1984 633CSi Rises again!

Document your Sixer project here.

Moderators: GRNSHRK, ron, bfons

songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

hornhospital wrote: Mon Oct 12, 2020 8:50 pm It tells me your temperature sensor that feeds the gauge is perhaps bad, or (unlikely) there's a local obstruction in the system adjacent to the sensor. I believe there is a test method for the sensor, but I can't recall what it is, and my Bentley manual is packed away.
So I checked out the Bentley Manual and the process seems very inexact, but then again, I guess they would know best. I'll try that out later. However, I did get to do a detailed measurement of various spots as the car heated up. I couldn't get it to go past about 5/8 on the gauge today after running the car for just over 40 minutes. Here are the readings.
Cooling System readings
Cooling System readings
Screen Shot 2020-10-12 at 10.36.59 PM.png (783.71 KiB) Viewed 10228 times
Does anything look awry? Ambient was about 78 degrees today when I did this.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
User avatar
olympia57
Platinum Member
Platinum Member
Posts: 1389
Joined: Thu Sep 13, 2012 1:47 pm
Location: Ireland

Re: 1984 633CSi Rises again!

Post by olympia57 »

Mmmm......
A word of caution when using these cheap low cost hand held thermometers .
They are sold as a one shot measure everything device but are anything but.
Consult the specifications of the unit to determine the FOV ( field of view ) as this will determine the target size and distance from the point of measurement .
Image
As you can see from above this has a huge effect on the accuracy of any measurement taken.
The target size MUST be greater than the FOV or you will be measuring anything else that is in the FOV , if it is a greater temp. then it introduces a positive error , if it is less then the opposite.
That's the simple bit .
The material and surface of the target is crucial to the accuracy of measurement , this is called emissivity .
Any Infra red measuring device must have an e adjustment to compensate for this .
Rubber , steel ( painted or raw) ,plastic , all types of alloys , water , oil ,SKIN , timber etc, ad nauseum all differ greatly , get the picture ? :roll:
I could spend a long long time explaining this .
Basically black is 0 and a shiny reflective surface is 10 but is not linear between the two. This has a HUGE effect on the accuracy and again is non linear across the measurement range of the device .
A crude , but reasonably effective , way around this is to either paint the surface black or wrap it in black tape and set the e value to 0.
Lastly the target "L" in the diagram above is circular . In essence the device takes a cross sectional average of the IR transmission in that area .
If you are not at 90 degrees to the target the circle becomes ellipsoidal and again greatly effects the measurement.
Most , if not all , of these low cost thermometers will have a minimum target size of 50mm + , I have a hand held with a MTS of 9mm and that cost over £3000 ( yep that's right ) ......
Just remember all this when the wee girl at the door of the supermarket you are about to enter points a £15 hand held thermometer at your forehead and says " Ah ,you're OK , on you go ...." [-o<
I hope this is of interest to you .
Good luck :lol:
Attachments
camera_fov.png
camera_fov.png (7.67 KiB) Viewed 10222 times
1981 635 series 1
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

olympia57 wrote: Tue Oct 13, 2020 9:13 am Mmmm......
A word of caution when using these cheap low cost hand held thermometers .
They are sold as a one shot measure everything device but are anything but.
Consult the specifications of the unit to determine the FOV ( field of view ) as this will determine the target size and distance from the point of measurement .
Image
As you can see from above this has a huge effect on the accuracy of any measurement taken.
The target size MUST be greater than the FOV or you will be measuring anything else that is in the FOV , if it is a greater temp. then it introduces a positive error , if it is less then the opposite.
That's the simple bit .
The material and surface of the target is crucial to the accuracy of measurement , this is called emissivity .
Any Infra red measuring device must have an e adjustment to compensate for this .
Rubber , steel ( painted or raw) ,plastic , all types of alloys , water , oil ,SKIN , timber etc, ad nauseum all differ greatly , get the picture ? :roll:
I could spend a long long time explaining this .
Basically black is 0 and a shiny reflective surface is 10 but is not linear between the two. This has a HUGE effect on the accuracy and again is non linear across the measurement range of the device .
A crude , but reasonably effective , way around this is to either paint the surface black or wrap it in black tape and set the e value to 0.
Lastly the target "L" in the diagram above is circular . In essence the device takes a cross sectional average of the IR transmission in that area .
If you are not at 90 degrees to the target the circle becomes ellipsoidal and again greatly effects the measurement.
Most , if not all , of these low cost thermometers will have a minimum target size of 50mm + , I have a hand held with a MTS of 9mm and that cost over £3000 ( yep that's right ) ......
Just remember all this when the wee girl at the door of the supermarket you are about to enter points a £15 hand held thermometer at your forehead and says " Ah ,you're OK , on you go ...." [-o<
I hope this is of interest to you .
Good luck :lol:
I appreciate the tips. I did take the reading several times before averaging and writing down result. Weird thing is that the engine running hot isn't consistent. It was hot once just sitting in traffic and then again when I was on the freeway. It didn't make a lot of sense to me. Something has changed as it went many years without an problem running hot.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
User avatar
TBM
Posts: 342
Joined: Tue Apr 15, 2014 4:05 pm
Location: California
Contact:

Re: 1984 633CSi Rises again!

Post by TBM »

Have you checked that your fan clutch is fully engaged? Common problem..If not fully engaged its of course not moving enough air.. If its bad may be a good idea to use either a high CFF electric fan on the front of the car (a pusher) or once you remove the fan clutch a low profile sucking fan on the inside of the engine compartment.. Its a bit of a job but the fan either works or it doesn't and in the future easier to diagnose or replace.

. When was the last time you flushed out the system with new coolant? If over a year ago you may have some kind of blockage (not likely, but you never know.) Coolant should be changed every year either way or it turns acidic. Also, you could just take the temp sensor out and run it outside the car in boiling wanter and measure the resistance to see if it's reading correctly (need to find what resistance readings should be..)
'84 M635csi
'86 Lotus esprit Turbo (sold)
User avatar
86_6series
Platinum Member
Platinum Member
Posts: 1069
Joined: Mon Apr 02, 2007 10:08 pm
Location: West Nyack, NY

Re: 1984 633CSi Rises again!

Post by 86_6series »

What I see is the temp for the upper rad is lower
than the bottom. Usually hot water enters the rad @ top
and cools as it works it's way down to lower hose.

Other than that, temps that you measured seem normal to me.

As said, it could be the guage or the temp sending unit to guage.
86-635CSi
96-MB C220 SOLD
98-740i
71-Continental MK3
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

TBM wrote: Wed Oct 14, 2020 4:56 am Have you checked that your fan clutch is fully engaged? Common problem..If not fully engaged its of course not moving enough air.. If its bad may be a good idea to use either a high CFF electric fan on the front of the car (a pusher) or once you remove the fan clutch a low profile sucking fan on the inside of the engine compartment.. Its a bit of a job but the fan either works or it doesn't and in the future easier to diagnose or replace.

. When was the last time you flushed out the system with new coolant? If over a year ago you may have some kind of blockage (not likely, but you never know.) Coolant should be changed every year either way or it turns acidic. Also, you could just take the temp sensor out and run it outside the car in boiling wanter and measure the resistance to see if it's reading correctly (need to find what resistance readings should be..)
Thanks for the input. I hadn't thought about the fact that hot coolant enters at the top. Perhaps checking the center top of the radiator is not an ideal place. I made special note of the area right by the temp sensor. As you can see, it hardly changes.

I thought about the viscous clutch, but the "newspaper test" tells me that it's not the culprit.

So yesterday I took the car to the office. On the way home, I ran an errand. All together, this was over 100 miles of driving. I watched the temp closely. I even got out once when it was at 3/4 on the temp gauge and measured the area by cylinder 6, which is what normally gets the hottest. I saw a reading of 223F, which is a bit hot but not surprising on a 91F day. Once again I checked by the temp sensor and it was 184F. Last weekend I put in a 80C (176F) thermostat so that reading didn't seem out of line. Also while running errands and driving home, I ran the A/C which had a measured 46F-48F blowing out the center vent the whole time. That made me happy since it was a hot day.

The weird part is that I could be at a constant speed on the freeway and it would rise close to 3/4 and then drop to just over 1/2 and then go back to 3/4, all within the span of 15-20 minutes. It just doesn't make sense!

I may try the coolant flush, just because I'm running out of ideas!
Last edited by songzunhuang on Fri Oct 16, 2020 8:40 pm, edited 1 time in total.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

86_6series wrote: Wed Oct 14, 2020 11:49 am What I see is the temp for the upper rad is lower
than the bottom. Usually hot water enters the rad @ top
and cools as it works it's way down to lower hose.

Other than that, temps that you measured seem normal to me.

As said, it could be the guage or the temp sending unit to guage.
The temp sensor test was mentioned in another thread as well. I actually have an extra one of those I could compare against. Turns out in the early days of fixing this car, I had suspected the sender and purchased a spare. However, when I install the spare many years ago, the reading was even higher on the gauge! This made me put the old one back. ;)

In any case, worth a shot since I have the bits on hand.
Hm, I just grabbed the spare temp sender and tried to get a resistance reading at room temp. I couldn't even get my multimeter to register anything. It's like the sender is an open circuit at room temp. I them ran outside and measured the unit installed on the car. It was 84F at the temp sender housing. Alas, when I measured the installed unit, nothing. Once again, it appears that it's an open circuit until it warms up a lot more.

Oh well, I guess I'll have to put them in boiling water and try a side by side.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
User avatar
sansouci
Platinum Member
Platinum Member
Posts: 1850
Joined: Mon Jul 08, 2013 10:52 pm
Location: Southampton NY

Re: 1984 633CSi Rises again!

Post by sansouci »

One good test is to pull the Tstat and put it in cool water that you then heat up. Get a high quality mercury thermometer (I think the kind used to make candy about $11 on Amazon) or maybe a dial type and run multiple heat-up/cool down cycles on the stove (your wife will think you've taken up cooking!). If it is not consistent, you have a lazy stat.
Ken
Last edited by sansouci on Tue Dec 29, 2020 8:59 pm, edited 1 time in total.
Sansouci
84 E24 633Csi Auto, Bronzit/PearlBeige 6997510
93 E32 740il M60 Auto, Alpenweis/Ultramarine
60 528i M30 5-speed Green/Beige (crushed)
71 240Z 4-speed White/Blue (rusty & sold)
65 396 Chevelle 4-speed, Marina Blue/Black (stolen)
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Running Hot? Throw parts at it!

Post by songzunhuang »

Ok, I know many of you have had this thought when a vexing problem isn't going away. As I was thinking about what bits of my cooling system hadn't been replaced yet, I thought of the fan clutch and the auxiliary fan temp switches. Although it wasn't clear that these were failing, I've never really had a reason to replace them, but I figured that they are OEM so it's about time. So off went the parts order. So here are a few observations.

I ordered a Meyle fan clutch. It fit well, but it sure is a lot different in design than the OEM Sachs. I hope it's a better design.
Looking at them side by side, you can see the following.
  • The Sachs doesn't have a perimeter aluminum ring.
  • The front of the Sachs is very different from the Meyle
It's a beautiful part when new. So shiny and nicely machined.
Sachs vs Meyle
Sachs vs Meyle
Fan Clutches.png (1.42 MiB) Viewed 10024 times
Here's the new clutch installed in the fan. You can see the changed front design.
Cleaned up and new.
Cleaned up and new.
Installed Fan Clutch.png (1.71 MiB) Viewed 10024 times
OK, so trying to get this installed onto the car proved to be a bit tricky. I needed about 1/8" more clearance to get the nut onto the engine, but there was no more room as the fan was tight against the radiator. Then it struck me! If I loosened the fan clutch mounting bolts, it would cause the center nut to move towards the radiator, giving me enough room to get this in! I broke out my new offset wrench and made quick work of this!
Trick to getting the Fan clutch nut onto the engine.
Trick to getting the Fan clutch nut onto the engine.
Install FC Trick.png (1.14 MiB) Viewed 10024 times
After being quite proud of myself for figuring out the trick above, I then installed the hi-temp auxiliary fan switch. This is the one that closes at 99C and kicks the electric fan on at the highest settings. Since I was getting some hot running issues, I figured having this fan kick in would help cool things down.
Red hi-temp switch
Red hi-temp switch
HI Temp switch.png (1.31 MiB) Viewed 10024 times
OK, so that all went well and I was excited to drive into my office (85 miles roundtrip) the next day. So how did it do?
Nothing changed. ](*,)

Mostly it was fine, but there were still a few times when it was showing a bit warm at 3/4 on the temp gauge. It was like that even though it was a mild 70F evening and I was on a freeway with plenty of airflow. So, I got some great new parts and eliminated a few more items, but I'm still getting intermittent readings of running warm. I think a system flush may be in my future.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
User avatar
86_6series
Platinum Member
Platinum Member
Posts: 1069
Joined: Mon Apr 02, 2007 10:08 pm
Location: West Nyack, NY

Re: 1984 633CSi Rises again!

Post by 86_6series »

My little trick to get clearance is to remove the rubber grommets
and bolt the radiator snug against the header panel. Gives me
plenty of clearance.

#4 https://www.realoem.com/bmw/enUS/showpa ... Id=17_0028

Something is not right that the top rad temp is
cooler than the bottom rad temp. I think that shows poor circulation
and the hot water is coming to the lower rad from lower engine and not being
pulled to the upper portion of the engine to flow down the rad.

For me I would do a rad flow check, and I know you don't want to hear this--
how old is your water pump? Is the impeller worn? Moving enough water?

Just brainstorming looking for a solution.
86-635CSi
96-MB C220 SOLD
98-740i
71-Continental MK3
User avatar
TBM
Posts: 342
Joined: Tue Apr 15, 2014 4:05 pm
Location: California
Contact:

Re: 1984 633CSi Rises again!

Post by TBM »

A "good" radiator shop can pull off the side tanks and look at that radiator and put it back together.

Also, might be helpful for the future if you used Evans waterless coolant. If you are pulling the radiator the flush will be very easy... You can even use a wet dry vac to evacuate the entire system after turning on the heater to also clear out the heater core. Have been running it in our vehicles for 5 years now. No scale issues, no converting to acid over time. Take the water element out. Many have their opinions but you also most importantly lower the bloody pressure in the system that ruins hoses and puts unnecessary strain on other components.
'84 M635csi
'86 Lotus esprit Turbo (sold)
User avatar
mdkohler
Bigcoupe Contributor
Bigcoupe Contributor
Posts: 79
Joined: Tue Sep 17, 2019 11:41 am
Location: Austin, Texas

Re: 1984 633CSi Rises again!

Post by mdkohler »

If anyone is interested, I no longer need my radiator.
Make me an offer.
IMG_8777.jpg
IMG_8777.jpg (117.61 KiB) Viewed 9981 times
IMG_8779.jpg
IMG_8779.jpg (68.11 KiB) Viewed 9981 times
IMG_8780.jpg
IMG_8780.jpg (79.27 KiB) Viewed 9981 times
IMG_8778.jpg
IMG_8778.jpg (70.14 KiB) Viewed 9981 times
Marc
Austin, Tx
1985 635CSi Euro
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Super Coolant Flush & crumbly fuel vapor hose.

Post by songzunhuang »

So we goto DefCon 4. It's time to clear out all the parts of the cooling system to make sure there's nothing stopping up the system.
The first thing is to use Liquid Moly radiator cleaner in the system. Fill it up, drive it around for 30 minutes and begin the drain.
This stuff only worked OK.
This stuff only worked OK.
FlushAgent.png (1.38 MiB) Viewed 9726 times
After the draining by pulling off the lower radiator hose, where the old coolant got all over me, I started taking apart all kinds of hoses. The hose to the heater valve, the hose to the plastic "T", the thermostat and the upper radiator hose. I used a hose and blew out each system forwards and backwards. I guess that's what a backflush is all about.
Time to get serious with the flush!
Time to get serious with the flush!
CoolantFlush.png (1.42 MiB) Viewed 9726 times
Then to complete the flush before assembling everything, I had to take out the block plug. I heard it's below the 5 and 6 exhaust ports. Crawling under the car, I was able to reach the plug by reaching in over the engine mount and using my 19mm socket wrench. When I first got it off, it was a trickle. Then I got a screw driver and poked in the hole, as I read in various recommendations, and sure enough, it gushed out. I did a bit more power flushing with my hose until the water was clear.
Getting the block plug from below the car.
Getting the block plug from below the car.
BlockPlug.png (1.46 MiB) Viewed 9726 times
I also replaced my blue silicon hose with a genuine BMW hose. Slowly but surely, they will all be BMW hoses.
Anyhow, I was focusing on the coolant when I smelled gas. Opening my gas fill cover and looking under the rubber gasket I noticed that my vapor hoses were wet with gas! The lower 2 hoses were wet and crumbly. Crap a diversion. I had to remove the gasket and cut off about 1/2" of the ends to t to some relatively good hose.
Time to address this issue. New hoses ordered.
Time to address this issue. New hoses ordered.
Fuel Lines.png (1007.46 KiB) Viewed 9726 times
Well, I also put back in my 75 degrees Centigrade thermostat and the OEM temp sensor. Recall I had replaced both of these to no avail. I refilled the system with OEM BMW coolant and distilled water then went for a drive. So far, so good but the real test will be when I drive it into the office next week. Oh, I also noticed that my passenger side axle boot is torn now.

Man, these cars keep you very busy.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
User avatar
hornhospital
Platinum Member
Platinum Member
Posts: 2931
Joined: Sun Jul 11, 2010 6:46 pm
Location: Silverhill, AL
Contact:

Re: 1984 633CSi Rises again!

Post by hornhospital »

It's good to see you posting an update!

What's the thing above the cut off hose ends in the photo? I see the Roundel on it, but don't recognize it.
Ken Kanne
'84 633CSi "Sylvia"; '85 635CSi "Katja";'85 325e "Hazel Ann"; '95 M3 "Ashlyn"
87Shark
Posts: 177
Joined: Sun Mar 10, 2013 5:14 pm
Location: Peoria, AZ

Re: 1984 633CSi Rises again!

Post by 87Shark »

Gas neck cover, just inside the door. It hides the hoses and kind of supports the filler neck.
'87 L6, Arctic White
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

87Shark wrote: Sun Dec 13, 2020 8:13 pm Gas neck cover, just inside the door. It hides the hoses and kind of supports the filler neck.
Yes, that is it! Luckily FCP Euro has these on sale right now. I was shocked to get a replacement for $14. All over eBay, that little rubber part is about $37-40. I also couldn't stomach the price of replacement vapor hoses. The OEM BMW hose 6mm ID x 11mm OD was priced at $70 a meter! This is freaking ridiculous. They call it a "fuel hose" and I would probably need at least 2 meters.

I looked on FCP and saw a 3rd party braided 7mm x 11mm for $6 a meter. I'm hoping the difference between 6mm and 7mm won't be too much for a hose clamp to overcome. Also, I think this is for fuel vapor capture, so I don't think it needs to be able to take super high pressure. We shall see.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

hornhospital wrote: Sun Dec 13, 2020 7:40 pm It's good to see you posting an update!

What's the thing above the cut off hose ends in the photo? I see the Roundel on it, but don't recognize it.
Hey Ken, I appreciate the encouragement. I've been really busy at work rolling out a new product and initiative with Nasdaq for diversity on corporate boards. It's been a heck of a task getting it to the announcement last week. https://www.equilar.com/press-releases/ ... rsity.html

Also, other cars in my family, Volvo wagon and a Volvo SUV, have all needed work as of late. The shark has been ignored a bit. Then last week, I drove my little convertible and noticed a weird rear end clunk. A bit of search on the web and I found a solution for that. So it needed attention too!

So basically, I haven't posted anything because I have had way too many distractions.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Cooling system - all seems well

Post by songzunhuang »

Well, it looks like the super flush of the system seems to have done some good. I drove about 86 miles round trip today and never did the temp gauge go above the 12 o’clock mark. It mostly stayed under around 10-11 o’clock. So I can mark that issue resolved and turn my attention to the torn boot on the passenger side half-shaft. All is well for a while.

Anyone need cooling system parts? I have a thermostat (new), temp gauge sender (new), viscous coupling (used). All good working order. My car just needed a super flush.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

CV Boot Driver side - replacing a torn boot

Post by songzunhuang »

Well, my boot came in and I replaced it. Fun thing was that it gave me an excuse to get a Milwaukee impact wrench. This sucker puts out 600 ft lbs of nut busting torque. I learned that I need to be careful. It made quick work of the bolts.
Milwaukee Fuel - 600 ft lbs
Milwaukee Fuel - 600 ft lbs
Fuel.png (1.16 MiB) Viewed 9607 times
Repacking the CV joint can be a big old mess!
Repacking the CV joint
Repacking the CV joint
BigMess.png (1.33 MiB) Viewed 9607 times
I've done this once before on the driver side, this wasn't so different, except I forgot all the steps and I had to relearn.
I got a chance to see the rear hub and suspension. It all looked nice and clean.
The passenger side rear hub.
The passenger side rear hub.
Clean Hub.png (1.4 MiB) Viewed 9607 times
Something nice about a clean CV Boot and re-installed, clean half shaft.
New CV boot and repacked joint.
New CV boot and repacked joint.
RT CV Boot.png (1.42 MiB) Viewed 9607 times
That was it for the e24 today. Then I had to go fix both steering rack boots on my Volvo wagon.
I did get a chance to take the e24 out for a drive to get some pickup food. It was just fine and I can rest easy that my CV joint isn't getting all kinds of dirt in it.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Fuel vapor hoses - the actual sizes

Post by songzunhuang »

Well, today I tried to install the new hoses I got from FCP. It turns out that the hoses I got were conti brand, but the sizes were wrong. In reality, the hoses probably could have worked, but I didn't have the right clamps to make this work. Here is what I learned in pictures.

So this is the vapor canister fastened by 2 x 10mm screws. Looking at it closely, it just looks like a thick plastic container to me. I don't think there's actually anything inside of it. Also, note that all the hoses are fastened by oetiker clamps. I didn't have any of those and that's what ultimately made me stop.
Vapor canister
Vapor canister
Canister.png (915.18 KiB) Viewed 9515 times
Here's the original hose on the left and the hose I got on the right. Never has 2mm looked like such a big difference. The original hose turns out to be 6mm inside diameter and 9mm outside diameter. The hose I got was 7mm inside and 11mm outside. Weird because realoem said the hoses were 11mm outside. Oh well.
Original an potential replacement.
Original an potential replacement.
Hoses.png (1.01 MiB) Viewed 9515 times
So I tried the replacement hose on the steel tube out of the gas tank. It seems like it'd work just fine actually, but I didn't have any oetiker clamps so I didn't want to leave things in pieces. Once again, I just cut off a little bit of the old hose and reattached it. I'll order some oetiker clamps and go at this some other time.

As for the gas filler gasket, that worked fine. I sprayed some water on it to help slide it over the filler cap.
New BMW gas filler gasket.
New BMW gas filler gasket.
Gas filler gasket.png (861.99 KiB) Viewed 9515 times
This was easy to install and it worked perfectly.
New gasket installed.
New gasket installed.
GAsket installed.png (874.28 KiB) Viewed 9515 times
So, I learned something but didn't get a lot accomplished on the shark. I spent the rest of the afternoon installing a new driver side axle on my daughter's station wagon. That's another story.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
User avatar
jps635
Platinum Member
Platinum Member
Posts: 872
Joined: Sat Feb 27, 2010 10:50 pm
Location: NZ

Re: 1984 633CSi Rises again!

Post by jps635 »

Yeah you need the right hose. You mind find it difficult to get the otiker clamp tool on the hoses connecting to the filler tank neck in which case look at using fuel injection hose clamps. They pull up evenly and don't cut into the hose like a jubilee clamp. Also be aware is easy to break the hose tales on the expansion tank when removing the old hose.

While you're at it check for corrosion around the vent tube connections, is where the commonly corrode.
'85 635csi JPS (RA2-66)
User avatar
sansouci
Platinum Member
Platinum Member
Posts: 1850
Joined: Mon Jul 08, 2013 10:52 pm
Location: Southampton NY

Re: 1984 633CSi Rises again!

Post by sansouci »

Closet Milwaukee fanboi?
Sansouci
84 E24 633Csi Auto, Bronzit/PearlBeige 6997510
93 E32 740il M60 Auto, Alpenweis/Ultramarine
60 528i M30 5-speed Green/Beige (crushed)
71 240Z 4-speed White/Blue (rusty & sold)
65 396 Chevelle 4-speed, Marina Blue/Black (stolen)
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Fuel Vapor Hose replacement

Post by songzunhuang »

Hey, bigcoupe crew! Life with the big coupe has been blessedly boring these past several months. Aside from replacing wipers in May, and then the hose replacement in April that I am finally capturing here, nothing else has transpired. The car has been reliable and performing well. It's getting a bit warm and I ran my A/C and everything is blowing chilly.

For the sake of completeness, here's the boring fuel vapor hose replacement documentation.

I got a Oitker clamp tool and an assortment of clamps to do the job.
All the bits needed to replace the vapor hoses.
All the bits needed to replace the vapor hoses.
HoseReplace.png (1.47 MiB) Viewed 6984 times
As luck would have it, I didn't have enough hose to replace all of the original ones. I just did the top and bottom ones, which were in bad shape on the fuel tank filler end. Oh, little trick that you can see in the picture with the top hose. Since the new hoses were slightly bigger than the original, I cut the tips at an angle to feed it through the rubber seal. I then grabbed the angled part fed through with a plier to pull it the rest of the way. Cut the end at 90 degrees and then it's all set for re-installation.
Hoses on tank side. I didn't replace the center.
Hoses on tank side. I didn't replace the center.
HoseTankSide.png (1.37 MiB) Viewed 6984 times
Here's the hoses replaced on the vapor canister side and the new clamps in place. You can see how the new hoses have a slightly larger diameter.
Vapor canister side of the hoses.
Vapor canister side of the hoses.
HoseVaporSide.png (1.26 MiB) Viewed 6984 times
So I went ahead and buttoned everything up and for 2 months now, no more fuel vapors have been detected. So I think it's good for now.
So tip to the forum. You don't have to purchase the wicked expensive OEM braided hoses to make this work. The Conti-tech braided hoses are a fraction of the cost ($6 compared with $60 for the same length) and they work just fine.

OK, that's it for now. Nothing to do on the car really as it's running well (except for that bumpy idle). I'm beginning to ponder the leather interior. The leather is getting hard and the seam splitting. Hmmmm....
Last edited by songzunhuang on Sat Jun 12, 2021 11:47 pm, edited 4 times in total.
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
Jlc
Posts: 287
Joined: Sun Sep 27, 2020 5:27 pm
Location: Atl

Re: 1984 633CSi Rises again!

Post by Jlc »

How much of each line did you wind up using. I have a gas oder and will replace all lines as well am considering having tank checked/flushed/lined.

I’m thinking of using fuel injection line clamps on all lines - any issue with that?
85 635CSi Bronzit
songzunhuang
Platinum Member
Platinum Member
Posts: 1224
Joined: Fri Aug 14, 2015 10:04 am
Location: Silicon Valley, CA
Contact:

Re: 1984 633CSi Rises again!

Post by songzunhuang »

Jlc wrote: Sat Jun 12, 2021 10:17 pm How much of each line did you wind up using. I have a gas oder and will replace all lines as well am considering having tank checked/flushed/lined.

I’m thinking of using fuel injection line clamps on all lines - any issue with that?
Well, the lines came in 3 meter lengths from FCP euro. I ordered 2 x 3 meter lines. I don't know why but I thought that would be enough for all three. Well, it wasn't. There was about 6-9 inches cut off of each of the lines I replaced. As you can see, I replaced 2 of them. The middle one is holding up still. The others just crumbled away.

Here's what I purchased.
FCP fuel line order.
FCP fuel line order.
Screen Shot 2021-06-12 at 4.32.10 PM.png (34.31 KiB) Viewed 6974 times
It's easy to see why I had fuel smells coming from the filler. Here's what the hoses looked like when I removed them.
Crumbly old lines.
Crumbly old lines.
Old lines.png (955.45 KiB) Viewed 6973 times
The breather hoses off the fuel filler cannot be fasten with oetiker clamps. There's not enough room to work in there. Those are fasten with standard hose clamps, although very small diameter ones. I assume that's what you mean when you say "fuel injection line clamps".
-----
Song Huang
1984 633CSi
Last 7 of VIN: 6997383
Post Reply