sim's Volvo 245

All home-built FreeEMS implementations without a forum of their own, usually TA-based.
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Lucky Number Seven

Post by sim »

It works!

Today, after rewiring the CAS using two lengths of coaxial
antenna wire, and swapping in resistor plugs, I got the car to
run.

With a flat ten degree advance map, the engine started right up
and ran. Revving the engine causes the timing to flutter
slightly, no more than 1 degree back and forth.

I created an initial ignition map from Fred's truck map and the
information Nick Sayler researched while installing similar coils
driven by MegaSquirt on his Volvo 240. Nick explains his method
of deriving an ignition table from the factory service manuals
in his blog post here:
http://blog.nsfabrication.com/2010/03/1 ... on-curves/
Nick's car has a B230F, 200cc larger, but otherwise very similar
to the B21A.

With the initial ignition map, the engine runs very smoothly.
There is a noticeable difference from the old points set-up. I
took the car for a 20km drive, down a local highway, then up a
small mountain and then home. It is my normal shakedown route.

The car ran very well, except for occasionally losing sync. The
sync loss happened fairly randomly, but seemed to occur more
often at idle, and when well loaded. By the top of Burnaby
Mountain (about 300 metres of climbing) it had stumbled enough
times to generate a considerable backfire. Somehow, this caused
the #1 plug wire to come loose.

The coils are mounted to the valve cover, but the #1 plug wire is
only just long enough. I may need to get some longer wires.

After I pulled over and reconnected the wire, I made it home with
no further trouble.

The next step is to get set up for datalogging and diagnose/fix
the sync loss.

Video is here: http://www.youtube.com/watch?v=I5K-6jdSUeo
Thanks for putting it together Fred.
Last edited by sim on Sat Sep 17, 2011 9:39 pm, edited 1 time in total.
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

This week I replaced my dodgy USB-RS232 adapter with a new part
that uses a better chipset. It is a nGear product, SKU #17281
identified in dmesg by:
FTDI idVendor=0403, idProduct=6001

I attempted to use cutecom to record the data stream from the
EMS. All I could get was a half minute or so of data before the
serial connection reset.

I noticed also in MegaTunix that the connection would frequently
drop.

On a hunch, I disconnected the power ground for the coils
(not the logic ground, it's a separate pin on LS1 coils) and grounded
them instead to the cylinder head.

The engine ran much better, no longer losing sync.

My theory is that the high voltage from the coil had to go
through the coil ground in the loom and then back through the
block ground to the plugs. This induced a current in the power
line to the ECU which was routed alongside the coil ground. The
spikes overloaded the voltage regulators, which cut out and reset
the ECU.

It looks like this problem can be chalked up to my inexperience
in wiring a vehicle ECU rather than any flaw in the code or the
design of the ECU board. I'm pleased that the problem cost
nothing (but time) to fix and that the regulator did an
effective job of protecting the ECU from the spikes.

I'm now confident enough with the set-up to drive the car into
town for work.

The next step is making a set of det cans to listen for knock and
then playing with MegaTunix and the spark table.

Big thanks to everyone on the IRC channel for the support and
advice. I really feel like I'm standing on the shoulders of
giants.
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

I'm into my third tank of gas driving the 245 on FreeEMS sequential
spark.

The past weeks I've achieved some things:
  • Created circuits for BRV, IAT, CHT, and TPS on my board.
  • Jumpered R5 on the TA board to remedy the ~10% error in the
    ADC readings.
  • Removed the legacy "Wax Valve" from the head, and installed
    a GM CHT sensor in its place.
  • Installed a GM IAT sensor into the first elbow after the
    airbox in the intake tract (also a good place for a
    post-intercooler IAC, when I get there).
  • Built a wiring harness for the IAT, CHT and TPS sensors (TPS
    sensor will have to wait for EFI, but the wiring is ready
    now).
  • [off-topic]Replaced my aging water pump, and replaced the
    worn out rubber power steering pump bushings with poly
    bushings.
The BRV, IAT, CHT and TPS circuits have all been cribbed
shamelessly from the RavaAGE schematics. I ran out of 1N5817
Schottkys after building the BRV circuit (protected only on the
high side), so I used 1N5818 parts to protect the IAT, CHT and
TPS circuits.

I still need to clip the JB2 jumper on the TA board (to solve a
clock issue that I don't understand), it's a trivial
modification.

I've not yet built a set of det-cans, (to monitor, rather than
tune to the edge, for now), but I did acquire a set of
construction ear-muffs and I have hose. I need a hose tee, and
some time to set them up.

As it happens, my engine had a "wax valve" temperature sensor
installed in the head. It seems to be a placeholder, as is the
same valve on the B23E in our other Volvo -- it's not hooked up
to anything. The "wax valve" has the identical width and thread
pitch as a General Motors CHT sensor. The reach is almost the
same as well. I was able to remove the vestigial wax valve and
install a pick-a-part fresh GM CHT sensor in its place. Getting
the wax valve out was a tad difficult, a 22mm deep socket fit
over it, but the intake manifold was too tight to connect a
ratchet or even a wobbler extension to it (I couldn't find a
cheap 22mm crowsfoot wrench). I wound up buying a $2 7/8"
wrench, I bent the box end ~85 degrees with the flame wrench to
get the existing sensor out. The GM CHT sensor is skinnier, so a
deep 19mm socket installed it just fine.

A picture of the removal tool:
Image

The temperature readings are off, the IAT read 2c when ambient
was ~15c, but I haven't set up the firmware for GM sensors (or
anything) yet. It was very satisfying to fire the motor up and
watch the CHT values slowly climb as the engine warmed up.

Winter will be here soon, so knowing the temperature is critical.
Once the carb goes, so does the choke.

Soon on the agenda is acquiring a fuel pump. I believe I can use a
Walboro GSS340 pump as a sole in-tank pump. I'll need to create a
return line and mount the pump in the tank. I have a
tank+pump/sender assembly from a later K-jetronic car to adapt to
my purpose.

The car is running great so far. I am really looking forward to
fuel though. I want my idle RPM lower, and decel fuel cut, and,
and...
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

It has been a while since an update.

The car has driven an indicated 1984km since I installed the
FreeEMS ECU in mid September. The odometer reads about 10% low,
when it works, which is not all the time. It's safe to say that
the Snot Rocket has put at least two thousand kilometers on its
FreeEMS rig since installation.

The car showed a marked improvement over the points. I have not
yet tried to optimize the ignition advance, the car is happy
with the conservative map and I have a few things to do before
attempting to tune it.

I log all of my fill-ups, but the numbers are pretty meaningless
with a dodgy odometer. I do not think that I've picked up much
MPGs, though I'm fairly certain that the situation has not become
worse.

I've spent much of my Volvo time lately getting my partner's 242
sorted out for a three month trip to Thunder Bay Ontario, a very
cold place this time of year. Her '82 242 is doing great out
there so far.

Over the Christmas Break, I loaded new firmware into the car. In
the process of getting the Tupperware cap back on the EMS, I must
have tweaked the CAS cable. After the upgrade the car refused to
start, and exhibited symptoms similar to the sync loss issues I
had before switching to the coax cable.

Being busy over Christmas, and with a death in the family, I did
not get to diagnosing and fixing the problem until this week.

It happens that a ignition-only set-up has a few limitations. The
ECU could not hold sync, but the carburettor had no idea, so it
just kept mixing fuel into the air. Once while cranking, a random
spark managed to ignite the mixture in the exhaust system. The
resulting report brought a few of my neighbours to their front
doors. I'm sure that worse stuff could happen to a motor. I'm
looking forward to fuel control also.

The 12v feed for the CAS had pulled from the board. While this
was a problem, it was not *the* problem.

With help from a log taken while cranking, and Fred's input, I
narrowed the problem down to the inner track on the CAS wheel.

The excellent logging was a major factor in the diagnosis.

I went over the board carefully and it all checked out.

With the board on the bench, I could create a clean tooth log by
shorting the pins to ground.

The cable checked out electrically in the car.

I decided to pull the cable and re-do it.

Once the CAS cable was out of the car, it was no longer
continuous. The stress of bending it around caused it to fail
completely. I built a new cable out of stranded 18awg two twisted
pair shielded cable. With the new cable installed, the car fired
up on the first kick.

I'm very pleased to have a running car again, and I have no one
to blame but myself. The sold core antenna cable was electrically
suitable for the CAS signal, but was not physically suitable for
the environment.

The new cable is stranded and rated for 105c temps, I hope it
will be better than good enough:

Image

A side effect of all this crap is I made an effort to mount my
board securely in the Tupperware container I have been using. Now
the board and all the outgoing cables are attached to the deck,
and there is no strain on the board solder connections.

I used an old cookie tin to make a bracket to hold my board up
off of the Tupperware lid. I'm getting close to having the ECU up
and out of the passenger footwell. Look out turbo Citroen.

Image

On that note, congratulations to number eight and nine. What a
great start to the new year!

Another new development: I bought a low-mile 1995 Volvo 945 that
had been in a bit of an accident. The repair bill was worth more
than the car, so I got it fairly cheap. This 940 has a known good
pressurized fuel system for EFI, and a nice lightly used N/A squirter
block half the age of the motor in my wagon.

I'm keen to get fueling up and running, especially after having to
troubleshoot EMS sync problems with a carb. Worst case on fuel for
my car is six months out. It could be sooner.

Cheers all, and thanks again for the help troubleshooting Fred.
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

Tachometer Driver Circuit

My car, being old, has a tachometer that connects to VE- on the
coil and expects a brief high voltage spike resulting from the
field in the coil primary winding collapsing.

Under the new Coil Near Plug regime, the coils are driven by a
logic 5v signal, and there is no accessible flyback signal
available to drive the tachometer.

I built a circuit that takes a logic signal and produces a
flyback voltage using an optoisolator and a scavenged relay:

Image

Fred notes that it is not acceptable to drive an 82 Ohm
load directly from a CPU pin. He told me in IRC that I needed a
1k ohm resistor on the CPU pin. The optoisolator requires more
current to light up, so I intend to stick to my original plan of
using a line driver chip to drive the opto. DO NOT hook up
the above circuit directly to a CPU pin.

My board is pretty much fully populated at the moment, so I'm
going to either refactor the board, or add a layer to it. The
driver circuit for the tach out will be quite similar to the
circuits that drive the coils.

TODO: Post the driver circuit here when finished.

The tach driver is a pretty simple circuit. I decided to
implement it on its own board and locate it near to the cluster,
using the cluster +12v and ground and a shielded twisted pair
driving the LED in the optoisolator from the EMS. This ensures
that any flyback noise is kept well away from the FreeEMS board.

Here is the board I made, it has two female spade terminals that
I crimped, then wrapped a copper strand around, and soldered to
the board in positions that fit a standard relay. The board takes
12v and GND in, as well as an isolated 5v and GND signal from the
EMS, and produces a signal suitable for the tach on an output
wire.

Image
Image

After ensuring the circuit worked, I potted it in hot glue and
wrapped it in electrical tape. I'm still searching for a metal
candy tin that fits the whole business as an enclosure, I'm sure
I've seen one, but have come up empty so far. The Altoids tin in
the background is not deep enough.

Also, the relay buzzes a little bit. I pumped the mechanism full
of hot glue, but it was not enough. Hopefully a suitable tin box
stuffed with cotton will make it fairly silent. If not, the road noise
of a thirty year old 240 will drown out the buzz.

John Howe suggests ripping out the actual mechanism to ensure it
is quiet. I may do this in future. It is a good reason for making
the board mate to a standard relay, rather than soldering
something in.

For testing, I used an Arduino Uno to provide a 5v signal to the
board. The pulsewidth was the minimum (easily) possible with the
Arduino; 1ms. I wrote a quick and dirty program to sweep the tach
from about 4k RPM to 500 RPM, roughly 1k RPM at a step.

It was pretty cool when it worked after powering it on.

Image

(I have a video of the tach sweeping, but it is boring and it
sucks so just rest assured that it works).

All that is left to to do is mount the device to the back of the
cluster, wire up +12v and GND from the cluster, as well as 5v
signal and ECU GND, and implement a circuit on the FreeEMS board
to light up the opto LED as required.

Spankme wins, BTW, as he had a working tach mere days after
FreeEMSing his Citroen. I've taken months, but I also have 2k-kms
on my FreeEMS powered Volvo (tachometers are for the weak (I
really want mine to work)).

I really hope that my post helps other folks that wish to make a
"Old tachometer work with a modern EMS" or "Drive a
coil-connected tachometer with a logic signal" or "Hook up an
old-school tach to a modern engine" or whatever. Google was not
much help in searching for this (very simple) circuit, so here's
hoping that the next search finds this page.

Thanks Fred for writing the code to make the tach work.

[edited to ensure I don't mislead anyone with the 82ohm resistor]
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

PWM/Tach out/Remote console board

I've had the tach adapter built for a few weeks now, but I have
not yet hooked it up. There are a couple of reasons for that:
  • The main board is densely populated (for veroboard, at least) and
    adding stuff to it is pretty much impossible at this point.
  • The main board is a little fragile, and messing with it risks
    the car not starting next.
  • I've been fairly busy.
After some thought, I decided it was time to start another board
in the TA sandwich. I initially planned to use a board the same
dimensions as the main board. I acquired another board, and
soldered in a set of 50 pin two way headers. It became apparent
immediately that this would not work with all the wires and crap
protruding from the main board.

After a bit of head-scratching, I decided to make a small board,
with circuits in between the headers, and terminal blocks on the
outside for connections. The pigtails wired to the main board and
grouped into stiff bundles really started to cramp my style, I
figured that terminal blocks would make things compact and easy
to change in the future.

At first I came up with a set of things that I wanted to support
soon. The list was something like; Tach-out, WBO2, PWM for efan,
etc. After some reflection, I decided to keep it simple and limit
it to tach-out, and three 16bit generic PWM outputs. This would
allow me to use a single line driver chip, and get the most out
of it, while keeping it simple.

Here is the board I came up with:
Image

Another innovation is the use of color coded jumper wires. On
this board, red is +5v, white is ECU ground, purple is signal,
green is cable shield ground.

The ECU ground and the cable shield ground are isolated on the
board and brought out to separate pins on the terminal block.
Another lesson learned from the first board.

I installed a LED for each of the three PWM outputs and the
Tach-out as the LEDs on the main board have proven invaluable
when trying to figure out what is going on.

After building up the board, I realized that it would cover the
load/run switch on the TA board, and make flashing the thing even
more annoying than it already was. I thought about this for a
while. A toothpick epoxied to the slide switch would be in the
same theme as the rest of the build, but would likely just suck
in practice. I decided to spend an extra couple of hours to break
the load/run switch and the heartbeat LED out and mount them to a
switch panel in the car. I also added a USB connector to the
panel to make flashing a clean and no-dismantle operation.

Image
Image
Image

I got it all finished and installed in the car last night. I used
the load switch and the USB jack to update the firmware, and then
enjoyed the slow flashing of the heartbeat LED. The new console
panel rocks, and may be the best part of this work, even though
it came as an afterthought.

Unfortunately, the tachometer did not move. I still haven't
sorted this out, but I will. The LED on the new board did make
the appropriate rapid flashes, so the problem is likely solely
mine.

It was dark and late by then, and I gave up for the night.

Today, I discovered that I had lost my analog inputs. The car
sure ran for shit with no ignition advance...

I took everything apart and checked for shorts, and then gently
reassembled it all and crossed my fingers. It fired up and
worked. Another lesson learned, don't cram as much as possible
onto veroboard. Let the SMD jockeys work on density.

I've made inroads into keeping density down. I have connectors
soldered into a board I will use for fuel control in the near
future:
Image
That is a blank board test fit to the current rig. Should be lots
of room for four line drivers for the JBPerf P&H board, and WBO2
input, fuel pump relay output, and maybe an idle control circuit.

Another gain this iteration, a better soldering station for the
garage:
Image
It sits at just the right height for me at my stool, and has third
hand clamps and a iron holder screwed down. The whole thing
clamps in the vice. I still need to make a tower to stack those
spools of hook-up wire on. This is what I use to feed solder wire:
Image

All in all, I'm very happy with the new board, even though it
does not yet do what I built it for. I'll sort out the tach soon.

Oh, and the new rev-limit code is very fun! Thanks for that
Fred, and thanks for the assistance putting this all together.
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

By April, my homemade board had been in use for over six months.
The project was a success in that it ran the car, but it was a
big mess and was somewhat unreliable. The reliability seemed to
be deteriorating with time and use also.

I built a few electronics projects in my youth. I also spent
a few years working in a shop that built custom cable assemblies.
I'm pretty good with solder, and can read drawings.

However, the first incarnation of my FreeEMS board was easily the
most complicated electronics project I have undertaken.

It was built ad-hoc. It worked, but it proved to be
unmaintainable. I learned a bunch of things building the first
board. I learned what worked, but also, I learned many things
that did not work.

I often experienced strange problems after disturbing the first
board. Several times, the analog signals dropped out, and I'd
find myself driving with no spark advance.

On one test drive, I realized things were not right after pulling
onto the highway. I took the first exit into a marina. The
manifold was quite hot, and I reviewed the log in the parking lot
and saw the lack of advance due to the lost MAP signal.

After letting the car cool down, I started it up again and
gingerly headed back home. I just barely made it up the hill out
of the marina on 12 degrees of advance.

I didn't really expect to see the manifold glow until after I had
installed the turbo. I suppose it's good to get that milestone
behind me anyway.

A big part of the problem is how the analog circuits were laid
out on the board. I had packed everything into a tiny amount of
free board space and it was a disaster. A sub-two square inch
disaster.

In early April the car would not run right and I had had enough.
I decided to build a new set-up.

I made a trip into town to the discount electronics shop and
stocked up on parts.

One of the big problems was stress on the harness wires soldered
directly to the board. Not only were they unreliable, they
obscured much of the board and made it difficult to see what was
wired to where. I bought an assortment of screw terminal blocks
that would solder to the veroboard for external wire connections.

I wanted to space everything out logically in an ordered way.
After experiencing troubleshooting on the chaotic first build I
wanted to be able to easily trace any circuit. To do this I would
need more board space. I had a bunch of Samtec header blocks that
I had bought from their website. I bought a half dozen long strip
boards, with 26 longitudinal traces from my local electronics
dealer.

The old board mounting in the tupperware was done as an
afterthought and it sucked. I got a bag full of 25mm threaded
steel spacers, just about perfect for stacking boards. Everything
would be physically tied down. No fatigue stress would be
allowed. Hot glue, heat shrink tube, electrical tape, and zip
ties all help to make this happen.

I wanted to be able to painlessly disconnect the computer from
the car, and I wanted to eliminate any cable stress transmitted
to the board. I bought a number of panel mount DIN connectors
with mating cable connectors in different pin counts. I also got
a panel mount DB25 male, with solder cups, and a DB25 socket with
crimp on pins.

I drilled the corners of the boards, and also a piece of heavy
gauge sheet steel with nice rolled up edges. The steel face was
drilled for the DIN connectors, and I cut out holes to fit the
DB25 and DB9 by hand.

I wanted to have a well defined power and ground scheme between
the boards. I got a bag of PCB mount 90 degree male spade
terminals and set up a bus along one edge of the H1 wing on each
board. Each board gets Battery, 5v, ECU ground, and Shield
ground. The boards are daisy chained together with a simple wire
bus.

The grounds are all trees, so it is possible to isolate them if
necessary. There should not be ground loops.

The DIN and DB connectors were installed in the panel with
pigtails soldered on and shrink-wrapped into groups.

All connections to the internal boards are through terminal
blocks. The connector pigtails hook up to the terminal blocks.
This makes it easy to disassemble the stack of boards for
modifications, maintenance, or troubleshooting.

Image

The connector assignment is thus:

4 pin DIN: Power connector
12v (Switched by the main relay)
ECU ground (All grounds terminate individually at the head)
Chassis ground
Sheild ground

5 pin DIN: CAS sensor
12v
CKP
CMP
ECU ground
Shield ground

6 pin DIN: Ignition out
Cylinder #1
Cylinder #2
Cylinder #3
Cylinder #4
ECU ground
Shield ground

7 pin DIN: Injector out
(Future use)

DB25: Sensors/interface
BRV
IAT
CHT
MAP
AAP
TPS
EGO
Tach out
Panel load button
Panel CEL LED

DB9: Serial interface
Pass through to TA card

Image

The first board in the stack has the 5v power supply on the H1
wing and also contains the BRV circuit. Most of the H2 wing is
vacant. The voltage regulator is bolted to a heatsink and the
board.

The second board is the TA card. It is powered directly from the
power supply board below with a small cable that has a two pin
female connector that mates to the TA power header. (These wires
are the only off-board wires soldered to a board.) The TA card
DB9 connector is wired via a small harness to the DB9 on the
connector panel.

The third board contains the coil drivers and the CAS input
circuit. The ignition out uses the same hex line driver chip to
drive the smart coils as the first board did.

The CAS input, rather than using an inverting line driver to
condition the signal, uses a Schmitt triggered NAND to both
invert and clean up the signal from the CAS. The output of the
two NAND gates is fed to the remaining two channels on the line
driver. Each drives an LED and their CPU pin. In practice this
seems to be working as well as it did before. In theory, the
trigger will present a cleaner square wave to the ECU. The chip
count remains the same.

The fourth board is the PWM board detailed in a prior post. It
connects to the CEL and Load button as well as the tach out. The
tach function is not working, and I have not yet spent any time
troubleshooting it. The load button and the LED are greatly
appreciated. They make it easy to flash the ECU and to datalog.
The panel mount USB plug detailed earlier also makes
communicating with the ECU easy and fun.

The fifth board houses the analog input conditioning circuits.
The circuits are laid out in neat rows on the H2 wing.
The H1 wing is vacant.

Image

There were the inevitable kinks involved. I had initially had the
coils wired out of order, which not only made the engine fail to
run, it kicked back against the starter. (It is good to have an
over-engineered engine.)

Once that was sorted, I found a solder bridge between PT4 and
PT5. The symptom of this was the engine running on two cylinders,
with no spark in 3 and 4, but, when the parallel bench test is
run, all four plugs spark. Interesting, that one was.

One result of the false starts blew [a] hole[s] in the exhaust due
to igniting an exhaust system full of fuel mixture. I think I may
have roused some neighbors also. Perfect timing, as I have to
get the thing smogged this month. Ah well, smog will keep me from
running around on a "Harley" exhaust for too long, as it should.

Image

I have to say, the new rig is so much easier to work on. It can
be broken down to boards and examined. Everything is laid out
logically, well apart, and color coded, so it is easy to follow.

I can't report any difference, aside from the fact that the
sensors no longer cut out, but I feel a whole lot better about
the set up.

My sensor logs remain fairly noisy, so I still have some work to
do.
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: sim's Volvo 245

Post by Fred »

Bump for a sweet car! I'm jealous of your ignition setup. That is all.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: sim's Volvo 245

Post by sim »

The board rebuild made a big difference in reliability.

Circumstances have lead to the third ECU rebuild, another big
improvement.

I had an opportunity to buy one of the alpha version Jaguar
boards from Andy (DeuceEFI). It is now wired up and installed in
the car.

Image
It is a very nice part. The car seems to like it.

The Jaguar board has injector drivers built in. All that remains
between the Snot Rocket and fuel injection is installing the pump
in the tank, running some new hose, fitting a fuel filter and
swapping manifolds.

With any luck, the Snot Rocket will be squirting fuel properly
this month.
<@TekniQue> but in the end, it's code that makes a computer useful
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: sim's Volvo 245

Post by Fred »

Great news :-) I look fwd to a log from the new rig.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply