Vince,
Where are you buying small quantity of the KMI15/4 sensor?
Neither Digikey nor Mouser sell quantity one units. Looks like a
minimum quantity is around 800 pcs (T&R).
Admitted I could get samples for my Arrow FAE, but looking online at
arrow's web site, they are not stocked.
I hate designing stuff around "unobtainium" parts. Semiconductor
sector is picking up, but our lead times are being pushed out from
12 weeks to 20 weeks due to demand (to save money everyone dropped
inventories), so as things ramp back up you'll see allocation.
One tip for aircraft/automotive is to see if they sell a -40C to
+125C (automotive) version of the part. Even it you use the -40C to
+85C Industrial version, the device has been qualified for
automotive. Makes life much easier even if not running at +125C.
Oh stopped in at a Houston based customer who is using one of -40C
to +85C rated devices, he had the part running at 180C Ambient! I
was shocked it was running.
Tom W
Tom -
Buying? BUYING? I'm sampling, baby.
I am trying to order SAMPLES from NXP, but they are a little more
trouble than usual. For instance, TI is a simple matter of clicking
on a link and providing shipping info. I received a few sample TI
MSP430 uC DIP chips I use for prototyping. (I usually use samples
during design and testing, and actually purchase the final-design
parts.)
Your unobtainium comment is serious. I don't like weird components
either. If these parts are going to be 800qty lots, I think we can
just move along.
Your point on verifying temperature ranges is well taken. I'll look
to make sure -40/+125 parts are available when selecting components.
Oh, by the way EVERYONE, if I
*ever* start to go down a bad design/
test road, please PLEASE save me the trouble and SPEAK UP. I'm never
too proud to change when someone presents a valid case. I'd love
saving months of banging my head on the wall!
For the Flywheel CAS Sensor (#2) project, if anyone has a SIMPLE,
CHEAP component or package to recommend as an alternative to the NXP
KMI15/4 sensor, lay it on me. If the NXP KMI15/4 is unobtainium, then
I'll probably resort to standard automotive Hall Effect or Variable
Reluctance sensors and use one of the following:
http://www.onsemi.com/PowerSolutions/product.do?id=NCV1124DG
http://www.national.com/mpf/LM/LM1815.html
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/5410
(Of course I'll look at other recommendations.)
Back to the Returnless PWM Fuel Pump Controller (#1) project! It is
stalled looking for an alternative to Ford's FRPS (Fuel Rail Pressure
Sensor), minimum $50 each new. My design calls for two - one for each
fuel tank pump controller. Of course everyone probably saw my help
request for locating other automotive fuel pressure sensors - SPEAK UP!
-Vince Orton
Vince,
That is one reason I like using Digikey, to make sure I can get the parts! I also made sure the
parts my company makes are available from Digikey, as I've run into engineers who design stuff by
seeing what is available and ships overnight. THe Digikey warehouse is a cool trip on how things
are done (online order entered to completed box dropping in the shipping bin in something like 9
minutes!).
Napa, at one time, had a 'electrical catalog' where you could just scan down the book looking at
photo's. Not online anymore, as you have to select vehicle then look. Most vehicle OEM went to
'returnless' systems to meet evaporative emissions. Less plumping, less vapors.
With a return line to a small header tank, fuel returning before the pump, you can retain the older
pressure regulator. There is a myth that 'fuel gets too hot' yet, when measured we were within one
degree! Aluminum fuel line, aluminum header tank disappate heat, so the "hot fuel" doesn't really
exist, but always reported! I have high wings, both gravity feed to a common Andair R-L-Both-Off
Valve.
With a return loop system, the fuel regulator is engine vacuum modulated, less software to modify
the injector pulse width (if you hold a constant fuel rail pressure and engine vacuum changes, you
need to modify the fuel pulse width to compenstate for the pressure differential. Happens too fast
to modify fuel pressure... hence why the Ford Engine controller uses a floating point unit! With
MATLAB we did some cool demo's on modifying the fuel/spark on a cylinder per cylinder basis... which
is part of the reason for eTPU (Time Proccessing Unit)controllers built into the uC.
Tom
--
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Youtube key word UTUBPLEASE
Copyright 1998-2009 All world wide rights reserved.