Excel seat alternative - BMW

Post discussions on potential upgrades and modifications here, as well as documenting successfully implemented projects.

Moderator: Board Moderators

Post Reply
User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Excel seat alternative - BMW

Post by Excel SA »

I am replacing the seats in my Excel with those from a 2014/2015 BMW 2 series - F22 model. The BMW is a 2-door, 4 seater - the same as the Excel so comes with front seats allowing access to the rear ones.

Below are pictures of the front and rear seats from the BMW - on the seats I bought, the seat base moves forwards and backwards, up and down and can tilt front to back, the front edge of the seat base can be extended forward. The backrest angle changes as expected and the side "wings" of the backrest can be adjusted for width.

Image

Image

Things to look out for when buying seats

I missed a few things when I bought my seats - sometimes it's hard to pick up things when you're looking at something new....I was also thrown a bit by the fact that the seats were all electric, something I had not been expecting.

Airbags - these seats have airbags and seat belt tensioners. If the car has been in an accident these may have been activated. On the seats I bought, the driver's side ones had deployed. I have removed the seatbelt clip and tensioner on my seats, they are in the way, and I'm not sure the Excel belt will clip into them. The rubber around the seat clips stalk was obviously compressed once you looked for it - an easy thing to spot in hindsight.....the airbag itself is a different matter - the panel on the outer side of the seat back seems to get damaged and needs to be replaced. Cursory looks had made me think that the panel was just dirty, but after cleaning and a more thorough inspection it has been replaced - the leather colour is quite different and the stitching is not exactly up to factory standard. If the repair had been done better, this wouldn't have been a problem.

Some parts of the seat rail had also been bent a bit, the seats still moved okay, but I managed to straighten the bent bits out fairly easily with the help of a vice and some leverage.

I had negotiated quite hard on the price, so although missing the above is frustrating, the price was right, but I'd be more careful the next time....


Front seats - fit and installation

The seats I bought are fully electric - that wasn't what I was looking for, but a decent condition interior in red leather that will fit and be useful in an Excel doesn't become available too often. Manual seats may be different in some ways.

The front seats fit in the car, but fairly snugly - there is at best maybe 5 to 10mm of space on the sides of the seats, but the backrest side bolsters appear to be rubbing on the transmission tunnel a bit - this can change depending on how far back/forward the seat is positioned and how the rails are installed.

Height-wise things are also tight. If the seat is set back, the backrest can fold forward without a problem, but as the seats are moved forward, they also rise up (as the seat rails are sloped) and do get to a point where they won't fold forward fully as they catch on the roof. The head rest could be altered to fix this but would be some effort. Edited - I have found that the head rests can be lowered another 20mm or so without much work and keeping them standard - this seats them directly on top of the seat back - I will detail this later.

The height of the seat base is fairly similar to the original seat I think - my original seats are in pieces, and the suspension was torn on both of them. When I sit on the BMW seats, my head is very close to the roof - I seem to recall I had the same sensation with the old seats, but that was some time ago. I'm just under 6 foot, so if you're taller than that, it could be an issue. The seats I got are height adjustable, this is with the seats set at their lowest possible position. Once again, as you move the seats further back, headroom increases. I will make an effort to install the seats as low as possible.

Seat rails - the seat rails on the BMW seats I have are set closer together than the Lotus ones - about 20mm. The rails are also longer (110mm), and the mounting holes are further apart by 35mm. I have managed to get the seats installed using some fairly basic brackets, and a little fettling. I can get the full range of movement in the seats from far back to far forward where the narrowing of seat tub area becomes a problem.

Electrics

Manual seats will be exempt from this bit - there is a computer for each seat that needs to be dealt with to get the seats working, but it can be done according to some Google research. It is possible to easily test the seats by applying power to the main 2 wires, but this stops working about 5 seconds after power has been applied, and/or the seat adjustment buttons haven't been used. There are 5 wires going into each seat - 1 main power, 1 earth, 1 power feed for the seat back width adjustment and 2 data wires.

Weight

These electric seats are quite heavy - about 24.5kgs each, after removing the seat belt clip and tensioner. I believe the manual seats are a bit lighter, but not substantially less - levers and linkages replacing motors. The Excel seats are around 15kgs per side.

The next few posts will go into more detail on exactly what I've had to do to get the seats in and working.

Neil.
Last edited by Excel SA on Thu Aug 18, 2022 08:49, edited 1 time in total.

User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Re: Excel seat alternative - BMW

Post by Excel SA »

Removal of the seat belt clips and tensioners

I have edited the first post, and may do more - any changes I've done are in blue.

Below is a pic of the seatbelt clip and tensioner - this tensioner has deployed. Removal is by loosening the nut on the back of the Torx bolt circled in yellow on the right. There are two Torx bolts holding the runner to the frame - neither of these Torx bolts can be turned, one is the yellow circled one on the left, the other is behind the bracket - they are keyed into the seat frame - turning them will break things! The nut on the back must be loosened! The red arrow points to the bracket that the clip and tensioner are attached to.

Image

Below is a picture of the bracket - you need to remove this runner from the seat to get access to the head of the Torx bolt circled in orange fixing the bracket to the runner. The second rail to seat frame fixing is in the yellow box. The bracket does stick out to the side, and as space is at a premium, needs to be removed. On the electric seats there is a plastic bracket holding the forwards/backwards motor running between the two rails that needs to come out as well - this is held in with two plastic, expanding rivet type clips on each side.

My cheap Torx socket twisted trying to undo the bolt holding the bracket on, I ended up drilling it out - it drilled surprisingly easily, and the square nut in the runner was also removed.

Image

The wiring for the tensioner runs into a plug that slides into another plug on a bracket under the seat, and can be removed easily - there are two similar plugs, the other I expect is for the airbag.

Neil.

User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Re: Excel seat alternative - BMW

Post by Excel SA »

Installation

As the seat rails are different to the Lotus, I needed to make some brackets to install the seats into the car.

The front outer mounting hole was used without a bracket (front outer being on the door side of either seat). I did choose to elongate the hole to move the track slightly further out towards the door - there is a bevel on the edge of the hole in the bobbin that I widened the hole to.

Before:
Image

After:
Image

I made templates for the other 3 mounting points out of hardboard/masonite by installing the front outer bolt to keep the seat in place and sliding the template boards under the rail and on top of the bobbin, then marking the position of the rail and rail hole from the top, and the bobbin hole from under the car. I drilled the two holes out and then drew out the final shape and cut and tested before transferring it to steel. The steel I used is high strength Domex steel which I had lying around, the edge of the rails do sit over the bobbins still, or very close to the edge of the bobbins.

Image

I did the passenger side first, and then tried the brackets on the driver's side, but they didn't work very well, so new templates for the driver's side....don't be tempted to make an exact pair without checking first, it's not all perfectly symmetrical.

I have bolted the brackets to the seats with M8 bolts for now - the BMW rails were fixed with M10 bolts, so the hole in the rail is quite a bit bigger - 12mm square or so. I will test an M10 bolt and nut, to ensure there is still clearance under the bracket for the larger nut. This is the bottom of the passenger seat with the brackets attached.

Image

The bobbins do stick up above the level of the floor a fair amount (see pic below) so a nut and bolt can be used to tie the brackets to the rail without resting on the floor or being in the way of the bobbin.

Image

There is access to get 3 of the seat fixing bolts through the bobbin from the top, but the rear outside one sits too close to the seat rail - this mount will in all probability need a bolt with a low profile head to be welded to the bracket.

Clearance is tight - outside of the driver's seat:

Image

The driver's seat and minimal clearance to the transmission tunnel.

Image

So physically the seats fit at a squeeze, but with millimeters to spare.

Neil.

Pete Boole
Senior Poster
Posts: 3242
Joined: Tue Apr 04, 2006 21:28
Model: Elite
Colour: Monaco White
Year: 1974
Location: Nottingham

Re: Excel seat alternative - BMW

Post by Pete Boole »

Excellent! They are going to be soooo comfortable :D. Lots of lateral support as well - the std seats aren't great for that.

Pete

User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Re: Excel seat alternative - BMW

Post by Excel SA »

The lateral adjustment is very nice - I drove my sister's BMW with manual seats and the electric adjustment on the backrest on the weekend, you can tighten the seats up to grip you quite snugly.

Neil.

User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Re: Excel seat alternative - BMW

Post by Excel SA »

Electrics

If you want a relatively easy and sensible transplant, then the seats below would be a good option - all manual adjustments except for the backrest width adjustment which would need power - the green arrow points to the switch for adjusting this. There will be other wiring under the seat - airbag, seat belt pretensioner, data probably - seat occupancy sensors etc. as well as the wiring for the width adjustment. On the electric seats I have the power supply for this is separate to the rest of the seat, but there is only one earth supply. As far as I can tell power to the width adjustment is from a 10 amp fuse and this feeds both seats, but I'm not 100% sure of this....The switch controls a pump and solenoid valves - some seats have additional lumbar support adjustments controlled by the same pump, but more valves and switches.

Image

Wiring to the electric seats

If you're a sucker for punishment/looking for more of a challenge/have no other choice and want to do things the harder way, read on.

I have found a couple of issues with the electronics on the seats that stem from the boards themselves - they are identical items on the driver and passenger seats. One of them will not move the seat backwards and forwards using the switch on the top of the seat back, and the other one only works for 2 to 3 seconds at a time, then you have to release the button you're using and then push it again....knowing this I was prepared to take a few chances, but I am consciously incompetent at this stuff and am only copying others....please bear this in mind if you choose to do this as well.

Below is a picture of the wiring connections - I have cut and pasted a picture of the pins below the the connecter - the connector is not deformed...

Image

Pin 1 - Brown - earth
Pin 2 - Red/Brown - +12V from a 30 Amp fuse (I believe 30 amps are required due to the memory function when all 4 motors could operate at once to move the seat to the correct position - the memory function probably won't work as it requires other inputs via the data cables....)
Pin 3 - Orange/Green - data - CANH
Pin 4 - Green - data - CANL
Pin 5 - Red - +12V from 10 Amp fuse for back rest width adjustment - this feed is for both seats I think.

Connecting power to the 3 power pins - 1,2 and 5 will get the seats powered up and moving if they are all good to work, however, they will stop working after 5 seconds, or 5 seconds after you stop adjusting things...I did do some research on the internet before buying the seats, and from this thread ( https://www.bimmerfest.com/threads/info ... 057/page-4 ) and this one (https://www.e46fanatics.com/threads/diy ... e.1231324/ ) I found out there are two possible ways to keep the seat "alive".

1. Apply 9 volts to a pin of a chip on the circuit board - the chip is part of the CANBUS network on the car, and applying power to that pin keeps the network awake - I have no real clue about this stuff, and am grateful that others can work it out!

2. Apply an electrical pulse to the CANH wire (Pin 3 above) which keeps the network awake.

Option one seemed like the easier option - a resistor, some wire and a bit of soldering. That is until I saw the size of the components - they are all tiny! So I moved on to option 2....

The theory behind option 2 is that a low voltage (5 volts or so) pulse is sent through the data cable to the board which keeps it from shutting down. The links above suggested a 1 second pulse every 3 to 4 seconds would work. My thinking suggests a shorter pulse should do the trick - the network is capable of processing much, much faster pulses that that. I did some research and bought a coupe of items - they weren't expensive - under GBP 10.00 for both items and courier fees (we daren't send anything by post around here...)

1 x DC to DC converter to bring the voltage down to around 5V (maybe not necessary and a resistor could possibly be used after the pulse generator as the current is so small??? - I'm not expert so don't take my word for this.) This is the bottom module in the picture below - voltage adjustment is done my turning the brass screw and I initially thought I had been sold a dud as turning the screw had no effect on the output voltage (I'd paid and extra GBP for the fancy one with a voltage display...). It took many turns before the output voltage changed at all.

1 x pulse generator - this can be programmed in various different ways - I set it up to generate an on pulse, then be off and continue looping this way as long as power was supplied to the module. I chose this particular one as it is all solid state - no relays to click and it can supposedly switch endlessly.

Image

I connected the OUT+ of the pulse generator to the CANH wire, and had absolutely no joy in keeping the seat awake....I tried numerous pulse times, both on and off without success. The component does have an LED that switches on when it sends a pulse, and additionally I checked the output by connecting an LED to it, and it all worked as expected, but failed to keep the seats awake - I have no idea what I did wrong....

So I went back to option 1 with some trepidation....

Below is a picture of the board, with a green arrow pointing to the pin that needs power (I did check that there was 9 volts on the pin with the seat powered up). Did I mention that this stuff is all tiny? Apologies for the quality of some of the photos of the pc boards, phone cameras and me and my eyesight have their limitations.

Image

So that's Pin 7 on the TJA1043 chip.

I needed to do the following work under a magnifying glass, it seems to have worked and I don't seem to have cooked/broken/damaged or destroyed anything on the board that I can tell yet.

I soldered a wire to pin 7 - on the first board I soldered it onto the pin on the chip directly, on the second I thought it would be better to feed a wire (a very, very thin wire), through the hole in the board and then solder that - both worked - see pics below. I then added a 10K ohm resistor in series to the wire and then soldered that to the +12V pins coming into the board so that it's all self contained, I hot glued the wires in place. The resistor is to bring the voltage down to around 9 volts - on the one link the guy uses a 5.6K ohm resistor, but that was giving me almost 10 volts from a 12.6 volt battery, the 10K ohm resistor brought the voltage down a bit more, but it is still over 9 volts.

Image

Yellow arrow points to the wire coming through the board:

Image

The red circle is the connection to pin 7, the white one is the resistor, and the yellow circle is where I connected it all to the +12V supply coming into the component.

Image

There is a plastic cover that goes over the board, and all tests so far show that the seats stay active when power is supplied.

The plan is to run only on +12V power supply to each seat from a relay and that will then power the seat as well as the backrest width adjustment, and obviously an earth.

I have also come up with a cunning plan to supply power to the seats when required...that's next.

Neil.

User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Re: Excel seat alternative - BMW

Post by Excel SA »

Power

The seats obviously need power supplied to them to work - all the installations I came across seemed to be for 4-door cars, and power was only supplied when the ignition was turned on. This is not so useful here - I need to be able to move the seats before then so any back-seat passengers can get in, before starting the car. After throwing a few complicated ideas around in my head (involving car unlocking, timers etc.) I thought of an easier option - apply power when the doors are open - from the door switches. I also need power when the doors are closed and the ignition is on, so a combination of both door open and ignition on is required.

Simply joining an ignition power feed and a door open feed into one circuit will result in the power from each circuit ending up in both which would be a problem, and some possibly large currents finding their way through the joint. The simple solution to this (I hope...) is using diodes - these are used quite extensively throughout the car, and I have had to repair one that was blown already - see here search.php?keywords=diode&t=12508&sf=msgonly. The diodes act as one-way valves and stop one circuit from feeding into the other. The diodes used were 1A, 100V.

At tis point I was expecting to use the DC-DC converter and the pulse generator, and I have 2 other components that will need an ignition fed power supply (to change the electronic speed pulses for my speedometer, but not relevant here) so I made up a simple circuit board using veroboard.

Image

The green line in is from ignition - it feeds directly to 2 ignition out lines (the thin green wire joins the two output lines together), as well as crossing over via the green arrowed diode to the combined ignition/door open feed in yellow.
The red line in will be the door open power feed - I am tapping into this circuit already with the addition of remote locking/alarm. This crosses over via the red arrowed diode to combine with the ignition feed and should give power to the seats when either circuit is on, or both are on. The white link wire joins two tracks together to combine the circuits and for two separate outputs.

The thinner coloured lines drawn on the board give you an idea of how the holes in the board all join up on the other side of the board, which has a copper strip running across each row.

The two yellow outputs will now feed into switching two relays - one for each seat, rather than into the pulse generator.

As the car is still apart, and all the wiring currently in the boot, this is all theoretical for now......

Neil.

gusexcel
Junior Poster
Posts: 72
Joined: Tue Sep 17, 2019 19:57
Model: Excel SE
Year: 1987

Re: Excel seat alternative - BMW

Post by gusexcel »

Aren't they just DC motors?
I would think all that circuitry is for memory and all that hoo hah

User avatar
Excel SA
Senior Poster
Posts: 448
Joined: Thu Jul 02, 2020 06:11
Model: Excel
Colour: White
Year: 1987
Location: Johannesburg, South Africa

Re: Excel seat alternative - BMW

Post by Excel SA »

They are DC motors, but some with a twist....Most wiring diagrams and research I did show that there are 4 wires to each motor - 2 for power, 2 for control, and simply applying power won't do anything useful. In reality, some of the motors on my seats have 2 wires, at least one has 4. There are relays on the board to handle the current to the motors, so although it's probably feasible to control the motors directly, it would take some effort I think - new relays configured, polarity reversed for forward/backwards etc. assuming the motors can handle power directly.....I haven't tried to do that, and probably won't try either!

Neil.

Post Reply