MS3 + P&H + VR - another way

Main site: http://jbperf.com/index.html

MS3 + P&H + VR - another way

Postby aarc240 » Sat Feb 19, 2011 7:45 pm

Here's another way of achieving an MS3 with "sort of MS3X".

A brief rundown of the car and configuration:
It's a Datsun L6 [inline 6] with modified 280ZX injection, lo-Z injectors, GM 65mm throttle body, EDIS plus Chrysler coil pack ignition, electric fans.
Chassis is also fitted with front wheel VR sensor [actually it has more than one but not significant for this].
I had been running MS2e with batch injection using 2 of Jean's P&H boards and wasted spark, good but needed to be better.
I wanted MS3 for sequential fuel injection, had found wasted spark worked fine so really didn't need MS3X as it doesn't support lo-Z injectors and sequential ignition would be unused.
Cam signal was now required - easy as I still had a spare VR input available on Jean's VR board [one side already used for front wheel VR for other reasons]

So after a rebuild into a new case I have an "MS3 plus" as shown in the attached images.

Two P&H v1.1 boards, one VR board plus a small patch board all attached to the "lid" of the case.
The P&H boards use the same pins on the "MS3X" connector as normal for injectors, except that one [unused] IGN output has been usurped for the sixth earth return path for one P&H driver. They are configured as 3 on each, alternating in sequence between the boards and are driven direct from the MS3 board.
The VR boards use the usual "MS3X" input pins for cam and VSS
The patch board also uses the usual "MS3X" pins for flex fuel, datalog and N2O inputs.
The v3.0 main board is configured for EDIS [as per MS2e], 12V aircon input [from the "MS3X" pin via the green wire and user circuit to JS4], aircon output via FIDLE, fans on INJ1, N2O on INJ2, realtime barometer, table switch via SP4 and user circuit to JS7, tacho output via SP3 from user circuit and JS11, IAC stepper.
Power is supplied to the P&H and VR boards direct from the main board via the red & black wires but the patch board is supplied from the MS3 board.

The white spots on top of the MS3 IC's are adhesive - there's a clear acetate sheet glued on there with a folded "skirt" extending down almost to the main board to prevent abrasion of wires and subsequent shorts.
To ensure nuts don't fall off when in use all the brass bolts were over length and then cut off with diagonal cutters after assembly thus distorting the threads enough to destroy nuts on removal.

It uses WBO from a Techedge LD1 [analogue output] at present so the next change is to fit an I/O-X board from Jean into the same tall Hammond case and change to CAN input for WBO digital.
The I/O-X will also get rid of a couple of tacky data collection boards so that's a bonus!!

Externally, this connects to the car power supply via a custom relay / fuse box, and all earth wires connect via a substantial copper earth bus [75x25x6mm solid bar] which is electrically isolated from the chassis and then 2G cable direct to battery earth.
Attachments
MS3a1.JPG
MS3a1.JPG (255.09 KiB) Viewed 6685 times
Last edited by aarc240 on Mon Jul 25, 2011 6:26 am, edited 1 time in total.
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Mon Jul 25, 2011 6:10 am

A small further update.
The TechEdge LD1 is now gone, replaced with an SLC (actually an SLC/OEM) from Alan To at 14point7.
The decision to do this was driven by the support already available from Jean in the IO/X code for the I2C connection whereas the LD1 would have required more coding from Jean to support the LD1 serial connection for probably only one user.

I've also attached a Word doc of the wiring table I use.
The DIYAutoTune harnesses match up with this table on the main box (MS3 & MS3X connectors) but the rest is custom.
Feel free to use/modify in any way.

The common earth bus bar has been replaced with a more substantial version using a tin plated brass bar and tin plated brass screws.
There was some evidence of localised heating on the copper bar adjacent to a couple of earth connections which could be attributed to insufficient clamping pressure due to the short screws and thin bar section. This allowed some surface oxidising of the copper bar under the connectors.
Tin plating the original bar may have solved the problem but why take a chance?
Attachments
wiring_table.doc
(164.5 KiB) Downloaded 265 times
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Mon Nov 14, 2011 11:11 pm

I've been a bit slack on updating this!
Attached images show SLC/OEM modified slightly by addition of extra material to fit in the slots of the Hammond case used for the JB IO/X.
The black/white wire pair is the I2C data line to the IO/X board.
The black 'goo' is a liquid electrical insulation material, a b***h to use neatly but excellent for both insulation and vibration stress absorption.
Yes, the voltage regulator is laid down almost flat, next post shows it's a bit cramped in there.
Attachments
SLC-OEM2.JPG
SLC-OEM2.JPG (74.82 KiB) Viewed 6467 times
SLC-OEM1.JPG
SLC-OEM1.JPG (87.75 KiB) Viewed 6467 times
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Mon Nov 14, 2011 11:26 pm

And this is how it all goes together in a tall Hammond case.
The IO/X is pretty much fully loaded and most inputs & outputs are in use, primarily with data logging, unfortunately only by external laptop as the IO/X SD card isn't yet usable..
The add-on board for the IO/X has 6 EGT sensor inputs but is otherwise fully loaded.
It's amazing how many resources get used up when one car is used as a test mule for another!

As can be seen in the case, I've opted to use a DB15 for the SLC/OEM.
I chose this route after cursing the 8 pin microphone type connection of the TechEdge product which proved very nearly impossible to fit the cable to without cross threading when the unit is in a confined space.
DB9 would provide enough pins but a DB15 allowed me to use 2 pins for each of the heater power wires which has proven sufficient.

The patch board inside the MS3 case has undergone a rework so that the air conditioner input (12V from the OEM compressor activation circuit) goes to K2 via a protection circuit duplicating the MS3X N2O input, as does the N2O input going to PH7. Neither has a pull up resistor and hopefully the poorly documented CPU port at the other end of K2 can handle this.
Datalog input duplicates the MS3X input including pull up.
Flex fuel input now relies on the MS3 on-board protection circuit as the source of RF energy induced into that input has been fixed by moving the RF amplifier.
MAF is a new addition (adventure or complication?) and now uses the green wire to the main v3 board to JS5.

Sometime, I'll get a shot of how they both fit on the passenger A pillar after re-installing everything in place of the triple Webers.
Attachments
IOx1a.JPG
IOx1a.JPG (68.59 KiB) Viewed 6467 times
IOx2a.JPG
IOx2a.JPG (72.15 KiB) Viewed 6467 times
IOx3a.JPG
IOx3a.JPG (62.22 KiB) Viewed 6467 times
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Sat Dec 31, 2011 8:26 pm

Now that Alan To has supplied an SLC/OEM module with updated firmware, that module has been fitted with 3 SIL sockets.
The SLC/OEM carrier board has matching SIL plugs (ie square pins) so the module plugs into the carrier.
This has raised the module about 1.2mm, the consequence being that the CAN connection visible as a two pin header in the pics immediately above now interfere with the module on assembly.
No issues once it's in place but a real hassle to assemble so those pins are now gone.
Unknown yet how it performs as the I/O-x needs updated firmware to use the new SLC/OEM firmware and TWI addressing.
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Mon Feb 20, 2012 9:47 pm

Small rework of connections to free up JS11 for future use with JM knock sensor - assuming it works!
The doc mentioned above has been revised to separate the internal connections which will hopefully make it easier to keep track of the current state of the boxes.
Again, feel free to use as you wish.

It's going to be interesting to compare performance and fuel use once the TWI WBO connection is finalised and this gets back in the car.
Currently we have 192 RWHP in a 1.1 ton car with a good sized thirst, but that's only to be expected with old skool triple Webers and breaker point ignition.
Attachments
wiring_table.doc
(178.5 KiB) Downloaded 235 times
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Mon Nov 12, 2012 8:51 pm

About time this got some more data!
Baseline is a 2.8L inline six cylinder fitted with triple 45 DCOE Webers, custom headers, hi-lift street cam, ported head, 10.4:1 compression in a 1.1 ton car with 3.9 diff and 24" tyres. Used 98 octane fuel at 10.3l/100k (27.4mpg imperial, 22.8mpg US) on highway cruise, don't even think about the 'around town' guzzle (not actually measured but about 17l/100k).
After refitting the MS controlled EFI it is much nicer.
Horsepower dropped slightly from 192RWHP to 189RWHP (but that can happen just with changing weather!). Torque junped by 16% but more significantly the torque peak is now almost 600rpm lower without much top end loss, giving a wider power band. In fact if the old torque peak value is used and the same value found on the new curve, there's now slightly more than 800rpm increase in usable power band!
The knock sensor works.
I'm not entirely satisfied with the way the knock sensor software works as it appears to pull back the timing in one big step rather than progressively, not that this has been a problem, just that it seems to waste some of the potential. However, it has allowed ignition advance to be pushed closer to the limit.
Still using 98 octane fuel and consumption is now 8.1l/100k (34.9mpg imperial, 29mpg US) highway cruise, around town is only slightly heavier at 9.4l/100km.
Interesting side effects, the cam is not as noticeable either at idle or driving, there is no back fire/spitting like Webers, it's quieter and idles at 900rpm instead of 1200rpm.
Still haven't put the time into getting the EGT into use.
I can't thank Jean enough for the P&H boards used with the lo-Z injectors. Side by side testing of this set up against a stock 280zx using a digital temperature meter shows that using the P&H boards results in injectors running at least 40 degrees Celsius cooler!
Last edited by aarc240 on Thu Nov 15, 2012 11:25 pm, edited 1 time in total.
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby jbelanger » Mon Nov 12, 2012 9:16 pm

That's a good improvement in injector temperature! And I'd say that the torque and fuel consumption improvements are definitely worth the minuscule drop in peak power.

The better behaviour and fuel consumption are almost certainly due in good part to EFI but I wonder how much of the power curve change is due to the different intake. Do you have pictures of the new intake? Longer runners and a plenum would definitely push the power band to a lower RPM.

Jean
Image
jbelanger
 
Posts: 3578
Joined: Sat Oct 03, 2009 12:24 pm
Location: Quebec, Canada

Re: MS3 + P&H + VR - another way

Postby aarc240 » Tue Nov 13, 2012 7:42 am

The intake manifold is a ported 280z factory item.
The lower floor has been gutted and the EGR crap sealed off (as in welded shut) to get the largest plenum volume possible in a stock looking manifold.
Runners are smoothed and enlarged a little but mostly to port-match with the aim of keeping air speed high. They are pretty close in length but smaller volume to the IR manifold of the Webers.
The adapter from 65mm throttle body down to the 57mm manifold is unfortunate but the EGT's are telling me that it may actually have increased flow to the rear cylinders sufficiently to nearly equalise that. Maybe forcing a velocity increase at WOT was a good thing!
Definitely worth sacrificing a little top end, it accelerates like a scalded cat now!
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm

Re: MS3 + P&H + VR - another way

Postby aarc240 » Sun Aug 04, 2013 6:00 am

A few progressive and very worthwhile changes.
Manifold is now a Nissan NAPS type FI unit (originally used staged throttle plates) similarly gutted like the old 280z one and has a 40% increase in plenum volume resulting in a noticable improvement all the way through the range. Doing the TB adaptor was a bit tricky.
MS3x installed to facilitate sequential ignition with LS2 truck coils which has given much smoother idle and off idle acceleration.
Still using lo-Z injectors with P&H boards inside the case although that will soon change with a move to a larger case to enclose everything.
Subsequent to all that the engine has been pulled from the 260z and dropped into the 240z trials car.
Lost 140kg (now in a 960kg car) plus a 4.1 diff with the same tires. Not too many cars can get away from it on gravel!
aarc240
 
Posts: 95
Joined: Sat Feb 19, 2011 5:01 pm


Return to General

Who is online

Users browsing this forum: No registered users and 1 guest

cron