Page 2 of 2 FirstFirst 12
Results 16 to 28 of 28

Thread: Megaswitch v.megasquirttesting

  1. #16
    Super Moderator
    Join Date
    Mar 2011
    Location
    Camden, MI
    Age
    35
    Posts
    3,026
    the exact FET I was using was a STP65NF06, which is also a N-channel MOSFET and has a body diode the same way a IRFZ44 does. I thought the same thing until I burnt myself, did some google-fu and ended up here:

    http://electronics.stackexchange.com...-or-transistor

    the right diagram essentially has the body diode of the FET accounted for and is listed why it doesn't quite work(unless it's a zener). the STP unit's datasheet says the body diode functions as an avalanche, but I either did something very wrong or it isn't sufficient. the solenoid was also very non-linear in response, which is the opposite of what I expected. added in a 1N4007(as shown in the left diagram) I had from my diode selection and suddenly the solenoid operates with almost complete linearity from ~20% to 80% duty cycles and no more scorched fingers. without the diode, the solenoid was buzzing like crazy when commanding a duty cycle between the ranges of fully closed and fully open(which was something like 50 to 70%).

    something to test on the bench, if you really wanted to, I was much happier finding this out before calling a project finished.
    1995 Chevrolet Monte Carlo LS 3100 + 4T60E


  2. #17
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    Interesting, but I think the zener diode is making the valve more linear by changing the way the load is seen by the source. I'll have to think about that for a bit, maybe even do some testing A flyback diode is supposed to be invisible to the source and only be conduction when the coil is discharging immediately after current is taken away from the circuit .
    The man who says something is impossible, is usually interrupted by the man doing it.

  3. #18
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    So, since RS232 is pretty antiquated I wanted to get around the need to use a USB to Serial adapter, especially since every off the shelf adapter I've bought seems to either not work at all or be flakey.

    This is where that earlier modification of socketing the MAX232 IC (U6) comes into play. My plan was to replace U6 with a second DIP socket with the correct four pins jumpered to pass a TTL signal straight through from the db9 connector to the MS uC. I have an USB to TTL signal converter from Spark Fun, based on the FTDI FT232R IC.

    Initial testing of the idea:


    3 wire connection, ground, TX and RX:


    Test jumpers inside the MS:


    Replacement DIP socket to pass TTL coms straight through:


    Adapter DIP installed:


    While initial testing proved that it works for communications between the MS and the computer, there's an issue with early MSII daughter cards (which mine is) in that if there is no 5V present on the RX pin of the MS II daughter card at power up it goes into bootloader mode. This means that either the breakout board I have always has to be powered up, or a modification needs to be made to apply 5V to the RX pin when it's not powered up. Which is not a problem if I leave the breakout board in the MS all the time, but that was not the plan. I use this board for other devices as well, and I'm too cheap right now to buy another or several. lol So my idea of bypassing the MAX232 IC in the way I wanted was not going to happen. I came up with a few different ways to apply 5V to the RX pin without the breakboard installed, and then disabled once the breakout board was then plugged in, but this would be more proprietary than I wanted. ONLY my adapter would then work with the MS and that will usually not be a problem, but I didn't want to take any chances, so I went with an alternative, I created my own USB to RS232 adapter... well I've breadboarded it so far. I still need to make a more permanent version using some proto board.

    My adapter is the same USB to TTL breakout board I used earlier but I'm using another MAX232 IC to convert the TTL levels to RS232 levels that the MAX232 IC inside the MS can then convert back to TTL levels. yeah I know it seems odd to convert up just to convert back down again, but this way if my adapter isn't available for whatever reason and someone else's is, it will still work just as if it were normal. I also have a Blue Tooth TTL board that I used in a school project that I'd like to get working, but need to reprogram the baud rate before I can get that working and I need to look up again how to do that. lol

    the overall USB to RS232 adapter breadboarded and communicating:


    Just the guts of the main adapter circuit:


    I've had this circuit working on my bench for several hours how and it's worked flawless, zero drop outs with my main PC and only a couple drop outs with my laptop, but that's far less than with the USB to serial adapter I tested earlier today which is a Prolific based adapter, which are always flaky at best. I had drop outs every couple of minutes with the Prolific adapter, just monitoring basic coms.
    The man who says something is impossible, is usually interrupted by the man doing it.

  4. #19
    Fuel Injected!
    Join Date
    Apr 2013
    Posts
    498
    Yep, always use FTDI USB chip (which I believe this is what your breakout board is now ?!). Prolific drivers are not so great on Windows and even worse on Android :)

    You seems to be having fun with that MS Chris. Also looks like its getting closer and closer to make it to your car :D
    ALDLdroid author, Android application for ALDL ECU
    https://play.google.com/store/apps/d...roux.aldldroid

    Like us on Facebook

    Visit the app's Website

  5. #20
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    Yep, my experience is the same, FTDI or nothin' It just seems that people, myself included end up with Prolific junk. Oh well.

    I've just been playing with it on the bench to get used to the software, and figure out what makes the MS tick, and make sure everything works the way I expect.

    I've even upgraded the firmware. When I got this MSII it had FW version 3.0.3u, so, it's OLD, that FW is from 2009 if my findings are correct. I loaded the latest (why wouldn't I use the latest? lol) 3.4.0.

    I still need to register TunerStudio, I mostly want the custom dashes/gauge layout. I don't really care about auto tune, though I might as well use it with the registered version...
    The man who says something is impossible, is usually interrupted by the man doing it.

  6. #21
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    I figured I should update this a bit, since I have something to update.

    Well, the original plan for the MS was to install it on my LX9 in my Datsun, but since that's been broken since December with no sign of getting back together anytime soon, due to reasons, which are beyond my control right now, I installed the MS on my T-bucket (283+.030" V8) which I converted to EFI about a month ago using GM components, including a '7747 ECM. I used the Delco to start with because I knew that it should be close and could drive it on the stock tune, or tune it a bit better, and then switch to the MS. Both of my widebands were also loaned out in other people's vehicles at the time and didn't want to try tuning the MS blind, so another reason to start with the Delco at the time.

    Well there were mechanical issues that needed to be worked out and at this point should be pretty much solved. I got sick of the '7747 losing connection with the AutoProm and lack of reliable datalogging. There might be an issue with the ECM itself, but I was just sick of it. This was last Friday. I went home and started on an adapter to jumper the '7747 harness to the MS. I'll post pictures later.

    Sidenote: When I installed the Delco EFI I knew I'd be going to the MS shortly so I added wires at the time for the IAT and a couple extra grounds.

    On Sunday (I was lazy on Saturday lol) I finished the adapter harness and ran a vacuum tube for the MAP sensor that is built into the MS ECU. I was able to get the engine to start and run ok, in trying to tune the '7747 I discovered that increasing MAP AE worked very well to get rid of a lean pop through the intake with large throttle changes, so I started there by figuring out how to enable the MAP AE in MS. Turns out that there's a slider in the AE menu that controls the blend between TPS and MAP AE, and you simple slide it towards the MAP AE table...

    One thing was always a miss though, above about 3000 RPM, the revving became REALLY sluggish, and no matter what I tried with VE or spark nothing changed it considerably for the better.

    I checked base timing and decided to adjust it. I ended up with a weird thing where with the bypass connected the engine would not start, the MS was not getting or at least was not acknowledging any RPM (DRP) signal. Bypass disconnected and the engine would start and run the way that was expected.

    In this I decided to check whether the EST signal was good from the MS, which I used my JimStim and connected the EST signal to a spare LED on the JimStim board and observed the LED pulsing... for hours, in fact overnight the EST signal was still triggering in the morning... oopps... lol

    I also discovered that the MSExtra manual is wrong where the bypass control circuit is concerned. Using the board modifications and the external connections shown will have the timing control work backwards, as in ECU controlling timing during start, but not during run. *facepalm* This is why the engine seemed lazy above 3000 RPM, because it was running on base timing! In other documentation there is often a relay suggested to control the 5V signal on the bypass line, usually the relay is suggested to be controlled by the ignition switch, so that 5V is provided to the bypass wire only while running, so that base timing is still used at start. diyautotune.com suggests connecting the bypass wire straight to 5V and have the ECU control timing during start up.

    I went a different route. The built in settings for the HEI control will turn on the transistor that controls D16 (Q8) 5 seconds after the engine starts to run, or more specifically after the threshold that is set under "Cranking/Startup Settings" called "Cranking RPM" has been exceeded. IIRC stock setting is 700 RPM, I set this to 400 RPM. As I mentioned before the way the MSExtra manual suggests to connect it will not work correctly. I modified the circuit by installing a 2N3906 PNP transistor to provide the 5V on the bypass line that is needed for the ECU to control timing through the HEI module. I have not tested this on the engine yet, hopefully tomorrow I will be able to. On the bench it works it works perfect and I can't see the HEI module needing anywhere near the 200mA rating of the 2N3906 transistor.

    I will post pictures and hopefully a schematic tomorrow, once I get a chance to get them onto my computer.
    The man who says something is impossible, is usually interrupted by the man doing it.

  7. #22
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    Just to update this...

    I gave up on trying to get the 283 to run right. It was old and tired. I replaced the timing set, a friend of mine played with it for a bit, adjusting rocker arms because the lifters seemed to be pumped up and would not bleed back down making them act like mechanical (solid) lifters. At one point the engine was literally a V-twin, only cylinders 1 and 2 seemed to actually be firing. I gave up and left the car alone for a while.

    Last week I had holidays, and my Grandfather's truck had a good running 305 in it, that was destined to be replaced with a Vortec 5.7L that had been installed in the truck years ago but pulled due to a crack in the block for the starter bolt. That's fixed now and using a different starter now that uses different holes seems to have taken care of that issue. So I swapped the 305 out for the 5.7, and then took the 305 and swapped it in place of the 283 in the T-bucket.

    So now that it has a good running engine, I reinstalled the MS, and fired it up. It ran ok, then noticed that the bypass wire was still disconnected, I reconnected it and the engine stalled. Tried a few different things, stalled every time. Then used a meter to check the signal on the EST circuit and found there was no output, no wonder the engine would stall when it was looking for that output. Opened up the MS to find that the transistor driving that circuit had failed (small hole in it that had blown out). replaced that and the car runs well. Still needs some tuning but I have driven it everyday since Sunday, because I can and it's fun to do so now, unlike before when the 283 was in it and it would get worse and worse to drive as time went on.

    Couple pictures. The wiring will be cleaned up in the winter, it works for now, so it's good until then.


    Last edited by Six_Shooter; 09-25-2016 at 09:11 PM.
    The man who says something is impossible, is usually interrupted by the man doing it.

  8. #23
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    I've been playing with the T-Bucket and making some adjustments to the MS when I get the chance to.

    It's running well. I've added a lot of timing and the engine loves it. I can get away with this because the car is so light it doesn't really put much of a load on the engine. Still needs more though. It fires up every time the same way, little bit of an extended crank when cold, but then quick to fire on warm/hot restarts, runs as high as about 190 degrees F, when in stop and go or sitting and idling, cruising is generally down closer to 170ish.

    I've also been using my phone with MSDroid to make small adjustments and datalogs. It's been dropping out at certain points, usually just above 3000 RPM, but not every time. I have a couple ideas on things to try to get that to stop.

    So far the MS has been good. Unfortunately I can't really directly compare Delco to the MS in this application like I wanted to do when I started this thread since I never had the Delco running on this engine, well, I did, but for about 35 seconds just to get it to fire, since I knew the calibration would be close, and I didn't have one of my WBO2 sensors to install to use with the MS to get it tuned.

    But I can pretty much say that the outcome and use were pretty much what I expected going into this.

    Both systems have their pros and cons, I still like both systems for different reasons.

    The Delco I like the fact that it has a true RDF/LHM mode (Limp Home), that DOESN'T rely on the CPU being fully functional, or even a proper bin on the EPROM being present. The MS1 and MS2 don't have any sort of LHM, and the MS3 has a quasi LHM, that will allow substitute values if a sensor or multiple sensors fail, which is better than nothing at all, but still relies on the CPU to fully functional.

    The Delco is cheap to acquire, install and even tune, although like anything, you can throw good amounts of money at a Delco for tuning as well, such as using Emulators and other related bits.I also like the fact that you can buy a replacement Delco ECM pretty much anywhere, IF you ever have one fail. In my time of using Delcos I've only ever had 2 fail where it wasn't my own stupidity causing the failure, and even one of those failures, could be sorta blamed on me, when I had some rain water drip onto the ECM, where as if I had the ECM mounted where it should have been, it likely wouldn't have been an issue. The MS can be done inexpensively as well, especially if you buy used, but buying MS used is a bit of a gamble, due to the universality of the MS, it may need extensive recofiguing of the hardware to use on your application, and as long as you're proficient with reading schematics and soldering, that's a great way to go. The other issue is that someone who isn't that proficient may have caused damage to the MS ECU and now you need to deal with it or even find the problem.
    I've bought several MS1 and MS2 ECU and even a DIYpnp over the last year, the most I've spent on a unit has been about $170 USD, with shipping. The deals are out there. I've bought a couple locally, because again they were good deals. However this seems to be atypical, especially when it comes to the specific versions I have been buying. I just find them at the right time and jump on them. This again is where if you're proficient with reading schematics and soldering you can buy ECUs that are reported to be not working, damaged, etc and repair them. I bought one MS2 V3.0 because the guy bought a DIY kit, and wasn't very good with the soldering aspect, so he damaged the main board. He was upfront about this, but it came with a never been used MS2 daughter card. He never got to the part in the assembly where it gets plugged in before giving up and buying a pre-built unit. I figured at the very least I was spending just a little more than just buying the MS2 daughter card and it was worth it that way to upgrade one of my MS1 ECUs, even if the mainboard was a total loss. The board was pretty badly damaged in to spots, over current transistor for the fuel pump output and the MAP sensor pins. This worked out though since I want to install an electronic trans in one of my vehicles so I am re-configuring it for trans control use. I had planned to run that vehicle with '7749 and $59, but now I will likely also run it with MS, so that I can use the CANBUS between the UCU and the TCU to save on connections and be able to take advantage of some of the trans control features, I'll see when I get to that point.
    I bought a broken DIYpnp for about the price of shipping because that's all teh guy wanted. I haven't been able to verify everything that is wrong with it yet, but since I like to play with electronics I figured this was worth the price, especially since I've wasted more money on less fruitful en devours in the past. lol Since I can communicate with the CPU, I think I'm hoping that the I/O will all be good and hopefully just an external driver is bad. I need to set up a testing adapter to plug into my JimStim to feed it actual sensor signals though.

    Tuning:
    Tuning one is no easier than tuning the other IMO. Once the engine is up and running tuning is about the same between each, mostly working with the VE and spark tables, and making other minor adjustments to start up enrichments, AE, DE, etc are all very similar, and in some case just the same. The biggest difference to remember when tuning a Delco is that the VE numbers stop at 100 (99.8 actual IIRC), and any number above that gets truncated to the 100 value, where as the MS allows higher than 100 values in the VE table, I don't recall what the actual limit is, but I bet it's 254 (8 bit maximum). I know this max VE of 100 has caught some people in the Delco world a few times. Simple adjustment to the BPW, or the base calculation (differs in some Delco applications) and you can them get the values all below 100.

    So far I have not used any of the "extra" features that wouldn't be present on a Delco, no flat shit, or anti-lag, nitrous control, boost control, etc, since I have no turbo, nor nitrous and it's an automatic, so nothing to really control there. I will be playing with these features on the next MS equipped car, my Datsun, once I finsh assembling the engine which is nearly complete. I will be testing anti-lag, flat shift, etc

    So both systems are good, one requires more electronics knowledge, due to the needed internal modification (*except MS3Pro), where as the other requires more hardware knowledge to use in a non stock application. So pick your poison, and if you need some of the features that the MS has that the Delco doesn't, well then you'll need to look at another option, such as the MS3Pro, if you really don't want to get into soldering up circuits.

    Just to mention the MS3Pro, it has all of the features of the MS3 WITH the MS3X board added on, but in a moisture proof case with AMPseal connectors and nothing to modify on teh board to get any and all functions to work. What it does require though, is the ability to add on external circuits depending on what you are trying to run. Some injectors will require injector driver boxes (for Low-Z), or add on coil control boxes/ignitors, for "dumb" coils, relays for many of the simple on/off circuits. I've installed one of these on a friend's car that previously had a FAST EZ-EFI 2.0, and it runs SO much better with the MS3Pro, though we also did swap to MPFI at the same time, so I'm sure that was a factor as well. If I had the money I'd likely go MS3Pro, but I don't, so I'll use the DIY version, and that way I do get to play with the soldering iron and electronic bits I have on my bench which is fun for me.
    The man who says something is impossible, is usually interrupted by the man doing it.

  9. #24
    Fuel Injected! Dr_Grip's Avatar
    Join Date
    Apr 2015
    Location
    Germany
    Age
    44
    Posts
    116
    I have been told that the MS2/MS3 is much faster to react to changes in engine state (like throttle input) than a 7747 Delco unit, making throttle response almost as instant as on a carb. Can you confirm this?

  10. #25
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    I can't say that I've noticed any difference there.

    Honestly the processor speeds are much higher than need be for any of the three systems you mentioned won't affect throttle response. There are MANY, MANY other factors that will affect that more than processor speed.
    The man who says something is impossible, is usually interrupted by the man doing it.

  11. #26
    Fuel Injected! Dr_Grip's Avatar
    Join Date
    Apr 2015
    Location
    Germany
    Age
    44
    Posts
    116
    Quote Originally Posted by Six_Shooter View Post
    I can't say that I've noticed any difference there.

    Honestly the processor speeds are much higher than need be for any of the three systems you mentioned won't affect throttle response. There are MANY, MANY other factors that will affect that more than processor speed.
    OK, so I should rather focus on getting my AE settings right, I guess.

  12. #27
    Super Moderator Six_Shooter's Avatar
    Join Date
    Feb 2011
    Posts
    2,968
    Quote Originally Posted by Dr_Grip View Post
    OK, so I should rather focus on getting my AE settings right, I guess.
    Yes, AE, VE, timing, etc will all have a larger effect on throttle response than what system is used. :)
    The man who says something is impossible, is usually interrupted by the man doing it.

  13. #28
    Super Moderator
    Join Date
    Mar 2011
    Location
    Camden, MI
    Age
    35
    Posts
    3,026
    on a 7747, TPS is updated at 80Hz, correct? assuming the engine is running at 2000RPM and you jump from 10% throttle up to 100% in 1/10th of a second, the TPS is going to be read and acted upon 8 times in that scenario.
    the engine is going to complete 3.333 revolutions in that 1/10th of a second.
    the engine is going to make .417 revolutions between TPS updates at that speed.
    with a 6 cylinder, that's 1.25 ignition events between updates. with an 8 cylinder, it's 1.67.

    you would have to be able to feel 1 or 2 fuel/ignition events being less than optimal for there to be any sense of improvement. a hard miss is easy to detect, but a very mild misfire is awful difficult to detect.

    if the TPS read is done at 160Hz(some GM OBD1 ECMs do this), then even with a V8, you're at less than one fuel/ignition event before the TPS is read and accounted for.
    1995 Chevrolet Monte Carlo LS 3100 + 4T60E


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •