Injectors: P&H and SAT - Part 4

Len sabatine sabatine at epix.net
Wed May 24 03:14:30 GMT 2000


At 09:09 PM 5/23/00 -0700, you wrote:
>This is Part 4 for just the GMECM guys:
>
>Now that we've opined one mustn't mix P&H & SAT drivers and injectors
>willy-nilly, some astute pilgrim is bound to observe:
>
>"But wait, GM built at least one ECM that can drive either 4 SAT
>injectors per wire/driver, OR two P&H TPI (or one TBI) injectors on one
>wire/driver. Both BATCH-mode scenarios."
>
>So how do THEY do that? Altho I said you don't want to mix injector and
>driver types, that doesn't mean electronics can't be designed to drive
>either type, IF you can tell it which type it's driving. Here's what I
>mean:
>
>You CAN build robust, nay swarthy (heh) drivers that can pull open 2, 3,
>or 4 P&H injectors OR drive a bunch of (say 4) whimpy SAT injectors in
>parallel. GM did it for either 2 P&H or 4 SATs. The peak current
>requirements for the former (that is, 4 P&H) are much higher, but the
>thermal requirements for the latter are higher once you've got to keep
>them on for long periods. So what's a mutha ta do? Just design them for
>BOTH, but you have to KNOW (and telling the drivers would also be nice
>:) which mode you want them to run in.
>
>The problem is that this is perfectly economical if you're just dealing
>with two wires you're going to drive in either case. Two BIG drivers
>needed for either case, not a problem.
>
>So, OK, yes GM or anyone can make a single driver to batch drive
>either/both type of injectors, and quite a few at that if they wanted.
>They just didn't happen to want to spend the bucks to do that in *all*
>their batch-mode ECMs. Shhhhh, that's a good thing for moi; otherwise
>you wouldn't NEED any interface modules. Heh.
>
>But what about the SEFI case? You need there as many drivers as you
>have injectors. But if you're running BATCH you only need at most two
>burly drivers. So combine the two? To make a module that would drive
>BOTH, you'd have to have as many SEPARATE drivers, as you have injectors
>(to cover the SEFI case). That means lets say 8 drivers. That's OK all
>by itself; there are some really nice cost-effective integrated driver
>devices available for individual injectors.  But TWO of these drivers at
>least (for batch support), would require the ability to have current
>limits setable for 2, 3 or 4 P&H devices in parallel (this to support
>those guys upgrading from batch saturated to batch P&H for 4, 6, & 8
>cyl). So on those, no integrated drivers. OTOH, these two potential
>batch drivers need to be electrically and thermally extra beefy to the
>point of covering any batch-drive and current-sense requirements of
>multiple P&H injectors in parallel. A workable minimum requirement would
>be 4 TPI P&H in parallel on each of these two special drivers.
>
>So what do you end up with? A box that is BOTH sets of hardware, with
>very little overlap. If you use it for Batch, you paid for 6 SEFI
>drivers, their support devices and connector space, which you don't
>need. If you use it for SEFI, you paid for two pretty husky more
>complicated but versatile switchable drivers you didn't really need.
>Either way, you essentially paid for BOTH. That's OK if you wanna pay
>for both.
>
>Oh well, I guess we're back to "mixing it up doesn't pay". I just knew
>that was gonna happen.
>
>Gar
>  Geee, Don't comprende' why it's such a rub to make a "BOX" with
  application part numbers for each of the mentioned combinations.
  Oem's got lots of bucks to integrate functions,specially when they
  pump millions of parts. Soooo , little guys choose wisely and cover
  the most popular bases and keep costs down. That way folks with less
  funding can build cost effective combinations, plus avoiding all the
  advertising hype and paying for that as well. I feel many aftermarket
  builders,installers could use the hardware your're speaking of. IMHO,
I don't feel a FITS ALL can be cost Effective, although,others ritefuly
mention the convenience of a Single Application Part.

Len.



>----------------------------------------------------------------------------
>To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
>in the body of a message (not the subject) to majordomo at lists.diy-efi.org

----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org




More information about the Gmecm mailing list