is something like a 36-1 wheel on the crank and a single cam pulse. That gives you everything fast sync for starting, full sequential and good timing accuracy. It is a pain to get a 36-1 wheel onto the crank on these engines though, so I haven't seen a lot of MX5s/Miatas go that route (It's a VERY common setup on other vehicles though and very well tested with Speeduino).

Does the 48-pin connector have enough spare pins to run true sequential spark and fuel? I figure I can ditch the CEL and the two self diagnose pins, just would need one more.
It does for fuel (The California models run sequential as standard) and these boards have the drivers there already for it. You just need to cut 2 traces, which are designed for this.
Ignition is trickier though as there wasn't ever a version of these that had 4 drivers from the factory. These PNP Speeduino boards only have 2 ignition drivers on them, so they're limited to wasted spark, which is what the car is wired for. The generic v0.4 Speeduinos have 4 ignition outputs, but aren't as straightforward to wire in as the PNPs.
noisymime is offline  
Old 05-29-2018, 08:30 PM
  #420  
Junior Member
 
noisymime's Avatar
 
Join Date: Jun 2015
Posts: 75
Total Cats: 11
Default

Originally Posted by lsdlsd88
I only used open loop idle control and no issues, I can't really feel any difference from the stock ecu.
In my experiences I've found that 90%+ of people will be totally fine with open loop. It's what I run on my car and what the original ECU uses as well, so it should be enough. It's very easy to tune and typically delivers good results without much effort. The one thing missing up until now has been a 'high idle' input that can be used to compensate for things like the AC being turned on, but this has been added in this months firmware

Keep in mind that the IAC in these cars defaults to being partially open as a safety fallback position. So even if you turn idle control off completely, you tend to still get a good idle, just slightly high when it's warmed up.
noisymime is offline  


Quick Reply: Arduino as ECU?



All times are GMT -4. The time now is 03:22 PM.