Taking the PFI plunge soon

[brian617]

I'm betting it has to be the new body style 94-95 s10 pickups and the 95 blazer and jimmy. Forgot they put the CPI in the 94-95 s10 4x4s...

[brian617]

I think it is a good plan to build an MPFI system with abundant / affordable parts, and well documented .xdf / .adx files!

dave w

I agree, $OD would be the best CPI .bin to start with.

[PJG1173]

I'll check my schematics later today.

Mathematically it's possible to get 6 MPFI injectors to match the impedance of one CPI injector, see example below. By adding a 12 ohm resistor across both the terminals of 16 ohm MPFI injector yields an impedance of about 6.85 ohms. The second example shows the impedance of 6 MPFI injectors with an impedance of 6.85 ohms, yielding a parallel impedance of 1.14 ohms. Without an actual measurement of the impedance of a CPI injector / MPFI injector, I'm not sure what the actual resistor box values will be.

I think it is a good plan to build an MPFI system with abundant / affordable parts, and well documented .xdf / .adx files!

dave w

I agree totally. the more I think about it the CPI firing strategy is the way to go on this vise batch firing 2 banks of injectors. Obviously the CPI motors worked so no need to reinvent the wheel. this post by 1project2many indicates a variable tuning valve that could be an issue. http://www.gearhead-efi.com/Fuel-Inj...ll=1#post11352

I was thinking along the lines of wiring them in a series/parallel circuit to match the resistance. It might be able to get it close to the correct impedance without using resistors.

[PJG1173]

I agree, $OD would be the best CPI .bin to start with.

I think I have a BBNT memcal at home. finding it may be a problem though I may have used it in by buddies CJ5 304 TBI conversion. might explain why I had to disable the flag for the knock sensor

[PJG1173]

interesting reading. so if this article is correct:
-CPI fires injectors 3 times per crank revolution
-The injector driver circuit uses the peak-and-hold method, providing 4 amps to open the injector and about 1 amp to keep it open.
-The resistance specification is 1.9 to 2.1 ohms

http://www.tomorrowstechnician.com/l...uel-injection/

this article seems to support that
The assembly uses a single low impedance (1.5 ohms) "maxi injector"
http://www.automotiveforums.com/t462...ne_info__.html

and this CPI Vortec 4.3 liter 1.5 +/- .2 ohms
http://www.pirate4x4.com/tech/billav...ector_info.pdf

[brian617]

I think I have a BBNT memcal at home. finding it may be a problem though I may have used it in by buddies CJ5 304 TBI conversion. might explain why I had to disable the flag for the knock sensor

BHRD is the one I've been looking through, it's on the $OD page.

[1project2many]

I lost track of some of this but the CPI valve does not need to be present. It was a feature on the early engines that allowed a broader power curve. It worked well. My '95 Blazer out-pulled the '96+ models when towing.

When putting high impedance injectors together in batch configuration it is possible to control them with either saturated mode drivers or peak and hold strategy. The equivalent resistance of six CSFI injectors at 13 ohms is 2.1 Ohms. This is close to the same value as (8) 16 Ohm injectors which is traditionally handled with saturated mode drivers. Choosing peak and hold mode works because the driver circuitry will never reach the current threshold to switch to "hold." The driver will stay in "peak" mode while the injector is commanded on and will not overheat. Injectors with resistance of 13 Ohms or higher would be perfect for this application.

[PJG1173]

I lost track of some of this but the CPI valve does not need to be present. It was a feature on the early engines that allowed a broader power curve. It worked well. My '95 Blazer out-pulled the '96+ models when towing.

When putting high impedance injectors together in batch configuration it is possible to control them with either saturated mode drivers or peak and hold strategy. The equivalent resistance of six CSFI injectors at 13 ohms is 2.1 Ohms. This is close to the same value as (8) 16 Ohm injectors which is traditionally handled with saturated mode drivers. Choosing peak and hold mode works because the driver circuitry will never reach the current threshold to switch to "hold." The driver will stay in "peak" mode while the injector is commanded on and will not overheat. Injectors with resistance of 13 Ohms or higher would be perfect for this application.

the intake should be here Monday at the earliest so I can measure the resistance to see what they are at. I have found conflicting info on the injectors included with this intake.

[PJG1173]

So the question remains batch fire 2 banks of 3 or fire all 6 at once like a CPI. I am leaning towards the CPI strategy, but have been unable to determine if just changing the bit in the bin to pfi/cpi will give me the correct 3 injector firings per crank revolution required. I have yet to find anything that says this strategy needs a physical mod/jumper on the memcal.

So I guess my real question is what are the pro's and con's to each?

[Fast355]

I lost track of some of this but the CPI valve does not need to be present. It was a feature on the early engines that allowed a broader power curve. It worked well. My '95 Blazer out-pulled the '96+ models when towing.

When putting high impedance injectors together in batch configuration it is possible to control them with either saturated mode drivers or peak and hold strategy. The equivalent resistance of six CSFI injectors at 13 ohms is 2.1 Ohms. This is close to the same value as (8) 16 Ohm injectors which is traditionally handled with saturated mode drivers. Choosing peak and hold mode works because the driver circuitry will never reach the current threshold to switch to "hold." The driver will stay in "peak" mode while the injector is commanded on and will not overheat. Injectors with resistance of 13 Ohms or higher would be perfect for this application.

Yea the older 92-95 CPI engines really do perform better than the 96+ vortecs. The cam is similar in specs to what came in a F/Y car LT1 and they have that much more open intake with the tuning valve. Combined with the factory high stall converter and 3.73s the little S10 Blazer I have worked on with the motor had a nice power curve from off-idle through redline. I kinda wish GM had made a V8 version of that intake with the hotter cam and offered it on the L31 trucks. If you scale the 3/4 350 up the 200 HP/260 TQ becomes ~265 HP and ~345 TQ and a table top flat power curve compared to what the L31 was.

[dave w]

The pro to CPI is it will be almost factory, which is a good plan. The con I see to using 6 saturated injectors instead of one peak-n-hold injector is possibly needing a resistor box. On a stock or nearly stock engine, the resistor box will not effect injector response time that would effect performance. There might be a need to tweak the "injector" bias or offset voltages, possibly a con? It might be a good plan to bench test the injectors with the resistor box, and verify the "electronics" with an oscilloscope. Maybe compare the "electronics" of a factory OBD1 V6 PFI system that used either 1227730 or 1227727 to compare injector response times. Measuring the "electronics" would be a good plan.

dave w



Peak and Hold:

Peak/hold injectors are low impedence, usually in the 1-4 ohm range. When the ecu calls for fuel to be injected, it sends out voltage via the wire clips until a certain current level is reached (the peak part) (varies by injector size, company). For the duration of 1 pulse width, that current is slightly reduced and held (the hold part).

Advantages of this design:

• minimizes the injectors "on" time, resulting in faster response

Disadvantages of this design:

• increases coil heat, which can lead to failure over time

Saturated:

Saturated injectors are generally higher impedence than peak and hold, running in about the 10-15 ohms range. Unlike peak/hold, a saturated injector remains "on" for the entire pulse width.

Advantages of this design:

• reduces heat

Disadvantages of this design:

• slower response time
• can't handle large CC or lb/hr styles due to limitations in its speed.

[PJG1173]

The pro to CPI is it will be almost factory, which is a good plan. The con I see to using 6 saturated injectors instead of one peak-n-hold injector is possibly needing a resistor box. On a stock or nearly stock engine, the resistor box will not effect injector response time that would effect performance. There might be a need to tweak the "injector" bias or offset voltages, possibly a con? It might be a good plan to bench test the injectors with the resistor box, and verify the "electronics" with an oscilloscope. Maybe compare the "electronics" of a factory OBD1 V6 PFI system that used either 1227730 or 1227727 to compare injector response times. Measuring the "electronics" would be a good plan.

dave w



Peak and Hold:

Peak/hold injectors are low impedence, usually in the 1-4 ohm range. When the ecu calls for fuel to be injected, it sends out voltage via the wire clips until a certain current level is reached (the peak part) (varies by injector size, company). For the duration of 1 pulse width, that current is slightly reduced and held (the hold part).

Advantages of this design:

• minimizes the injectors "on" time, resulting in faster response

Disadvantages of this design:

• increases coil heat, which can lead to failure over time

Saturated:

Saturated injectors are generally higher impedence than peak and hold, running in about the 10-15 ohms range. Unlike peak/hold, a saturated injector remains "on" for the entire pulse width.

Advantages of this design:

• reduces heat

Disadvantages of this design:

• slower response time
• can't handle large CC or lb/hr styles due to limitations in its speed.

If only I had the test equipment to do that

[brian617]

I kinda wish GM had made a V8 version of that intake with the hotter cam and offered it on the L31 trucks.

I wish they had made a small block version of the the Vin "L" L29 intake. That's a sweet intake setup.

[1project2many]

the intake should be here Monday at the earliest so I can measure the resistance to see what they are at. I have found conflicting info on the injectors included with this intake.

A former co-worker ordered one of these marine intakes. We decided he would be better off to use GM injectors. We could measure the resistance but could not find the voltage curves or flow rate. We also decided to forgo the TB. We made an adapter to use an L31 (IIRC) TB. Too bad he never finished the conversion. :(

I wish they had made a small block version of the the Vin "L" L29 intake. That's a sweet intake setup.

Heck, they should have continued it with the 96+ 4.3. The '95 cam with the csfi intake and Vortec heads is a strong performer. The dual runner intake would have really upped the ante.

[PJG1173]

A former co-worker ordered one of these marine intakes. We decided he would be better off to use GM injectors. We could measure the resistance but could not find the voltage curves or flow rate. We also decided to forgo the TB. We made an adapter to use an L31 (IIRC) TB. Too bad he never finished the conversion. :(


Heck, they should have continued it with the 96+ 4.3. The '95 cam with the csfi intake and Vortec heads is a strong performer. The dual runner intake would have really upped the ante.

http://www.pacificp.com/forum/viewto...er=asc&start=0 shows them as Delphi EV6 24lbs/hr. @ 43.5psi. units with USCAR connectors. hoping to find a part number I can research when it comes in. kind of doing this on a limited budget or I would just upgrade to a 98+ computer and get HPTuners or efilive. I do have a set of stock injectors LS injectors from an 02 Camaro. I also have a L31 and 454 TB if needed but I was going to compare them all to see which would work out the best or be made to work the best. this motor requires 600+ cfm I'm at 630cfm now so I'm looking to stay in that ball park. I have also read that the MAP sensor needs to be changed since it has an integrated IAT. not sure why that would be a problem if you are only wiring the MAP in it.