E46Fanatics

E46Fanatics (http://forum.e46fanatics.com/index.php)
-   DIY: Do It Yourself (http://forum.e46fanatics.com/forumdisplay.php?f=27)
-   -   Any Electronics designers want to assist in a VERY cool mod? (http://forum.e46fanatics.com/showthread.php?t=772701)

Nachos23 06-19-2010 02:00 AM

Any Electronics designers want to assist in a VERY cool mod?
 
Some of you may have seen my previous posts in regards to retro fitting a DCT shifter from a 2010 650i to my Steptonic e46. http://forum.e46fanatics.com/showthr...381&highlight=

With my black interior, and black CF trim, should look sweet! Not my car; someone that tried and almost got it there......

http://i905.photobucket.com/albums/a...i/DCTine46.jpg

BMW's current lighting system on the e46 uses a common power, switched ground design. PRND LEDs are lit when the selector switch closes the circuit to ground.....

See schematic below.

http://i905.photobucket.com/albums/a...onicSwitch.jpg


The challenge is that the DCT shifter uses a common ground, switched power design to illiuminate PRND LEDs. I've traced and determined the pin connections for the DCT shifter and got ambient lighting on the shifter to operate through car dash lights. Got green LEDs to illuminate by applying Dimmer power directly to applicable pin and varying intensity (best appearnce @ around 6vdc; don't know spec of LED installed).

http://i905.photobucket.com/albums/a...lumination.jpg

Where everybody previously had failed is getting it to operate in the car.

The circuit I have in mind would use the switched ground connections of the current design to trigger a micro relay or opto coupler which in turn would provide approx 6vdc to the LEDs (drop voltage across resistor). Replicate circuit 5 times PRND and M/S. Simple design and I am an electrical engineer by trade, but been out of industry for close to 20yrs and sorta behind on mre technology available to me. Spent evening browsing Electrosonic catalogue and think I've got the relay identified....

Just need some assistance in fine tuning and double checking to ensure the components selected are appropriate for application.

Just a little of your time to help make this work........ Any takers? :idea:

frogg320 06-19-2010 03:00 AM

Isn't this what transistors are good at?

Nachos23 06-20-2010 06:37 AM

Transistors are being used to close the circuit, but how to apply to use as an isolator? Don't know of they'll sink enough current to drive the LED..... THat's what I'm hpoing someone on this forum can assist with.....

Arty 06-20-2010 07:00 AM

PNP should do it.

http://www.theowlnest.com/PNP-NPN.jpg


http://i88.photobucket.com/albums/k1...s/650gear2.jpg
The 10k Pull up is to make sure the transistor base comes up above the emitter when the switched GND is not there. TBH even without it the base will probably float up to the emitter voltage.

Edit:

1) rinse and repeat the circuit for the other LEDs.
2) The pull up will stop any noise accidentally activating the LEDs. Unlikely to happen on such a short wire in a sparesly populated area.

Edit 2:

The Bentley manual looks wrong compared with the OP's description of how the cct behaves. It says in the diagram that a voltage below rail 15 is needed on pins 2-5 to turn on the LEDs.
That appears to be the E46 schematic.
This 6er schematic has replaced it now.

howie1971 06-20-2010 10:18 AM

I am sure you gave seen this from Delmarco but if not it might help http://forum.e46fanatics.com/showthread.php?t=760259

You sound like you know what you are doing so I am sure you will work it out.

Howie

Nachos23 06-21-2010 11:24 PM

Quote:

Originally Posted by Arty (Post 11898712)
PNP should do it.

http://www.theowlnest.com/PNP-NPN.jpg


http://i88.photobucket.com/albums/k1...s/650gear2.jpg
The 10k Pull up is to make sure the transistor base comes up above the emitter when the switched GND is not there. TBH even without it the base will probably float up to the emitter voltage.

Edit:

1) rinse and repeat the circuit for the other LEDs.
2) The pull up will stop any noise accidentally activating the LEDs. Unlikely to happen on such a short wire in a sparesly populated area.

Edit 2:

The Bentley manual looks wrong compared with the OP's description of how the cct behaves. It says in the diagram that a voltage below rail 15 is needed on pins 2-5 to turn on the LEDs.
That appears to be the E46 schematic.
This 6er schematic has replaced it now.

Perfect. I was worried about current draw and was looking @ using an inverter connected to a CMOS Bilateral switch (4066), but sounds like overkill based on your mods. The e46 schematics don't actually show the LEDs (couldn't frind them anywhere on WDS or Bently's), but they are pulled done by the transistors. I know it shows a 15vdc rail (actually battery voltage), but the part of the schematic that's missing is the circuitry that leads to the LEDs. I measured the vcc of the LEDs and it's around 2vdc (current design).

When open, the ground floats anywhere upto 7vdc (typcially around 2vdc), when closed; about .7vdc..... How much current can a std tx source (LEDs apparently like around 20mA)? I hooked up the LEDs to the dimmer power and ran from full power right down to 0vdc. They looked best @ around 6vdc.... Didn't measure current draw; should havem but was getting late and quite dark.....

Quote:

Originally Posted by howie1971 (Post 11898870)
I am sure you gave seen this from Delmarco but if not it might help http://forum.e46fanatics.com/showthread.php?t=760259

You sound like you know what you are doing so I am sure you will work it out.

Howie

Delamarco and i have been talking on this since he started (my Sig is in his credits) and we got it working on the bench @ that time. It was my interest in his DIY that got me started. Problem is; that NO ONE had gotten it to work off the bench and installed in the car. One of the other members up here and I have have spent considerable time tracing out the circuit and determining pin assignments (which BMW dealer couldn't find although there was a 2010 650Ci sitting in the showroom with this shifter installed.... LOL). Delmarco's work to retro the actual shifter has been extremely helpful in many aspects of the project. Gave me alot of great ideas. Going to use "P" button on top to replace current mechanical shiftlock, with electronic shiftlock; addressing numerous safety concerns....

Nachos23 06-21-2010 11:56 PM

looked more carefully @ your schematic :eek:; feel like :banghead:. I used to pop something like this off in a few minutes and here I've been banging my head to create such a simple circuit to perform the task. I'll check a couple of Trans specs to make sure it'll sink the required current (not sure what that is yet; will measure); it'll be based on visual appearance of the LEDs during day and night operation. Might install variable resistor initially so I can play with voltages... Ambient lighting for the display will be hooked to dimmer lighting.

Going to run electronic shoft lock from brake in series with "P" switch to provide secondary electronic locking so the shofter cannot be inadvertantly pushed into the wrong gear. It will require that you press both the Brake and "P" button to release electronic shoft lock and change gear. Was concerned if in an emergency braking situation that the shifter (with current mechanical shiftlock no long present) could be pushed into R or P because the mechanical shift lock is gone (DCT Shifter is fully electronic). Nice eh?? :woot:

Your design is great, but it actually has to interface the e46 shift selector to operate the lights in the DCT shifter (so not quite where you placed the circuit, but PERFECT DESIGN!!! :thumbup:).... The Cathode connections will actually connect to the Gear Selector lighting pins as follows:

1 P
2 M/S
3 D
4 N
5 R
6 ???
7 ???
8 P switch
9 P switch
10 Grnd
11 Power (dimmer)
12 Hall sensor (intended electronic shift release; but I prefer the P button)

Since I'm obviously :excited:, any recommendations on an appropriate PNP model to use; or just a standard low power transistor?

Thanks for all your help BTW!!! :idea:

Nachos23 06-22-2010 12:19 AM

Sorry about the cellphone pics..... but here's some of the intial fitting pics. The reason for the covered PRND lights is to "visualize" what it will look like after it's finished.

http://i905.photobucket.com/albums/a...erfitting3.jpg

http://i905.photobucket.com/albums/a...erfitting2.jpg

http://i905.photobucket.com/albums/a...erfitting1.jpg

Thoughts?

Arty 06-22-2010 05:25 PM

Quote:

Originally Posted by Nachos23 (Post 11905356)
l I'll check a couple of Trans specs to make sure it'll sink the required current (not sure what that is yet; will measure); it'll be based on visual appearance of the LEDs during day and night operation.

Any old transistor will be ample. BC327 or something. The junction on that will be able to sink up to 0.625W or 800mA. The LED's in the knob look like 20-50mA jobs. No problem at all for a common PNP. You may even be able to get a multi-device single part and have less of a rats-nest in there.

Quote:

Originally Posted by Nachos23 (Post 11905356)
Might install variable resistor initially so I can play with voltages...

I'd say a max of ~5V is a good idea. They will look much brighter in the car at night than on the workbench, and will last longer.
Quote:

Originally Posted by Nachos23 (Post 11905356)
Your design is great, but it actually has to interface the e46 shift selector to operate the lights in the DCT shifter (so not quite where you placed the circuit

From your first post, I understood there was a unique ground connection made for each position of the gear lever, and that is shown in the diagram as the switches marked PRND. As each switch is closed as you move the gear lever, it grounds that individual transitor base, turning on the 5V to the knob for the LED of your choice.

All the others are pulled up to 5V by the 10K resistor pull up, turning them off. This is because just (5V-0.6V)=4.4V on the base will light the knob LED. Although you say the line floats to 7V, so you may not need the 10K, as touched on in post 4.

Arty 06-23-2010 01:09 PM

http://i88.photobucket.com/albums/k1...s/650gear3.jpg

Nachos23 06-27-2010 01:54 AM

Hey guys. Got the circuit working on a breadboard. Playing with voltages, etc to make it look the best day and night. Using PN2222A transistor to convert the open ground of the e46 lighting to a switched power; common ground required to power DCT shifter.

Going to get leather professionally stitched @ my tailor to close opening in front made to accomodate signifcantly larger DCT shifter. Going to cover-up the current PRND M/S lights on console using that spray on plastic stuff (can't remember name). Then assemble original circuit board and inverter circuit board into a small enclosure which will mounted to left of shift selector under console.

Hopefully have it done tomorrow (Sunday)........ Will post more pics when it's further along.

firefirefire90 06-27-2010 01:58 AM

looking forward to the progress!

Arty 06-27-2010 09:59 AM

Quote:

Originally Posted by Nachos23 (Post 11926635)
. Going to cover-up the current PRND M/S lights on console .

In that case, if you are ditching the E46 LEDs {and if they are not used for the dash} you can simplify.


http://i88.photobucket.com/albums/k1...of650gear2.jpg

profor 06-27-2010 08:01 PM

Check if LEDs are not fed by PWM voltage. You mentioned around 6V which is not normal in car and this reading YOu can get with normal multimeter and PWM signal.

Nachos23 06-27-2010 09:01 PM

Quote:

Originally Posted by Arty (Post 11927164)
In that case, if you are ditching the E46 LEDs {and if they are not used for the dash} you can simplify.

http://i88.photobucket.com/albums/k1...of650gear2.jpg


Actually, before I ohm'd out the circuit, that was what I was hoping to accomplish, however, the M/S LED is also open ground and comes from connector X1600 (white one) see schematic above. Goes from transmission control unit to steptronic switch through lighting circuit board. It's that signal that has me thinking I need the interface board.... But could use a single transistor for just M/S LED if your suggestion for shift selector works. Need to verify how the switch is located/functions... ie completely isolated as shown in schematic, or does it connect to chassis ground where mounted? Need to verify visually and electrically. If it's isolated, maybe I can use your suggestion for selector switch and use single PNP transistor to power M/S LED...... Pouring out there tonight, and RFP due Tuesday night, so probably will check Wed night and report with hopefully a functional shifter.....


Quote:

Originally Posted by profor (Post 11928625)
Check if LEDs are not fed by PWM voltage. You mentioned around 6V which is not normal in car and this reading YOu can get with normal multimeter and PWM signal.

Well, accoridng to the current e46 schematic its fed by 15vdc. What I can't find on ANY schematic is the ton of additional circuitry that exists on the current lighting circuit board, that include, coils, caps, diodes (in addition to LEDs), etc..... They must drop voltage down. Also, on the current design, there is both CanH (15vdc) and dimmer voltage being fed into board. New DCT shifter has only 1 connection for Ambient lighting on shifter plate and it must be connected to dimmer voltage..... There is no ambient lighting being fed to DCT shifter when headlights/running lights off. Only illumination during day is green PRND M/S LEDs..... They do not seem to be impacted by ambient lighting circuit (separate powers; common ground).....

Nachos23 06-29-2010 04:34 PM

Transistors aren't pulling voltage up enough. I need about 5-6vdc for correct illumination. When I tested the circuit using PNP (2n3906) the Emitter voltage was 0.7 vdc when forward biased (turned on) which is matched @ Collector..... Is there a way to "pull up" the voltage @ Collector to 4-5vdc? If I could isolate grounds on LEDs problem is easy. Its providing the 4-5vdc that's seems to be eluding me...... Could use relays, but looking for a lower current drawing solution.... or Bilateral CMOS gates, but those need to be driven by a TTL inverter (now voltage reg would be required)...... Driving me crazy. Going to try with Micro relays I think; just got to find time to buy (not easy to find)...... thoughts?

Arty 06-30-2010 01:43 AM

When you ground the base, the emmiter-base voltage is -5V. This means the transistor is biased hard on. The junction os saturated. This has the effect of connecting collector to emitter, with like about 0.1V drop. The 5V present at the emitter is present at the collector. With only -0.7V of b-e voltage, the transistor will be Biased class AB and ready for action but not fully saturated, which is where you want to be.

About the PWM, it is a good point, and another reason not to use a 6V rail as the life of the LED will be exponentially reduced as it is overdriven.

Nachos23 06-30-2010 08:25 AM

what ends up happening is the collector is pulled to 0.7 vdc when I close the switch (selector lever) and ground base; its Collector voltage is 5vdc when transistor off (emitter @ 0 vdc). The emitter voltage matchs the collector voltage @ 0.7vdc (which is not enough to power the LED; testing has shown I need about 4.5vdc to get LEDs to light). How can I pull-up the collector voltage to 5vdc..... I'm missing something basic. tried resistor in series on base signal; no effect..... Now I know why I took my degree and went into IT..... all 1s and 0s.... LOL I'm still designing digital circuits, it the analog crap thats got me stumped.

Also, since you obviously know what you're saying, I could go with your previous suggestion in regards to isolating selector switch and applying power to it, but how whould you handles the M/S signal coming from TCU? X1600 is a 3pin connector and pin 2 seems to control M/S LED, however pins 1+3 seem to go through the lighting panel and pass through a diode. not sure if I can just jumper pins 1+3 using a diode, or plain jumper wire. If I'm to eliminate the PRND circuit brd, then I'd have to address that signal..... Tomorrow is Stat holiday in Canada, guess I'm going to have to test on my car and find out.

howie1971 06-30-2010 10:01 AM

I admire your tenacity Nachos23 it sounds frustrating. I am following this thread with interest as I am interested in retrofitting the DCT or M3 shifter. Its a bit beyond me so I was looking to purchase the mod. I am not sure if it helps but the guy at the aftermarketbulbshop.com (I think they are a sponsor) said he can do both on an E46.

However although the lights work its still a mechanical shift. So do I understand it right that you are trying to make the PRND electronic as well? I have just bought the paddleshift wheel/mod for the steptronic so I am excited by that but what I would really like is to be able to engage P when sitting at the lights via a button. The P button on the DCT shifter would be ideal so I hope you get it working. I have asked a couple of people whether they can do it but no success yet.

Goodluck
Howie

Nachos23 06-30-2010 03:08 PM

Quote:

Originally Posted by howie1971 (Post 11939593)
I admire your tenacity Nachos23 it sounds frustrating. I am following this thread with interest as I am interested in retrofitting the DCT or M3 shifter. Its a bit beyond me so I was looking to purchase the mod. I am not sure if it helps but the guy at the aftermarketbulbshop.com (I think they are a sponsor) said he can do both on an E46.

However although the lights work its still a mechanical shift. So do I understand it right that you are trying to make the PRND electronic as well? I have just bought the paddleshift wheel/mod for the steptronic so I am excited by that but what I would really like is to be able to engage P when sitting at the lights via a button. The P button on the DCT shifter would be ideal so I hope you get it working. I have asked a couple of people whether they can do it but no success yet.

Goodluck
Howie

Thanks Howie;

The M3 shifter is easy. same circuitry as our e46's..... I thought the DCT would be similar; that's where I was wrong. LOL.....

I've purchased small micro relays this afternoon and will try tonight or tomorrow to finish project. I also am going to do some physical inspecting of current mechanical and switch setup. Depending on how the M/S switch functions, this could be an easy mod *and I've spent alot of uneccessay time). The Steptronic connector has 4 pins on it. I know 2 are to identify upshift and downshift (hall effect sensors; BMW loves them). The interesting part are the other 2 pins that seem to go through lightinng panel and then on to ground. Hoping I can bypass that connector and then just run power to LEDs; nothing to it.... Don't think it'll be that easy, but if I can bypass the lighting panel and go direct from connectors, then I can just feed power to connectors through resistor bridge. going to try and figure this all out and come up with the easiest solution......

As far as ANYONE doing this sucessfully; NOPE I'm gonna be the first!!! (then make retro-fit kits)

Going to use the P button to REPLACE mechanical shiftlock with Electronic shiftlock. The current mechanical shiftlock is the button we press to release the gearshift. There is no such mechanical lock on the DCT as it is pure electronic. It comes with a electronic RELEASE button on side of shifter, but I'm going to use the P button instead; prefer it's placement.

Today we need to press both mechancial shiftlock and brake to release shift lever; I'm gong to force using the brake and P button to release sift lever...

To hook-up paddle shifter, connect the 2 wires from the paddles to the middle 2 wires on steptronic connector; easy; if backwards, reverse wires thats it.


All times are GMT -5. The time now is 06:28 AM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.
(c) 1999 - 2011 performanceIX Inc - privacy policy - terms of use