|
Hi Tracy,
Thanks for the info. You mentioned there are
issues greater than the isolation of power between the primary and secondary
injector supplies. I would imagine some other issues could be generated by
the back EMF as the injector coils collapse. The question is, do the
diodes correct the other issues that you did not mention?
On the relay, do you have experience using solid
state relays in this type of application? I ask, because I have seen
problems with SSRs in noisy AC applications. Those were applications using
SCR based SSRs. With the SCR type, noise can cause the SSR to turn off
before the supply wave crosses zero. I do not have experience with the FET
type, but have to wonder if some of the back EMF generated by the coils
collapsing could drop the source/gate voltage on the FET below the turn off
threshold momentarily.
If the diodes and or SSR do not fully remedy all
the issues, would a driver board external to the EC2 that splits the
injector control into individual channels be an option. Power FETs might
be a good fit, similar to the FETs built into the EC2 for injector and ignition
control. I would think a P channel would be needed due the sinking control
level from the EC2. The P channel compliment to the IRL530N that's used in
the EC2 is a IRL9530NPBF. Not being a board designer, I do not know if
this would be a sensible approach, but for your edification I have attached a
diagram of a possible circuit. Would be interested in your thoughts as
well as comments on whether I got the device selection and polarities
correct. One element that is missing from the diagram is probably some
form of suppression (I believe you refer to as snubbers) to handle the
noise being generated from the coils. If this board looks like it would
work, would the snubber modification to the EC2 still be a benefit? (would
hate the see FedEx loose the controller again.)
Thanks,
Joe
----- Original Message -----
Sent: Monday, October 15, 2007 3:15
PM
Subject: [FlyRotary] Re: EC2 20B Injector
Power Circuits
As Al said, yes, these diodes are necessary. I'm still
studying this situation and determining what the best configuration is for the
20B version. I have been hampered by a number of problems encountered on
my 20B installation (too many to mention here but I'll talk about a few in a
minute).
Here is what is certain so far. There needs to be an
isolation diode in series with EACH of the 6 injectors, not just the two power
feeds to the primary and secondary banks. There are more important
reasons than the power isolation issue that Joe mentioned. In
addition to these diodes, there should be a diode (regular, not schottky) from
the junction of the secondary injector power feeds to ground. I am also
changing the spec on the 20B staging relay to a solid state type.
Attached is the preliminary schematic for this configuration but it is still
under study until I can verify everything on my engine. This is the
Renesis crank sensor version of 20B controller but the same changes apply
to the standard crank sensor as well.
One delay I had was due to my oil cooler springing a leak. As
you may recall, I made one out of an AC evaporator core.
Thought the core had failed like Rusty's had but it turned out to be
the way I installed the fittings on the modified core so I was able to fix
it. The leak was hard to see as it was hidden and did not show up
anywhere on the cooler itself. I thought it was the gear drive and tore
that down to replace the rear oil seal (didn't help of course). Then I
thought I had another mysterious engine problem that caused it to run rough at
random times so chased my tail for awhile until I found out that the throttle
body was inhaling the oil spraying from the oil cooler. It went on like
this for quite a while. Just now getting back to checking out the diode
and relay changes I mentioned. BTW, this cooler (and associated
duct) is working incredibly well. I can run the engine at power
settings high enough to maintain level flight (as determined by fuel flow
readings) while static on the ground at 90 degrees F and oil temps
stay in the green (~185 F). Water temps are almost as good (~200
degrees). This makes me confident enough to not bother putting the spray
bar cooling for initial flight testing (if I ever get there).
Also been out of contact with the list for awhile due to Wildblue
satellite Internet problems.
Tracy (Back on dial-up, Arghh.......)
On 10/14/07, Al
Gietzen <ALVentures@cox.net>
wrote:
Joe;
You are
right. We discovered this crossfeed issue some time back. Is
applies only to the 20B version as the 13B has separate drivers for each
set. Install Schottky diodes in the circuits for each injector.
Tracy is aware of
this, and has installed the diodes in his as well.
Al
-----Original
Message-----
From: Rotary
motors in aircraft [mailto: flyrotary@lancaironline.net] On Behalf Of Joe Ewen
Sent: Sunday, October 14, 2007 3:20
PM
To: Rotary motors in
aircraft
Subject:
[FlyRotary] EC2 20B Injector Power Circuits
I have been testing my wiring
harness from the EC2. I found a behavior that I did not expect.
The manual drawing (EC2 manual) show a single injector supply that splits at
the Injector primary and secondary disable switches. On my
installation, I have 2 separate feeder circuits, supply Primary comes from
battery 1, and supply Secondary comes from battery 2. I did this
for redundancy. If I had a failure such as a short to ground in either
circuit, its related overload should trip, but the other circuit should
remain active.
While testing the injector
circuits, I noticed what I would term an abnormality. With both
injector disable switches in the enable position, I turned on the power for
the primary circuit (I have LED indicators on all circuits that annunciate
power present) and the secondary circuit also showed power present even
though the power switch for the secondary circuit was off. The same is
true when the secondary circuit is engaged and the primary is off.
Initially this led me to think I
had miss wired something in the circuit. After checking out the
wiring, I found no issues. So I took a closer look at the wiring
diagram in the manual. It turns out the power was being back feed
through the injectors to the opposite circuit. On the 20B version of
the EC2, the injector control lines for each rotors injectors are junctioned
going into the EC2. For example the Rotor 1 Primary and Secondary
control lines are junctioned into a single input point at the EC2.
I can see several situations
where this may lead to undesirable operation. For example, if the
Primary circuit had a hard short to ground, it would take out the primary
circuit's overload (CB or Fuse.) Since the primary and secondary
circuits are electrically connected on the control line, the other circuit
would also trip. Another situation that could occur is a medium to
high resistance short. Current would then flow uncontrolled through
the unaffected circuit's injector to the affected circuit's injector then to
the resistance short. This could keep both injectors on 100%.
While this situation is unlikely, it seems possible.
I believe the 13B version
uses separate control lines for each injector, so this in not likely to be
an issue on the 13B version.
It appears that isolation diodes
are needed to prevent these possible problems. Before making any
changes I have to look the circuit over more closely, as well as a call to
Tracy on the next service day. My question to the group, especially
anyone with a 20B, is am I looking at this wrong?
--
Homepage: http://www.flyrotary.com/
Archive and
UnSub:
http://mail.lancaironline.net:81/lists/flyrotary/List.html
|