[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: P&H Injector Interface



On Sat, 20 May 2000 21:27:50 -0400, "nacelp" <nacelp@bright.net> wrote:

>I'd have to shot myself if I told ya,
>Note to others, pays to save some parts numbers I throw out.

OK, fair 'nough, but I'd be saving part nos. too, if I thot that someone
could source them! The stuff in the Packard catalog is what you can get
NOW. That's what we have to live with, ain't it?

>*My*, numbers are the older numbers used when 730s (and it's "clones") were
>in production.

Yup, understand. All I was after was trying to equate them to current
production numbers. If you're confident that they're the same as the
female MP terminals shown in the Packard catalog, that's good enough fer
me.

>You realise how much research your skimmin off here <g>...

Heeelllll yeah I do. And I appreciate it mucho. Why else do ya think I'm
returning to the ole waterin hole so often? But I'd be pullin my tail in
with shame if I didn't think I wuz passin back some stuff of reasonable
value, mate.

>Well, hmm, I just finished some stuff where we ran some (3)  2.4s  static,
>and the only down side was the injectors getting hot.

Yup, understood. The DRIVERS in your box didn't get hot, the injectors
did. This is very valuable eXperimental data. Says that the driving
devices your "conversion guy" provided were low enough impedance, that
they're pretty much outta the picture, from a bulk resistance
standpoint. That's ideal. But can you (or the little guys, even) get
their fingers down into the connector contact and see how much hotter
they're getting? Nope, you cain't FEEL stuff like that, cuz as soon as
you put even an elf's finger on the contact, it's cooled by the added
mass; what matters is how hot they're getting when sittin there all by
their lonesomes. And for that, you can't tell but by following the
connector mfg's. current capacity guidelines. Thas all I'm sayin.

>At light
>cruise you at like 3-4 msec at 2000 rpm, gotta be lots of time for things to
>cool, I'd think.

Yup, you're probly right, and many would be fine with that. But if
you're making sumpin and following std practice, you cain't use those
cases as design criterion. Ya know.

The pins on the ECM connector are simply not up to the job, from an EE
standpoint. Gotta face it. (Unless of course I messed up and mangled the
analysis; that's ALWAYS possible, which is why you gotta sanity check
everything).

>Heck ya got a sensing resistor in there anyway, what's a couple .1s ohms
>gonna change opening/closing rate.

VERY good point. The CURRENT is made certain by the driver, if it's a
good one (which is all the injector solenoid cares about). I shouldn't
have suggested the injector is gonna pull in poorly *of necessity*;
you're right, that's not really the case, the driver will pull down hard
enough to make sure the current is what it should be, if it can. BUT,
there's only so much a driver can do (rhymes with "there's only so much
a mother can do" :), and when you run outta headroom, the injector's
gonna wilt. In this case it's a low-side driver, pulling down a load to
get the current to a certain point. Gawd help you if you've put in
enough contact resistance that a perfect ground won't get the current
you want. BTW, the drivers are, of course, never a perfect ground.

>Build one, ship her over, and I'll run it thru a ringer so tight I'll stand
>behind it if its passes my test.  We gotz lots of batteries waiting to give
>there all in the search for truth here.  Or pencil me out what ya want if
>other then the pdf ya offered of the cherry drivers

Well, let's get it to at least pass muster in the mind's eye, before we
go for the batteries. Lots of things batteries cain't see.

Also, Cherry devices aren't likely to be usable here, if we need to
drive 4 P&H devices on one line. AFAIK, nobody makes an integrated
driver that can do that, at the moment. It's discrete time, at least for
the batch-fire ones.

Gar


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