Hi-Res code = ohmymotherfuckinggodamazing
#21
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,046
Total Cats: 6,607
As compared to Hi-Res on a 3.0 board? No dedicated ground plane, thinner traces, no proto area, no overcurrent protection...
Oh, you mean disadvantages to HR as compared to LR assuming both are on a 2.2 board. None. The only functional difference between HR and regular SnSExtra is the injector PW resolution and lack of Lo-z drive capability.
Oh, you mean disadvantages to HR as compared to LR assuming both are on a 2.2 board. None. The only functional difference between HR and regular SnSExtra is the injector PW resolution and lack of Lo-z drive capability.
#22
Elite Member
iTrader: (11)
Join Date: Jun 2007
Location: Overland Park, Kansas
Posts: 5,360
Total Cats: 43
As compared to Hi-Res on a 3.0 board? No dedicated ground plane, thinner traces, no proto area, no overcurrent protection...
Oh, you mean disadvantages to HR as compared to LR assuming both are on a 2.2 board. None. The only functional difference between HR and regular SnSExtra is the injector PW resolution and lack of Lo-z drive capability.
Oh, you mean disadvantages to HR as compared to LR assuming both are on a 2.2 board. None. The only functional difference between HR and regular SnSExtra is the injector PW resolution and lack of Lo-z drive capability.
So even with a flyback board with HR you lose the ability to use low-z injectors? Good deal, I'll look into giving that a shot.
#23
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,046
Total Cats: 6,607
What the HR code does is to increase the resolution of the time calculation on the injector drive. It does this by repurposing one of the CPUs high-precision hardware timers to do the fuel calculation. In the base code, this timer was used to do the PWM calculation for the injector current-limiting, and the fuel calculation was done with a lower-precision general purpose timer.
The Flyback board is nothing more than a couple of transistors which, in the 3.0 schematic, are already on the main board. The overall design of the circuit is identical between the two, only the location of the parts changes. Be it a 3.0 or a 2.2 with Flyback, the same CPU resources are used to control the injector drivers. If those resources are unavailable, the drivers cannot function in PWM mode. Also, understand that the "Flyback circuit" is purely a damper circuit that is needed to protect the main drivers when they are operated in PWM mode. An unmodified 2.2 board could be run in PWM mode with lo-z injectors, but the drivers would probably be damaged.
However, there is a board which replaces the injector drivers altogether and does the Peak & Hold current limiting in hardware, without intervention from the CPU. It will thus work with HR code. It's made by the same folks that brought us the JimStim: http://www.jbperf.com/p&h_board/index.html
They're using an LM1949 IC, which is a special purpose IC designed specifically for doing low-impedance injector control. It's a beautiful design, you just feed it the regular injector signal, and it takes care of doing the current ramp-down after the initial opening period.
I think Abe is planning to try it as he's been having electrical issues with his on-board injector drivers.
#24
Junior Member
Join Date: Feb 2007
Location: 11368 miles from where i would like to be
Posts: 269
Total Cats: 92
Actually (from the horses mouth) it's "about 0.02 - 0.03" ms (20 - 30us) fuel resolution. MS2 does it to 0.6666us in hardware and if you look at the comparison chart between different ms variants here :
http://www.msextra.com/ms2extra/MS2-...l.htm#features
It still says 100us* where * = "hires can do better than this, but we aren't specifying how much"
I just thought it was unfair on MS2 to claim ms1e hires can match it for fuel accuracy. It simply can't. Light years ahead of ms1 ah la plain though, so your excitement is justified :-)
It's worth noting that the software accuracy behind the hardware output makes a big difference too. Case in point, ms2 b&g vs. ms2e, ms2e has extensively rewritten math to make use of the hardware accuracy that ms2 b&g just doesn't have.
Fred.
http://www.msextra.com/ms2extra/MS2-...l.htm#features
It still says 100us* where * = "hires can do better than this, but we aren't specifying how much"
I just thought it was unfair on MS2 to claim ms1e hires can match it for fuel accuracy. It simply can't. Light years ahead of ms1 ah la plain though, so your excitement is justified :-)
It's worth noting that the software accuracy behind the hardware output makes a big difference too. Case in point, ms2 b&g vs. ms2e, ms2e has extensively rewritten math to make use of the hardware accuracy that ms2 b&g just doesn't have.
Fred.
#25
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,046
Total Cats: 6,607
Exact specs are hard to come by, but some numbers seem to show up more often than others.
To quote Phil, James & Co. from the MS1E-HR manual:
The MS1 uP's code, be it MS-Extra or standard B+G code, is capable of controlling the injector's pulse width down to 0.1mS (1000th of a Second). (...) The HR code has a resolution of around 0.035mS
Fuel control to 1 µsec (100 times more resolution than MegaSquirt-I)
I'm sure there is point of diminishing returns in this, though I don't know where it is exactly. I doubt that increasing the resolution to, say .01µs would buy you anything palpable over the MS2's stock 1µs resolution.
The point is simply that going from MS1-Extra to MS1-HR, and gaining three times the fuel resolution, is a significant change.
#26
Actually (from the horses mouth) it's "about 0.02 - 0.03" ms (20 - 30us) fuel resolution. MS2 does it to 0.6666us in hardware and if you look at the comparison chart between different ms variants here :
http://www.msextra.com/ms2extra/MS2-...l.htm#features
It still says 100us* where * = "hires can do better than this, but we aren't specifying how much"
I just thought it was unfair on MS2 to claim ms1e hires can match it for fuel accuracy. It simply can't. Light years ahead of ms1 ah la plain though, so your excitement is justified :-)
It's worth noting that the software accuracy behind the hardware output makes a big difference too. Case in point, ms2 b&g vs. ms2e, ms2e has extensively rewritten math to make use of the hardware accuracy that ms2 b&g just doesn't have.
Fred.
http://www.msextra.com/ms2extra/MS2-...l.htm#features
It still says 100us* where * = "hires can do better than this, but we aren't specifying how much"
I just thought it was unfair on MS2 to claim ms1e hires can match it for fuel accuracy. It simply can't. Light years ahead of ms1 ah la plain though, so your excitement is justified :-)
It's worth noting that the software accuracy behind the hardware output makes a big difference too. Case in point, ms2 b&g vs. ms2e, ms2e has extensively rewritten math to make use of the hardware accuracy that ms2 b&g just doesn't have.
Fred.
So I have MSPNP9093. i just installed it recently. I am installing my 550cc injectors. So I just need to download the Hi res and burn the firmware. Then everything works the same. Just more precise.
I was told my injectors were RC550CC from flying Miata. But how do i make sure they are rc550 and High ped. A red neck did own it.
Last edited by Toddcod; 06-16-2008 at 12:43 PM.
#27
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,046
Total Cats: 6,607
Anyone have the exact link to the download. I don't wanna screw it up. This isn't something I wanna search or try to find the right one. Trust me. I would appreciate the help.
#28
Junior Member
Join Date: Feb 2007
Location: 11368 miles from where i would like to be
Posts: 269
Total Cats: 92
That is all I was getting at.
To quote Phil, James & Co. from the MS1E-HR manual:
The MS1 uP's code, be it MS-Extra or standard B+G code, is capable of controlling the injector's pulse width down to 0.1mS (1000th of a Second). (...) The HR code has a resolution of around 0.035mS
So, regular MS1 or MS1Extra has a resolution of about 100µs, MS1Extra-HR has a resolution of about 35µs (three times more accurate)
Actually (from the horses mouth) it's "about 0.02 - 0.03" ms (20 - 30us) fuel resolution.
The point is simply that going from MS1-Extra to MS1-HR, and gaining three times the fuel resolution, is a significant change.
Additionally and somewhat off topic : ms2 res = 0.666us, not 1us :-)
When someone asks me about them I say this :
MS1base <= MS1Extra < MS1base HiRes <= MS1Extra HiRes << MS2 < MS2Extra
Which I think is a fair summary most of the time.
Fred.
#30
Former Vendor
iTrader: (9)
Join Date: Jan 2008
Location: Bay Area, California
Posts: 929
Total Cats: 9
Damnit, I really have to go do this. It takes like 5 forking minutes, and im still putting it off. I guess I just dont want to remove the MS from the car if I need the cap mod.
Are you guys just reusing your low-res msq files, or re entering tables by hand into the hi res version?
Are you guys just reusing your low-res msq files, or re entering tables by hand into the hi res version?
#32
Junior Member
Join Date: Feb 2007
Location: 11368 miles from where i would like to be
Posts: 269
Total Cats: 92
ini files on the other hand, NEVER try to use those! MS has direct memory addressing through serial and a wrong ini (as it says when you try) will/can ruin everything :-)
Fred.
#34
I have been experiencing a driveability issue whereby the car would jerk and stutter if I was coasting downhill with the throttle just cracked open a hair. My assumption was that it was due to the overrun fuel cut option which I had enabled. I had tried tweaking the fuel cut parameters with no success.
After loading the high-res code and re-tuning the VE map for it, the issue appears to be gone !
Additionally, I now have a stable idle at 13.5:1 whereas it would oscillate wildly with the 029y4 code if I tried to lean the idle above 11:1.
I am impressed with what the high-res code has done for my setup.
After loading the high-res code and re-tuning the VE map for it, the issue appears to be gone !
Additionally, I now have a stable idle at 13.5:1 whereas it would oscillate wildly with the 029y4 code if I tried to lean the idle above 11:1.
I am impressed with what the high-res code has done for my setup.
#35
My pulse gauge reads neither. Any ideas?
__________________
Best Car Insurance | Auto Protection Today | FREE Trade-In Quote
#37
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,046
Total Cats: 6,607
"1.7" instead of "1.745" was just an example, to illustrate the use of one decimal place vs. three. "2.1" is a number with one decimal place of precision. Also, as kotomile pointed out, it says right up at the top of your screenshot (on the titlebar) that you are running 029v, which is not a HiRes build.
#40
Anyone have any intrest in spending a few mins on the phone with me and walk me thru the install? Id appreciate it big time and would be sure to return the favor somehow. Im just affraid that I might blow something up.
__________________
Best Car Insurance | Auto Protection Today | FREE Trade-In Quote
__________________
Best Car Insurance | Auto Protection Today | FREE Trade-In Quote