When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I have decided against the 3-way valve in favor of a Y and two separate 1" pex ball valves. That way they will each will flow straight through instead of being a 90 in one postion. It also allows me a lot more flexibility in positioning the valves since the plumbing will be more straightforward.
I got the feed line completely run from where the rear pump will be up into the bay. It was no easy task...that hydraulic hose is 1.25" OD and not exactly flexible. I ran it from basically the rear bumper up and over the muffler heat shield, over the rear subframe, through the fuel pump box, and then on top of the underbody plastic where all the other lines are hidden. Only had to trim a few places on some of the plastics. At the front of the car I ran it right along the ABC lines and managed to snake it up right between the headlight and A/C accumulator or dryer or whatever that is. I plan to put the Y right there and locate valves somewhere in that little bit of space behind the driver's side hood latch there. Then run the bypass loop on top of the radiator and then down to Y into the second pump feed line. The other line will go to the heat exchanger. Then I'll be re-doing the plumbing for the intercoolers with Y fittings instead of the stock T pipes and from there, run a return line back to the trunk tank. Lots of places the hose just barely fits. I'm super happy I was able to use it instead of just heater hose (have had bad luck in the past with heater hose kinking and such).
Sounds like a big pain. One of the reasons I was hoping not to do a trunk tank is adding all of the plumbing. Do you have a lift or are you doing this on the floor?
Sounds like a big pain. One of the reasons I was hoping not to do a trunk tank is adding all of the plumbing. Do you have a lift or are you doing this on the floor?
Fortunately I'm a mechanic by day and my father owns the place, so I have shop access outside of business hours for personal projects. I'm very lucky. It was on a lift, and I doubt whether I could have snaked the hydraulic lines over the rear subframe without loosening the back two bolts (it fits with it tight, but feeding it through there was very difficult until I loosened the bolts). It could definitely be done with heater hose much easier (about 1/4" smaller OD) but I really only want to do this once. And like I said, heater hose is really easy to kink, or to be damaged abrasively or punctured/cut. On previous install, heater hose would deform just from the zip ties used to secure it, lol. It's pretty well impossible to kink or hurt 3,120psi two-wire hydraulic hose.
It's only twice the weight per foot of a quality heater hose and similar price so we're talking a minimal amount of extra weight...maybe 15 lbs more than the heater hose would have been. Marine hardwall would be nice, but it's 2-3x the price and can still kink/be damaged a lot easier.
Ok, that is by far the worst/most challenging/frustrating thing I've had to do to the car so far. By that, I mean installing the Ys for the feed and return to the intercoolers, and routing the return line over the rear subframe. There's really only room for ONE line above the frame, not two. So I think one is ovaled/squished a little bit, I am going to have to do a flow test once I get the last couple connections sorted out. At this point I have all the lines run to the trunk, and everything behind the intercoolers done. I just have to install the two valves and the new line to the heat exchanger, and the bypass line. I really hope there are no leaks or issues, it's a huge chore getting to any of the connections and all the hoses are such a tight fit I don't think they're coming back off without breaking something (the Y fitting are plastic...quality plastic, but plastic nonetheless...I couldn't find any steel ones in 3/4"). Anyway, I'll post more and some pics later.
I took some pics but I'm exhausted (been at it all night) and have to be back at work in about 6 hours so I'm gonna get to bed and upload them later.
Ok, so in hindsight I should have taken more pictures during the actual install progress, because once everything is together, you can't really see any of it. Essentially, I removed all the stock piping behind the intercoolers. I did end up cutting one T out of it to be able to connect the filler cap inline, and re-used a couple of the heater hose elbows where I needed a tighter bend than I could convince the hyraulic hose to make. I only decided to retain the filler neck because I already have the adapters to use that for a vacuum fill port, whereas I would have to add/fabricate something in order to do so at the trunk tank. Plus, I had already installed the expansion tank and didn't want to now remove it and have an empty hole in my false firewall.
Basically, the stock feed pipe is intact to the rear of the passenger side head, and I put the feed Y right there on the elbow coming off of that pipe. The return line Y is also right in that area, as I ran the line through the grommet that goes through into the cavity that holds the A/C inlet box. So the lines from each Y to the passenger side intercooler literally have to make a pretty tight U-turn to go into the intercoolers. This was a HUGE pain to get to fit...there is no way it could have been done with non-molded heater hoses without creating kinks. The driver's side wasn't much better because I had to squeeze two 1.25" OD hydraulic hoses under that wiring protector and then make a very tight curve up into the intercooler.
To top it off, the barbs on everything are quite large and as you know if you've ever worked with it, hydraulic hose does not expand like regular heater hose. Even with lube, it took all my might to get the hose onto each connection. And of course, it takes an abrasive chop saw to cut so I had to keep running back an forth from the car to the chop saw taking small increments off of each hose each time to get the length just right. This was a HUGE project. My biggest concern is that I coul have overstressed one of the plastic Y fittings in the process and that there could be a crack as a result. Won't be able to tell until I finish the tank mounting and run the valves and new hose/bypass hose.
I decided to wrap the intercoolers in Audio vibration dampening mats. They are basically sticky rubber with aluminum foil backing. The aluminum should help to stop radiant heat from the engine and the rubber should provide some insulation. They ende up with 2-3 layers of 80 mil mats. It doesn't look that great, but the good news is with the new inside-the-manifold water/meth setup, I can actually get the engine cover back on so who really cares what it looks like underneath?
Overall, I'm quite happy with the install. The lines under the car are completely hidden/covered by factory plastics with the exception of inside the driver's wheel well where it follows the ABC lines. They are secured well away from anything that moves or could cause a problem.
I didn't get any undercar pictures just yet, but basically I just have to mount the pump, mount the tank, cut the lines to size, and install the fittings on the ends. I will also have to figure out wiring for the pump. I was thinking I may just run one pump full-time and have the second one activated by the boost switch.
I can't stand the thought of that hose possibly being pinched above the rear subframe, with how much work I've put into improving flow through the rest of the system. I think I'm going to install around 1/4"-1/2" spacers. Probably via large diameter washers as a temporary measure until I know what thickness I need to take the pressure off that hose and then I will have some machined.
FWIW I doubt a small flattened area will create much flow restriction is a system with such long lines to begin with. I don't have anything to back that up with other than my instinct. Given the hose you used, the area for flow probably did not change, just it's shape. If you you were pumping a highly viscous fluid it might be more of an issue, but I doubt it with coolant/water.
FWIW I doubt a small flattened area will create much flow restriction is a system with such long lines to begin with. I don't have anything to back that up with other than my instinct. Given the hose you used, the area for flow probably did not change, just it's shape. If you you were pumping a highly viscous fluid it might be more of an issue, but I doubt it with coolant/water.
I know this is exhaust, not coolant, but air behaves as a low viscosity fluid when moving through plumbing.
I agree with you, it *probably* isn't a big deal. But I don't have any good way to determine how squished it is (very difficult to even get a finger or mirror up into that area with the subframe bolted up), and as much work as the rest of this has been I really want it to be right. Even 1/2" subframe spacers isn't going to have a significant effect on alignment or anything. I haven't looked too closely at the height sensor attach points to determine if it will change ride height or if it would just be changing ground clearance, but in any case it's an insignificant amount and would give me a bit of extra peace of mind.
I purchased a Peirburg CWA50 pump off Ebay which was delivered without a plug or wiring. I found a wiring diagram which I duly connected up but when trying to get the pump to operate on the bench, it doesn't work. My multi-meter found that the power terminal and the ground terminal seem to be connected. I checked both against the pump housing with the same result. Looking at the manufacturer's website they infer that the electronics in the pump itself will only turn the pump on when a certain temperature is reached. That still doesn't explain the short circuit between the main terminals. This is the circuit diagram I found which is for the pump to be controlled externally as it must because the SL55 only has a power and ground wire for the pump. Before I start complaining I thought I would check this with you folks. Any ideas? Cheers.
Did you measure resistance to ground with your meter? Usually with a power probe type device pumps and electric motors and such will still read ground on both terminals, but the resistance to ground will be a lot higher on the + side.
Did you measure resistance to ground with your meter? Usually with a power probe type device pumps and electric motors and such will still read ground on both terminals, but the resistance to ground will be a lot higher on the + side.
I just checked again and the resistance is the same for both terminals against the pump housing at 0.5 ohms. That would confirm something isn't right. The vendor advised that it takes 3-10sec for anything to happen when connected to power which was from my charger which showed 12.8 volts. He also confirmed that my hook-up was correct. Nothing happens after 20sec. I notice that the visible sparking is the same whether with just momentarily touching the charger leads together or after hooking up the negative and doing the same by touching the positive charger lead to the positive on the pump. I am waiting on an answer from the German manufacturer. I will keep posting. Cheers.
I just checked again and the resistance is the same for both terminals against the pump housing at 0.5 ohms. That would confirm something isn't right. The vendor advised that it takes 3-10sec for anything to happen when connected to power which was from my charger which showed 12.8 volts. He also confirmed that my hook-up was correct. Nothing happens after 20sec. I notice that the visible sparking is the same whether with just momentarily touching the charger leads together or after hooking up the negative and doing the same by touching the positive charger lead to the positive on the pump. I am waiting on an answer from the German manufacturer. I will keep posting. Cheers.
That sounds like a defective pump to me, but who knows.
Small update on my project:
The rear subframe has pilots from the car so I can't use spacers without also extending that pilot. I relocated one of the hoses a bit...it makes the routing less than ideal, but doesn't pinch it nearly as badly in the new location.
Going to literally velcro the rear pump to the frame of the car for now. It's pretty much a wedge fit where I am putting it so it won't take a lot to hold it in place. Don't really care about vibrations from it since it will only be activated by a boost switch. Velcro will allow me to fine-tune the placement, and I will build a proper bracket when I get the final spare tire tank built.
I'm kind of rushing through it right now because I would love to get it to the track tomorrow night (if weather cooperates). Just pretty much gotta wire the pump and run the two new lines up front (pretty much one actually, I think I can re-use part of a stock line), then fill/leak test the system.
Ok, so I made some good progress tonight. Still don't know if I'll have it done for the track tomorrow night...I should make it if I don't hit any major snags with the final details (basically if I don't have any leaks, lol). I have the pump "mounted" and plumbed. Just have to wire it up. I got the Y and valves into the engine bay and the hood actually shuts. Time will tell whether or not it is going to hit/wear on the underhood insulation. I had to grind down and polish the barbs on the valves to get them to fit inside the 3/4" hydraulic hose. They are actually 1" pex valves, which is only slightly bigger than 3/4" barbs. Basically, they are the largest valves I could have used and very nearly match the ID of the hoses. The feed line comes up behind the driver's headlight into a Y right there and then splits, one line to a valve that then attaches to the stock line coming from the intercoolers to the hot side of the heat exchanger. The other line is the heat exchanger bypass loop. It goes across the bay on top of my now very tilted radiator (due to the giant new heat exchanger), with a valve installed in the most convenient place I could find space-wise. It goes back down behind the passenger headlight and tomorrow I'll figure out how to Y it into the heat exchanger outlet/stock pump inlet.
All I have left to do up front is figure out how to Y into the stock pump now. It's going to be a real trick since space is so tight down there. I may end up utilizing some more of the stock molded hoses to achieve a nice smooth 90 or two where needed.
The only thing I'm really not happy with is the feed line from the tank to the rear pump. It hangs too low, and really needs a tight 90 to hug the underside of the spare tire well. I drilled the hole larger and I will be able to tie it up some without kinking, but not all the way. I may have to re-think my plan regarding the whole tank inside the spare tire thing in order to come up with some plumbing that isn't going to hang down and be so visible.
I will probably paint the valves black or something to make the install a little more stealth. But I'm pretty happy with those locations, it will make the changeover pretty convenient as opposed to having them underneath the car hidden in the corners of the bumper or something.
Wow, a lot of work put into your setup. Only thing I wonder about is your hose hanging so low under the trunk. I'm sure it won't drag, but just looks a little ugly like that. Would be nice to run it out the side of the tire well right into the pump. I'm guessing that wasn't an option with your tank.
Wow, a lot of work put into your setup. Only thing I wonder about is your hose hanging so low under the trunk. I'm sure it won't drag, but just looks a little ugly like that. Would be nice to run it out the side of the tire well right into the pump. I'm guessing that wasn't an option with your tank.
With the spare tire in place there is no decent routing except right through the bottom like that. It might be possible if I run a pickup tube and run the line out the top of the tank and down but I wasn't sure how well that would work in terms of pump prime. I hate that it hangs. So much. I have gone to great lengths with this car to keep as much of a stock/classy appearance as possible. It is better now that it is tied up a little way. Maybe only level with the mufflers now. Not sure what to do. Track is closed due to weather so now I have time to think it over some more.
Didn't realize you still had the tire in there. Maybe a 90* fitting? There are some nice ones with smooth bends if you look for an AN type fitting.
I have grandparents in their 90s about 450 miles away that I go to see every couple months, and access to a family cabin in Colorado that I go to twice a year or so, so I do put a lot of miles on the car and want to be able to retain the spare just in case.
My vision was more or less that once I get the permanent tank fabbed, it will have a line attached to the bottom that could be disconnected under the car and then pull straight up out of the hole without any issues. If I put a 90 on there, the hole would have to be huge. Or I'd have to figure out some way to build in some slack to pick the tank up far enough to disconnect it at the bottom. But that may not be the best plan, or even necessary.
I have been thinking more and more, and with the less than optimal routing anyway, that it would be best to just have both lines at the top of the tank so it wouldn't even necessarily have to be disconnected to access the spare tire (if I could build in enough slack in the lines and still get the carpet to sit flat). Just a pickup tube running down to the bottom of the tank. Or even just shut off valves on both the lines and the tank for removal with no spillage.
Ok, so I fixed the low-hanging feed line. It now runs through into the trunk out of the way of the tire somewhat, as opposed to through it from underneath. Which means my spare tire tank will have to have a pickup tube, but I think it will work out just fine and will certainly make it easier to remove the tank if I were to get a flat. I had to buy a molded heater hose to make it work, I couldn't believe it was $20 for a little bit of hose, but whatever...it will look a lot better.
I also got the Y into the front pump figured out. There wasn't room to run a bit of hose, the Y, and then the other hoses. So I took the pump head off, cut the stock barb off, cut down the Y, and epoxied it to the pump housing. This worked out well, but I suspect if I ever have to pull the hoses off it might break and have to be redone. Unfortunately, when I went to leak-test the system I found out that I did crack one of the Ys behind the intercoolers when I was trying to get the lines on the intercoolers. So that will have to all come apart again and be fixed before I can fill it up and report on the performance. Pics:
Ok guys, I just vacuum filled the system and did a flow test. Actually, I did 4 flow tests (one in each possible configuration). My setup was very simple...I essentially temporarily ran some hoses from my rear shut-off valves to a 5-gallon bucket and turned the pump on. Started a timer and at 5 seconds (needed both my hands so I had to start the timer and still have time to set the phone down, grab the hose, and have a few seconds left to time the jug swap), put the return hose into a gallon jug and recorded the ending time it took to fill, then subtracted the 5 seconds. Crude, but at least a ballpark test. Results:
With just one pump running and the bmw X3 radiator in the loop, it took 13 seconds to fill the 1 gallon jug. With one pump and the bypass loop open (radiator loop closed to simulate ice-box run at the track), it took 12 seconds to fill the gallon jug. With both pumps running through the x3 radiator, it took 9 seconds. With both pumps running through the bypass loop it took 8.5 seconds.
So to convert that to gallons per minute I was getting 4.6gpm (heat exchanger), and 5gpm (bypass) with one pump. And 6.6 and 7 gallons per minute with both pumps (they are effectively in series). I'm very pleased with these results. This was with a 13.8V power supply on the battery with the car not running to ensure consistent voltage for all the tests. I did take two videos comparing the best flow scenario to the worst flow scenario and I will post links just as soon as I get them uploaded.
It's kind of hard to tell the difference from the bucket videos, so I took some reservoir flow videos as well. They are all super short because I was uploading them without wifi and didn't want to eat up a ton of data.
One pump through heat exchanger (lowest flow) bucket:
I think that will be adequate agitation for optimal heat transfer between the ice and water. Very pleased. Can't wait to build the actual spare tire tank version now and get my trunk space back, lol.
In my RHD SL55 I have installed a tank where the starter battery was (now in the trunk) in which the water level will be the highest point in the I/C system. I will be using a Pierburg CWA50 pump which I have a couple of ways to plumb. Should the pump transfer water to the I/C then into the tank which then will feed the H/E, top to bottom (or reverse?) then back to the pump OR should the pump transfer water first to the H/E bottom to top (or reverse?) then the I/C then the tank back to the pump intake. It would seem the back pressure would be less with the first way.
Also, if I run both the OEM H/E and the new H/E should they be plumbed in parallel or series? I have scoured these threads and find conflicting information. Thanks in advance.
05-24-2016, 03:42 AM
