Paul,

 We ran the "bench test" on the EC-2 today, turning the CAS trigger wheel
 with a variable speed Shop Smith wood lathe.  This allowed rpm adjustment
 from approx 1900 rpm to over 6500 rpm.  I've enclosed some pics of the
 setup (IMG_0358.jpg) and the related scope traces (IMG_0365.jpg).
 Voltage
 measurements at the sensor pins (before resistor) were as high as 41v,
 but
 were a max of 8.25 at the zener diode/22k resistor.  The majority of the
 tests were with the CAS sensor gap set to .040 as this is what Bobby
 Hughes
 measured the gap at on his Renesis.

 The really strange thing, which I need Tracy's opinion on, is that the
 EM-2
 would only read correct RPM up to about 5800 - 6000 rpm.  This is
 about the
 same place where the engine starts cutting out.  At that point the
 readout
 went crazy and would indicate anywhere from 0 to 6300, and anything in
 between, and constantly changing.

 We put the scope on the injectors and have included those traces too
 (IGM_0363.jpg).  From what we could tell, the signals looked good up
 to 7500
 rpm.

 We also tested that the coils/plugs were firing up to about 8500 rpm
 with no
 obvious problems.  Of course, this could change when under compression.

 Another thing that we discovered is that RPM's on the EM-2 would not
 indicate correctly unless we powered the injectors during the test.
 RPM's
 were verified with a handheld laser tach.

 Since all of this began after converting my EC-2 to Renesis style CAS, I
 think my next step is to borrow Bob Darrah's EC-2 and try it on my
 engine,
 using the second gen CAS.
 Mark

 Nice Scope.
 http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=14408&lc=EN

 Something is going on in the software. I suspect it is a time
 related interrupt problem. Tracy may be using the software
 to decode the RX8 wheel and the processor is just not
 fast enough.

 Good work guys. Way to go. Now we are getting some where.


 I will do a littler hard ware decoding of the RX8 wheel
 this weekend and see what I can do.

 I do agree that unshielded RX8 CAS wires will pick up
 interference. I had 3 foot wires on there and
 hooked up Craig's function generator with a square
 wave at 3 Mhz and a 3 foot long antenna. I got
 about 1 volt peak to peak noise on the CAS wires.

 The first thing you must do is shield that CAS
 signal wire.


 Paul Lamar


 Hey Mark;
 Grounding/shielding are common problems with both stock and aftermarket.
 Is that CAS wire shielded?
 Does Tracy recommend grounding the CAS wire to the ECU?
 Any ferrous metal nearby could also be bad.
 Cheers
 Cary


 Paul,

First, yes, the CAS wires are shielded.  They are run separately, away from
other wires.  But I can only separate things so far.  However, as you can
see they could be shortened a bit.

Here's another shot of the injector trace.  They are 60#/hr units x 6 ea.
 Since the engine was not actually running, the EC-2 would be seeing
atmospheric pressure, and so both sets of injectors were firing.

Mark

OK. My mistake. Kill that last message. The injector is on when
the voltage is near zero. The injector duty cycle is near
90%. No more fuel is possible at that fuel pressure.

Looks like the FET drivers need damping
diodes as you have that huge negative spike. I don't think that
has any effect on the how the system runs however but it is a major
source of noise.

At what RPM was this data recorded?

I scaled the jpgs so I could read the scope settings easier.

Paul Lamar

Paul,

I don't know the exact rpm, but it was pretty high.  My system does have
diodes in the injector circuit, added at Tracy's direction.  He may have
some other suggestions after reading my email and seeing the scope traces.
(I hope he's reading his emails)

Right now I need to finish installing the Renesis water pump.  This is
requiring new plumbing and a re-do of the alternator belts.  So, it is
taking longer than anticipated.  If I don't hear from Tracy by the time I'm
done with this, I'll try calling him.

Mark

With the CD 4049 or CD4050 you don't need zeners
less than 15 volts.

OK I'll assume the injector pulse was timed with the RPM
on the scope read out.

With jpg IMG_0363_1 that is 15 ms. Or about 4000 RPM.
Does that sound about right? You will need to test it
higher than that to determine if the pulse width is
OK. Try 6500 if you can.


Are you using the 2nd gen front cover Mark?
If you or anybody else want to simulate a 2nd gen
CAS system on a RX8 front cover let me know.
Of course the RX8 front cover does not have
the old fashioned distributor like CAS.
I have a way of doing it.

I'll ber available just about any week end between
now and OSH to help with the Hytek.

Paul Lamar