D-Jet Bench Setup, Let the experimenting begin (again).... Options

[pbanders]

See attached photo, this is my D-Jet bench setup. The black box on the left is an EFI Associates 1400 D-Jet tester (circa 1972) that measures the injection pulse width, simulates the trigger contact points over a range of engine RPM's, and provides reference values for the CHT and air temp sensors, as well as the throttle switch operation. The ECU is an 044 that I've used in the past for all of the oscilloscope traces on my web page, I can access any circuit in the ECU with it. I've got a TPS connected to the harness in the middle of the picture, as you turn it it you can watch the injection pulse width increase, then when you stop, you can see the pulse with relax back down (there's a delayed effect as well as an immediate effect). I have an air temp sensor plugged into the harness, and instead of a CHT, I have a variable resistor (half of a pot) that lets me simulate the engine temperature. That's a NOS 043 MPS in the upper right, with my vacuum pump attached to simulate engine load.

I've also got an interposer box that I can put between the ECU and the wiring harness that can be installed into the car so I can monitor all of the ECU pin connectors while driving the car. My current box is a PITA to use in the car, my old design sucks. I've got an idea for a better one that I'm going to have to build.

I've also got a Fluke calibrated pressure sensor that I can independently monitor the manifold pressure, or the pressure at any other point in the vacuum system. I can datalog the output from my DMM to get any of the pressure values as a function of time.

Now, just gotta do something useful with it all. BTW, working on the 914 is one of about a million things I'm trying to do simultaneously, so I tend to get to it irregularly, just hoping that's more than "never". I plan to look into the hot start problem first. I'm going to take some data when it happens on the car (i.e. measure the voltage at the CHT through the sequence of events), bench simulate it, then try to come up with some simple solutions. The idle stabilizer will take longer to do.

[stugray]

Nice! That looks a lot like many of my various workbenches I have around here!

You could invest in a USB oscilloscope for under $100 and it would make a great tool for that setup.

Please find an electronic replacement for the MPS so we can start using different camshaft profiles

I have given that some thought and the concept is relatively simple.
Read the Manifold pressure, and simulate the output of the MPS under the same circumstances.
The concept is really simple using a microcontroller to read a MAP sensor, then an analog output to drive the ECU with a waveform that is similar to the original.
The last bit is the hard part.
Simulating the actual response of the MPS accurately is tricky partially because it swings above & below GND, so you need a output op-amp with a +/- DC supply that has the range you need.
Still not that difficult to build.
I have worked through the concept and even have the MAP sensor and postage stamp sized microcontrollers that would work, and a +/-12VDC-DC converter.

If Brad is interested in experimenting, I might pitch in some parts and some code.

[pbanders]

Nice! That looks a lot like many of my various workbenches I have around here!

You could invest in a USB oscilloscope for under $100 and it would make a great tool for that setup.

You can't see it in the photo, but to the left is a full-on '70's era Tek oscope with storage and 4 channels, as well as a set of plug-ins that include a precision pulse width measurement module. I also have a full electronics lab where I've got a Seeed DSO Nano v3 scope, logic analyzer, bus analyzer, etc., and all the parts and stuff for doing sensor work on Arduino, Raspi, PICAXE, and XBee based systems. I just finished building a XBee sensor network for the house which uses a Raspi for the controller and data aggregation node, it's monitoring for leaks at my water heaters and monitoring motion at my side gates. Sends me a text message if any leaks occur or if the gates are opened, as well as a daily summary. I'm going to be expanding it in the future to monitor more systems, use mysql for storing the sensor data, and providing several ways to alert me and to display the system status on a web page.

So, yeah, I'm into electronics (IMG:style_emoticons/default/biggrin.gif)