Stock ECM upgrade?

[darkfa8]

I found this on Conley Racing's web site:

The Buick ECM [Electronic Control Module] is set up to drive high impedance injectors. The stock injector’s impedance is 16 ohms. (A high impedance injector.)

In the past low impedance injectors have been used, successfully with the stock ECM. However, the [FP/INJ] fuse in the fuse box must be upgraded to a 20A fuse to handle the additional current load.

I have read numerous times that if you're going to use the 55lb/hr and larger low-impeadance injectors you have to have the ecm upgraded..well, after reading this on Conley's site, can someone tell me why you need to change something in the ecm?

I don't want to spite any of the vendors, however I'd like to know why this is done when according to a relatively reputable source, the only change needed is a fuse.. which makes sense since the coil in the injector when going from a high impeadance injector to a low impeadnace one just alters the current load/operation.

Or, is the stock ecm injector circuitry not sturdy enough to handle the lower impeadance without some sort of chip change, wiring upgrade or something?

 

[Russ Merritt]

Probably for the same reason you don't put tinfoil on a blown fuse and stuff it back in. Had it happen at 100mph when my computer was getting upgraded for ME-R, a friend let me use his spare computer, it wasn't modded for 55s and it worked till I floored it.

 

[KendallF]

Dan, you want to diss my cheap MAF pipe but you can't pay for an ECM upgrade?! That's not fast with class!

You may be able to get away with running low impedance injectors w/the stock ECM for a while, but the drivers aren't meant to handle the current load. And, as Russ mentioned, as duty cycle increases, so does current load, so they're likely to fail at WOT..

The driver mod replaces the stock injector drive transistors with peak and hold modules that are designed for the low impedance injectors.

 

[darkfa8]

since i posted, i found the module and zener diodes used to upgrade the ecms to allow the use of the lower impeadance injectors .. just did some searching through the DIY-EFI sites.

As for me affording anything...

In terms of the maf pipe issue, I have a kenne belle deal on my car now (came with car when i bought it), but will be putting on a custom fabbed up aluminum peice so i can move the maf sensor to the inlet pipe and just make the inlet tract to the turbo more "clean" and free of obstruction. My cold air kit is a Big Mouth kit from Buick From Hell.

Back to the ECM upgrade, I am very good with electronics and I frequently do wiring and assorted electrical work on cars here. I'm aware that the majority of semiconductors are cheap, and in this case a Injector Solenoid Driver and some zener diodes are quite a bit less than $150.

There are plenty of people out there who you can pay to build you something, fix or modify.. but when you do it yourself, its just that much more satisfying. I don't think there is anything wrong with that, especially in this hobby.

 

[TurboJim]

If you were so good at el;ectronics, you would have answered your question already. You must not understand current, voltage and impedance.

Lets take Ohms law.... E=IXR

Well, we know volts and resistance, lets get current.

14v / 16ohms is what about .8 amps.

Now P = I x E so .8 x 14v 11.2 watts. This is roughly what the ECM and its circuit are designed to flow and dissipate.

Now Take a low impedance injector

14 / 2 is 7amps. 7amps x 14 = 98watts

Mind you this is for ONE INJECTOR. multiply this all by 6.

Now we have a circuit designed to flow 4.8amps nominal, flowing 42 amps, and trying to dissipate 588 watts roughly where it was designed for ~70 watts. Now les cranmk the RPM's up so the transistor has NO cooldown time. What you think is gonna fail?

So what are we gonna do? Were gonna install drivers that are designed to handle these types of current. Thats why they do what they do. IN a p/h setup, the driver "senses" curent demand and ramps down current when the injector has opened. If you watched an injector event on a scope you'd see what I mean. Anyway, the P/H driver takes the current from maybe 7amps at injector open, to 1 amp or so to hold it open. This reduces current demand, and keeps everything colol in there.


My math might be off (I never said math was my deal), but you get the jist. If you DONT, I suggest you buy an ECM already done, or listen to whoever told you to "Just put a bigger fuse in it" and HOPE that the ECM doesnt shut down while you're making power or you'll be out WAY more than the $150 the mod costs.

Wiring a cooling fan, or soldering new leads on a coolant sensor doesnt make you electronically proficient.

 

[Buicksx2]

Yeah............what Jimmy said.......he took the words right out of my mouth of course I woulda divided by pie in line 9 but either way both our formulas Jim work out the same.

Point being is spend the money and get the upgraded ECM, you'll be glad you did.

 

[darkfa8]

i appreciate you explaining all that at length. i should have just gone back and deleted the post since I found the components needed to properly do the job.

What I found is a Cherry Semiconductor, now On Semiconductor CS453 Injector Solenoid Driver or equivalent Motorola MC348454-2. This is a monolithic integrated circuit designed for medium current solenoid driver applications. It has a 4.4amp peak, 60Vdc Peak Transient Voltage, Low Saturation Voltage and operates over a 4.5vdc to 24vdc range.

It reduces the load current by a 4:1 ratio, so for one 2.8 ohm Siemens 55# (ohm rating measured by Ron Gregory) injector, at a more realistic 13.6vdc at the battery with a decent alternator, we take the (13.6vdc/2.8-ohms), or 4.8amps/4 [4:1 ratio] =1.2amps, 13.6vdc x 1.2amps=16.51 watts

Whereas, with the stock type high impeadance injectors:

13.6vdc/16-ohms = 0.85 amps, 13.6vdc x 0.85 amps = 11.56 (4.95 watts less than using the Cherry driver and 55# injector)

The Cherry driver has a max operating temp of 257*F and the heat-sink portion of the TO-220 case would be affixed to either the existing internal aluminum heat sink of the ecm, or the actuall ecm case. The heat sink tab also acts as a functional ground.

The other thing is that the load current dissapation is handled by a appropriately sized Zener diode (Zener Clamp), so as not to damage the driver itself.

As per the spec sheet:

The fundamental output characteristic of the CS453 provides a low impeadances saturated power switch until the load current reaches a predetermined high-current level and then changes to a current source of lower magnitude until the device is turned off. This output characteristic allows the inductive load to control its actuation time during the turn-on while minimizing power and stored energy during the sustain period, therby promoting a fast turn-off time

Also, the '87-'90 2.0 turbo cars and '88 2.3 quad-4 cars had two injectors per driver (two drivers) and they run in Peak and Hold (low ohm) injectors. Obviously, DFI systems accept the low impeadance injectors too, so it can be done, I'm just trying to learn myself how to make the change to a stock ecm.

The software/components' output doesn't care about how the injectors are driven, so long that whatever driver is used is compatable with the output signal of the ECM, and it can integrate into the voltage supply and heat dissapation characteristics of the install/application.

There are some people on the DIY-EFI board who are using a GM Ecm to control a out-board injector driver.

This ECM modification isn't copyrighted or trademarked. Anyone who wants to nail me for taking one less ECM worth of work from Bailey or anyone else who does this for money.. I'm sure Bob has better things to do than complain about people who figure things out for themselves. I'm not trying to make money on this, I'm trying to save money on a piece of work that I know i'm capable of performing myself.

I made another post on a MAF pipe thread, about being "cheap", "tacky" .. Rich commented on that one...

Jim, you know full well I'm not Richie Rich, but while I'm not at the track every for test-n-tune trying to squeeze every ounce of power out of my car, and risk breaking it, heck in the 7 years I've had a car, I've only raced at a track 5 times (never broke anything at the track)... I am trying to find ways to upgrade the car to a level where I want it and to where I know I'd be capable of running in the et-range I want, and drive the car home without having to change anything more than maybe slicks, boost level, timing and fuel pressure.. or not even that much if and when I goto propane injection.

I drive the car too... bare minimum of 600 miles a week, every week.

 

[TurboJim]

Originally posted by darkfa8

Jim, you know full well I'm not Richie Rich, but while I'm not at the track every for test-n-tune trying to squeeze every ounce of power out of my car, and risk breaking it, heck in the 7 years I've had a car, I've only raced at a track 5 times (never broke anything at the track)... I am trying to find ways to upgrade the car to a level where I want it and to where I know I'd be capable of running in the et-range I want, and drive the car home without having to change anything more than maybe slicks, boost level, timing and fuel pressure.. or not even that much if and when I goto propane injection.

I drive the car too... bare minimum of 600 miles a week, every week.

I told you how to meet your goals. Not to read how everyone else did it, add the ways all up to do em all like yer Monte days, but listen to one or two guys that know. Dont listen to me, I dont really care, but you can put 2000# injectors in there driven by flux capacitors and the car isnt gonna be fast, You should be running 11.8 before you even get the bright idea that you need more injector.You whine to me that the monte boys caused you spend all this dough on **** that didnt make your car faster or prettier or whatever the complaint du-joir was, but look at what your doing now. Well my car runs 13.5, its making about 325hp, I am gonna do this and that despite being not necessary for your power level. THEN you are gonna whine the the car didnt respond like XYZ's did. I mean wtf is wrong with the 009's you got in there? Do you feel the need to upgrade cuz it sounds cool to say you have 55's in the car? Because i tell you what mr richie rich, you could be driving as nice, and be running that et on STOCK injectors...hell 325hp isnt even pushing the stockers yet. Next is PROPANE??? You should be running 12.90's on friggin PUMP GAS ya ding dong! You dont need a $500 propane kit!!! Just thought you said I dont have a lot of money!! Jeezzz....

So before you start whining about not having the $ to do an upgrade, think about all the useless **** you've done to that car, that did NOTHING for driveability, power or mileage. Then think about how much you woulda spend listening to the advice I gave you when we first met, which was a PROVEN combo, or hell, any ONE person.

I guess I really shouldnt reply to your posts, I dunno...you seem to get my goat most of the time.. guess its the ask a question, dont listen to answer, whine about the upgrade not working, ask the masses, get same answer you got from the start, hear about how poor you are, ask about another useless upgrade, over and over thats getting dry. I'm done wit diss one..

 

[Buicksx2]

Jim as for your recent post, I woulda still divided by pie in line 9. Bottom line is this you and I agree on the above and both our formula's would have worked out again, guess thats why we are gurus.........lol.

Sighhhhhingly Yours,

 

[darkfa8]

all i asked was a relatively simple question for the "gurus" in here... and i wind up getting my head blown off, because I might implement it in my car... gezz, I didn't even say I was putting it in my car.

I don't need it, just like I don't need a engine rebuild, but doesn't mean I can't ask questions to learn how to do it, or how stuff works.

What the hell is the point of this board anyway?

 

[ijames]

The short answer to your question is that the stock driver transistors in the ecm cannot reliably handle the larger currents from low impedance injectors operated in the saturated mode. Some people have tried it and gotten away with it, some people have tried it and let magic smoke out. The fuse may or may not need changing - I've never blown mine with 72's but I know someone else who has.

As for power dissipation, there was a thread several months ago that went into gory detail on this but I'm going to summarize since Jim didn't get it quite right (sorry, Jim) and I'm bored waiting for Emmit Smith to break Payton's record .

I'm going to use 16 ohms for stock injectors, 0.3 V drop across a saturated transistor, and 13.8 V for the alternator output voltage. Any one of these can be argued but let's not, okay ? (13.8 - 0.3 V) / 16 ohms = 0.844 A, and 0.844 A * 0.3 V = 0.25 W dissipated in the ecm driver and 0.844 * (13.8 - 0.3 V) = 11.4 W dissipated in the stock injector. Total power in the ecm is 6 * 0.25 W = 1.5 W (all these powers are assuming static injectors), and the total current is 5.06 A.

For the low impedance case I'm going to use 2 ohms for the injector and 1 A for the hold current with peak-n-hold drivers (the cherry cs454 holds with 1 A, the cs453 holds with 2 A I think, and so far as I know neither are available any longer from Cherry Semiconductor and Motorola discontinued their parts a year before Cherry). Okay, 1 A * 2 ohms = 2 V drop across the injector, and 1 A * 2 V = 2 W dissipated in the injector while holding. In the ecm driver, (13.8 V - 2 V) * 1 A = 11.8 W dissipated per injector, or 6 * 11.8 W = 70.8 W total. This is a huge increase, from 1.5 W to 70.8 W. Also, while holding, the max total current is 6 * 1 A or 6 A, so the stock fuse is fine. Since only one injector at a time will be in the "peak" phase drawing 4 A, the max current total will be 4 + 5 * 1 = 9 A or still okay for the stock fuse.

The other way to handle low impedance injectors is to simply use driver transistors rated at higher current. This puts the injectors into saturation and slows down the closing delay (and thus limits the minimum pulse width), but the empirical evidence says that it still works with 72's. I can see the pw drift up and down around 2-2.5 msec at idle so I believe that they are on the threshold of being too slow, but the engine idles just fine to the ear and the exhaust gas analyzer. Bigger injectors would probably start to have idle smoothness issues. Anyway, (13.8 - 0.3 V) / 2 ohm = 6.75 A, and 6.75 A * 0.3 V = 2.03 W dissipated in the ecm driver and (13.8 - 0.3 V) * 6.75 A = 91.1 W dissipated in the injector. The total power in the ecm is 12.2 W, more than stock but not as much as the peak-n-hold case, but now the power dissipated in each injector is huge, 91.1 W - don't run these dry! The total current is 6 * 6.75 A = 40.5 A so the stock fuse should be replaced. However, like I said, I've never blown the stock 15 A fuse this way. I run at most about 75% duty cycle, and the inductance of the injector will limit the rise time of the current on each opening, and I guess I'm just not quite over that line. On a motor where the low impedance injectors were run near static I suspect the fuse would go. I also added an extra ground strap from the case of the ecm to the dash, to help out the ground pins in the ecm wiring harness.

Sorry I couldn't squeeze a pi in for you, Rich, but that would have invovled getting into the rise and fall times and I figured most peole had alread skipped over this message by now anyway .

 

[darkfa8]

Thank you Carl,

Would the National Semiconductor #1949 work since the Cherry/On and Motorola components are no longer available?

 

[ijames]

The cherry and motorola parts were 5 pin to220 case self-contained devices which made using them pretty easy (only needed a 2W resistor to replace the driver transistor). The ns lm1949 looks nice but it's an 8 pin dip package which needs lots of external components plus a darlington power transistor. You would pretty much have to do a board with 6 of these and 6 darlingtons, and the easiest place to put that would be to tap into the injector harness under the coil pack. I think someone on the gmecm or diyefi list has done a board like that, so go to www.diy-efi.org and surf til you find it.

 

[KendallF]

Originally posted by ijames
You would pretty much have to do a board with 6 of these and 6 darlingtons, and the easiest place to put that would be to tap into the injector harness under the coil pack. I think someone on the gmecm or diyefi list has done a board like that, so go to www.diy-efi.org and surf til you find it.

Yup..I have some of the 1949s...after I ran out of the CS453s and couldn't get any more I picked up some with the idea of making a box that'd plug in between the injector harness and the ECM. They're still sitting out in the garage somewhere..

 

[TurboJim]

Well Carl, while I wont argue your math or knowledge, My only fault was not taking any built in current limiting into GM's circuit design. I cant argue with your formulas HOWEVER, I have "let the smoke out" of MANY a transistor (3904 rings a bell? it was 6+ years ago), in semiconductors back in tech school by not using a limiting resistor. So that was maybe the only fault of my calculations.

Ah well, never professed to be an EE, just know enough to be dangerous

 

[ijames]

Jim, I'm sorry, I shouldn't have said incorrect because what you showed was right. The main thing you didn't do was individually calculate the power dissipation in the ecm versus the injector. The ecm is a closed box with no cooling fan tucked into an almost sealed kickpanel, while the injectors have gasoline flowing through them for cooling so they can handle lots of excess power.

 

[S2V6RACER]

Originally posted by TurboJim
Well Carl, while I wont argue your math or knowledge, My only fault was not taking any built in current limiting into GM's circuit design. I cant argue with your formulas HOWEVER, I have "let the smoke out" of MANY a transistor (3904 rings a bell? it was 6+ years ago), in semiconductors back in tech school by not using a limiting resistor. So that was maybe the only fault of my calculations.

Ah well, never professed to be an EE, just know enough to be dangerous

EE. I feel like the Micheal Keaton in Mr. Mom saying "220, 221 whatever it takes". While reading this thread.

 

[NICKG]

who should you use to get this upgrade?

do these cars idle ok (with correctly matched chip) with big injecters? who does this mod and is it reliable and how much does it cost?

 

[lrayzer]

I second that question, I can make more money in the shop fabbing something than spending hours figuring out how to let out the magic smoke, which is probably what would happen.

Preferred vendors or list members ??

 

[Mac in SD]

I've got 96-lb injectors on a Stage 2 274 idling great, running great on the street, with a stock ecm modded for low impedance by Turbo Bob Bailey, and a Brutal 6 chip. 3.5 LS-1 MAF with MAF Translator. (Haven't been to the track yet, and I'm sure there are bugs to be worked out, but it's pretty brutal so far. Certainly an order of magnitude stronger than my old 11.20/120 stock block set-up.)

ECM mod by Bob was $125, I think. Plus express shipping, totaled $140. About a 1 week turnaround.