View unanswered posts | View active topics It is currently Sat Nov 28, 2020 11:45 pm



Reply to topic  [ 34 posts ]  Go to page Previous  1, 2, 3, 4  Next
Separating the power and signal grounds on a V3 board 
Author Message
LQFP144 - On Top Of The Game

Joined: Sun Jul 13, 2008 2:58 pm
Posts: 331
Location: South Wales, UK
it may also be worth noting that the regulator U5 is bolted to the heatsink bar and requires insulating with a mica pad.

The battery voltage input to AD4-1 is taken directly from the collector of Q12 which introduces loads of unwanted noise even when not using the PWM mode, i dread to think what its like with Lo-Z injectors.
I rigged up a filter for the battery voltage reference as you don't need it to measure the noise, a 220uF cap from ground to the V12 side of R3 will smooth the input but for extra filtering cut the track from R3 to Q12 collector and place a 750R resistor from pin 1 of U5 to this new cap. This will make a low pass RC filter with a 6dB/oct above 1Hz.
Accessories switching on in the car with a low enough impedance to have any effect on Vbatt will still be compensated for but the spurious switching noise will be gone.

the heatsink bar is grounded by the main ground plane, I'm looking at insulating all the TO220 tabs, cutting the spider tracks to the grounded side of R37 & 38, cutting off the middle legs (collectors) on Q3 & 11 and using the tabs to connect to a second ground bar running on the underside.
This can then be connected via a very thick cable to a proper chassis earth.
Making sure obviously to ensure that Q9 & 12 are properly isolated as the tabs of these are connected to 12V.

I will need to ensure that the bolts connecting the tabs to the new ground bar are insulated from the top bar and the through plated holes in the PCB.

the ground sides of R37 & 38 will be reconnected to the new dirty bar either by wire and tags or if i can find that nice copper bar I had around here I may solder them.
will post pics soon. ;)

_________________
93'BMW 325is M50B25TU, Rebuilt 06/06, JE10.5:1, polish&port. Scorpion BB, K&N CAI, TEJ21 WBO2, '07 M3 Evo 18" 225F, 255R, EBC Kevlar, Bilstien Sprint, Polyflex. Head rebuild Oct'08, OEM+FSE FPR, MS2v3.0_DJB Custom, Extra 2.0.1


Mon Aug 25, 2008 11:32 am
Profile
Moderator
User avatar

Joined: Tue Jan 15, 2008 2:31 pm
Posts: 15371
Location: Home sweet home!
Good on you Dave! I'll be interested to see a master plan for all v3 users posted up at some point :-)

_________________
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!


Mon Aug 25, 2008 3:48 pm
Profile WWW
LQFP144 - On Top Of The Game

Joined: Sun Jul 13, 2008 2:58 pm
Posts: 331
Location: South Wales, UK
OK as Fred requested I have come up with a master plan to get all of the dirty noisy crap off of the power rails used by the MCU and sensitive stuff.
so here goes....

First we need to split off the high power ground return and provide another way for the injector low side mosfets to get the current back to ground.
to do this i chose to make a second heatsink bar on the back and strap it via very thick cables back to a solid chassis earth or preferably the battery.
so I got out my trusty plasma cutter (hacksaw will do) and pretty much duplicated the top bar only using slightly thicker ally (3mm).

Image

Image

I used double sided tape on the underside for insulation and ease of fitting.

Next after sticking the new dirty bar to the PCB, i fitted the appropriate fixings to hold the power devices to the bars, you will notice in the shot below that I have 2 long bolts and 2 short bolts. Check these wont foul against the case as it is ground!!!
The 2 short bolts are connecting the Collector tabs of the TIP42 transistors to the new dirty ground bar. The 2 long screws are electrically connected to the collector tabs of the TIP125 transistors these should be isolated from both heatsink bars as these will be connected to the new dirty 12V supply.
it would be quite unwise to allow any shorts to ground here! make sure your holes are drilled cleanly and any swarf removed.

to properly insulate the screw from the new dirty bar, use insulating kits designed for TO220 devices but for the TIP125's you need to improvise a little.

Image

you need 2 insulating "tophat" washers per TIP125 screw, the first one (black in this shot) needs to be mutilated to make a sleeve for the threads as shown below.

Image

Wind the screw all the way in (M3 are quite snug) and fit the other "tophat" to the under side between the dirty bar and the screw, hold that in place with a washer and nut and for the electrical connection to lug or bar use another 2 washers and a lock nut to clamp down, as shown below.

Image

Now you have all your new dirty connections solidly in place and waiting for heavy duty lugs, you can go about cutting the middle legs off the TIP42's and TIP125's also you want to cut the right legs of the current sense resistors. Do not cut the lead flush with the package, leave the thick section intact. once cut you can ether trim them flush with the PCB or remove them totally.

Image

now you have the problem that the ground side of the current sense resistors has nowhere to go so you fit a couple of links between the right pin of the resistors to the remaining stub of the TIP42 middle (collector) pins. This now takes the current flow to your dirty bar on the back of the board instead of the ground plane that is use by everything else.

Image

As you can see there are 2 links fitted, I advise using insulated fairly thick wire as it carries up to 14 amps and runs under the CPU card (you don't want any shorting!)

Now on my setup I had strange injector behavior where every few seconds the injector pulse width would drop to virtually nothing and then begin to fire as normal. Using the real time display screen in megatune i could see the battery voltage fluctuating rapidly reaching 17 volts, it was when the voltage peaked that the injectors turned off briefly.
To filter out this crap i cut the track suppling VBatt to R3:

Image

and put an RC filter using 750R and 220uF. The result is a minuscule voltage drop of 0.1V and it filters out anything above 1Hz at 6dB per oct, this will still allow compensation for electrical accessories turning on and off, but will stop erratic behavior.
All the components are mounted on the underside of the PCB, I get the 12V from pin 1 of U5 and the ground from the proto area. the connection in between the R and C is the topside of R3.

Don't be tempted to just put a big cap across the ADC input as it is used by the bootl oader on MS2 and will trigger it on bootup.

Image

Now as some of you may be aware I have a Bosch 3 wire PWM idle valve, this requires a push pull drive. I initially used 1 channel of the stepper motor chip on MS2 to provide this drive. after looking at the nasty crud coming back on the ground because of it. I have decided to put in a couple of mosfets on the new dirty bar to separate the noisy PWM valve and sink it back to dirty ground.

To get the PWM signal into the stepper chip I lifted Pin 3 and strapped it to pin 34 of the dip 40:

Image

the output of that 1 channel are naturally push pull, being lazy I just used those 2 outputs to drive the mosfets via a couple of 820R resistors (in the heatshrink tubing):

Image

You will notice a red wire running under the bar that is to beef up the supply to the sanctioned 12V it runs to the now unused TIP125 pad.

For the positive supply (overkill bar) i wanted to use this :P :

Image

but the actual power requirement here is very limited on Hi Z injectors, so i will use some modest sized lugs, that I will acquire from work tomorrow and post up some pics.

If you have any questions or have noticed something i may have missed please let me know, I have tested on the bench and it all works fine and it is electrically separated so in the car it should be hunky dory! ;)

_________________
93'BMW 325is M50B25TU, Rebuilt 06/06, JE10.5:1, polish&port. Scorpion BB, K&N CAI, TEJ21 WBO2, '07 M3 Evo 18" 225F, 255R, EBC Kevlar, Bilstien Sprint, Polyflex. Head rebuild Oct'08, OEM+FSE FPR, MS2v3.0_DJB Custom, Extra 2.0.1


Tue Aug 26, 2008 12:32 am
Profile
Moderator
User avatar

Joined: Tue Jan 15, 2008 2:31 pm
Posts: 15371
Location: Home sweet home!
http://forums.hybridz.org/showthread.ph ... 713&page=5

Popular thread apparently :-)

Well done Dave and GearHead!

Fred.

_________________
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!


Wed Nov 12, 2008 2:36 pm
Profile WWW
LQFP112 - Up with the play

Joined: Sun Feb 03, 2008 9:30 pm
Posts: 120
Location: Chicago, USA
what about the anodes of D6 and D20? Shouldn't these be pulled off the ground plane as well?

Also, if you do not install the current limiting stuff (Note 1 on the schematics) or the PWM stuff (Note 3), the modifications become even less complicated or invasive. You do not have R37 or R38, nor do you have Q9 or Q12 (the TIP125s). I do not install the current limiting stuff anyway to ensure rapid opening times nor the PWM stuff b/c I run HiZ injectors.

Gearhead


Wed Nov 12, 2008 9:40 pm
Profile
LQFP144 - On Top Of The Game

Joined: Sun Jul 13, 2008 2:58 pm
Posts: 331
Location: South Wales, UK
gearhead wrote:
what about the anodes of D6 and D20? Shouldn't these be pulled off the ground plane as well?

Also, if you do not install the current limiting stuff (Note 1 on the schematics) or the PWM stuff (Note 3), the modifications become even less complicated or invasive. You do not have R37 or R38, nor do you have Q9 or Q12 (the TIP125s). I do not install the current limiting stuff anyway to ensure rapid opening times nor the PWM stuff b/c I run HiZ injectors.

Gearhead



Not really necessary as they are for ground referencing only, the current path is through the transistor and down to the dirty ground.

This works perfectly even when driving Lo-Z injectors, the kick back used to produce up to 20 Volts on the same rail that is monitored for V Batt compensation! even worse because of the current flowing back through the standard ground system the ground on the board would float under load to 2 volts above actual battery -ve.

With the above modification/upgrade both digital and dirty grounds are solid and the battery monitoring is accurate to 0.3V, the chances of getting the MS2 working out of the box for the first time improve dramatically as you are just dealing with the variables you wan't not spurious noise.

So if you have a v3.00 PCB I highly recommend you separate your power rails, grounds and filter the V Batt monitor line, after all you want the damn thing to work for the £200/$400 you paid for it!

_________________
93'BMW 325is M50B25TU, Rebuilt 06/06, JE10.5:1, polish&port. Scorpion BB, K&N CAI, TEJ21 WBO2, '07 M3 Evo 18" 225F, 255R, EBC Kevlar, Bilstien Sprint, Polyflex. Head rebuild Oct'08, OEM+FSE FPR, MS2v3.0_DJB Custom, Extra 2.0.1


Fri Nov 14, 2008 11:38 am
Profile
TO220 - Visibile

Joined: Mon Nov 17, 2008 12:35 pm
Posts: 7
Hi,
Really really appreciate you hard work and sharing.
I am new in this forum, but have follow the above instruction to modify my V3 board.
What I discover is that if I attached the Battery Voltage sensing on pin1 of U5, the MS2 could be always frozen.
But If I attached it to the V12 directly, everything will be fine.
I have spent one day for checking any short or my chassis harness, nothing abnormal found.


Mon Nov 17, 2008 12:45 pm
Profile
LQFP144 - On Top Of The Game

Joined: Sun Jul 13, 2008 2:58 pm
Posts: 331
Location: South Wales, UK
The V Ref line is normally connected to +12V via a potential divider 50K on the top and 10K on the lower half.
all i have done is use a resistor in line and a cap to smooth the fluctuations the output of which connects to the 50K of the potential divider.
Are you talking about connecting the MCU pin directly to U5 pin1?

_________________
93'BMW 325is M50B25TU, Rebuilt 06/06, JE10.5:1, polish&port. Scorpion BB, K&N CAI, TEJ21 WBO2, '07 M3 Evo 18" 225F, 255R, EBC Kevlar, Bilstien Sprint, Polyflex. Head rebuild Oct'08, OEM+FSE FPR, MS2v3.0_DJB Custom, Extra 2.0.1


Sat Nov 22, 2008 12:24 am
Profile
TO220 - Visibile

Joined: Mon Nov 17, 2008 12:35 pm
Posts: 7
Find out the real root of my fault.
The connection is exactly right.
The reason of MS-II frozen is that :
I set up. 1 squirt per cycle, 1 cylinder, and then chose Missing Tooth Trigger.
The equation on the MS-II code will have problem in calculation.


Tue Nov 25, 2008 3:57 am
Profile
LQFP144 - On Top Of The Game

Joined: Sun Jul 13, 2008 2:58 pm
Posts: 331
Location: South Wales, UK
calkai wrote:
Find out the real root of my fault.
The connection is exactly right.
The reason of MS-II frozen is that :
I set up. 1 squirt per cycle, 1 cylinder, and then chose Missing Tooth Trigger.
The equation on the MS-II code will have problem in calculation.


Ahhhh! now I understand. :)
So you aren't having difficulty running with my modification on your V3 PCB and the problems you are seeing, are related to software/configuration settings.

Thats good I was getting worried there for a second :lol2: I found that after the mods the noise detected by the MCU was reduced from several volts of noise to a couple of mV. The difference to the ease of setup and tuning was well worth the work.

_________________
93'BMW 325is M50B25TU, Rebuilt 06/06, JE10.5:1, polish&port. Scorpion BB, K&N CAI, TEJ21 WBO2, '07 M3 Evo 18" 225F, 255R, EBC Kevlar, Bilstien Sprint, Polyflex. Head rebuild Oct'08, OEM+FSE FPR, MS2v3.0_DJB Custom, Extra 2.0.1


Fri Nov 28, 2008 8:17 pm
Profile
Display posts from previous:  Sort by  
Reply to topic   [ 34 posts ]  Go to page Previous  1, 2, 3, 4  Next

Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software for PTF. ColorizeIt.