[Diy_efi] 18t injector upgrade to more horses

[Perry Harrington]

> 
> Gonna share, or is that proprietary?  ;)
> | Adam Wade                       1990 Kwak Zephyr 550 (Daphne) |

Well, that's a good question.  I spent 150 bucks on it, just to look at it's
innards.  I don't even have a car to put it on!

I told the dealer I bought it from that I'd give him a copy of the information
I learned in exchange for selling it to me at cost.  He isn't a technical type
beyond his immediate need to make money.  I was rather disappointed at his
response.  He was merely looking for another company to make the same thing
cheaper and smaller for him so he'd have a better margin.  As it goes, these
things are $250 IIRC.

It's been a while, so I assume that several companies have scraped the design.

The information is from when I first instrumented the device.  Since then, I
understand a lot of why certain design decisions were made.  Suffice to say
that it seems to be a pretty complete and clever method.  I think a functional
duplicate could easily be made with less parts and lower cost by using an MCU
instead of discrete analog.

I don't have a schematic or other information on the device.  This information
came from physical observation and instrumentation of input vs output.

--Perry

Split Second BCS boost controller information.

Unit has 6 hookup wires and 1 jumper wire.

Pinout and wire color is:

ECU pin	35	Yellow/Blue stripe	TPS wiper connection
ECU pin 98	White/Black stripe	+5v (MAP sensor power)
ECU pin 101a	Gray/Blue stripe	MAP sensor scaled output
ECU pin 101b	Gray			MAP sensor input
ECU pin 104	Yellow/White stripe	Wastegate actuator solenoid GND return
ECU pin 108	Brown			GND

Observable parts list:
LM293	Dual comparator
LMC6484	Quad high CMRR opamp
LM3914	Bar/Dot mode analog display driver
IRF530N	HexFET
10k pot	Desired boost control
100ohm	1Watt resistor in series with FET and wastegate solenoid

TPS input is used to detect WOT.  It appears that the BCS interprets WOT as 3.0v
on the TPS input.  The unit will not control the wastegate below WOT.  This is presumably
for emissions compliance if CARB certification is desired.  Smog laws do not apply
at WOT.

The MAP input is used for controlling the wastegate circuit and LED bar graph.  The bar
graph output is tied to the MAP input.  Above 3.6v the MAP output to ECU is scaled
non-linearly.  It is hypothesized that the MAP input is never intended to exceed 3.6v.
At 3.6v input the boost gauge reads 14.5 psi.  This is not in line with known information
about conventional 2 bar MAP sensors.  Perhaps the VW sensor is equivelent to the MOT
sensor and is a 2.5bar.  Assuming a linear response to pressure a 2bar sensor would
output ~5v at 2bar (14.5 psi).  The MOT sensor output for 2bar would be 3.84v.  This is
close enough to the observed values to lead me to believe it is a 2.5bar.  The 2.5bar
sensor can read up to 21.81psi (150kpa positive pressure).

The unit controls the wastegate actuator in parallel with the stock computer.  A hexFET
is used with a 100ohm 1watt resistor to actuate the wastegate with 5vdc.  The wastegate
is actuated low; the ground is modulated to control the amount of bypass.  The unit
appears to have a wastegate adjustment range of 48% duty cycle to 90% duty cycle.  The
solenoid is actuated at approximately 50-60hz, this is in line with other solenoid
actuators such as the Ford ISV.

The controller has 2 inputs for controlling the wastegate.  The desired boost and actual.
There appears to be a small amout of lead that the desired boost (POT) can induce in the
wastegate control.  When the unit requests more boost than current, the wastegate duty
is reduced to 50%, allowing approximately half vacuum to the wastegate.  It is unknown
whether the wastegate is linear with application of vacuum.  As the boost nears the
desired value the unit runs in a closed loop to control the value.  The unit will swing
betweent 50% and 90% duty to compensate.  In actual use the duty would probably stay
within a fairly small variance around 68% to 74% duty.  When the amount of boost desired
is less than current, the wastegate duty is increased to 90%.  When the value drops down
to the desired value the unit runs in closed loop.

At the lowest end of the spectrum the unit excercises no control over the wastegate,
presumably allowing the stock computer to control it.  The manufacturer claims the lower
adjustment limit is 6psi, which is the stock computer level.  The unit has the
ability to drive the boost lower, but that would require more circuitry.

The boost gauge uses an LM3914 in application note form.  The reference voltage is 1.25v,
thus the 14.5psi LED is lit at max.  The range is .04v to 1.25v.  1.25v is reachead at
a MAP sensor value of 3.63v.

The other chips in the unit are an LM293, presumably for the wastegate oscillator and
duty control, and an LMC6484, a quad low input bias current opamp.  The datasheet for
the LMC6484 describes it as a high input range amplifier.  Upon observation of the board,
the chip is not being used as intended by the manufacturer.  There are no guard tracks
around the inputs.  It is thus hypothesized that the chip is merely being used for it's
near rail-to-rail output so they don't need a power supply circuit.
The opamp is clearly the analog comparator which controls wastegate duty cycle based on
the differential inputs of the boost pot and the map sensor.


-- 
Perry Harrington			Data Acquisition & Instrumentation, Inc	
perry at dainst dot com					 http://www.dainst.com/

Those who would give up essential liberty to purchase a little temporary safety
deserve neither liberty or safety. Nor, are they likely to end up with either.
                             -- Benjamin Franklin

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi