Aj's 818S in Houston

[Rob T]

Just check where you might want to jack the car from some day.

[Ajzride]

Just check where you might want to jack the car from some day.

I guess I'm a little perplexed. I'm talking about the braces inside the cockpit that run from the side rail down to the floor right in front of the seats, making a triangle at the front of the seat pan. Not sure how those would affect jacking the car up, unless you are saying I should be concerned about the folding in towards the cockpit if I use the wrong jacking point. If that is the case, I should be safe, I would always jack it up at the front or rear firewall, not in the center where the seats are.

[Ajzride]

I’m using the stock seat belts, but they just flop around on the floor. Because they have the 90 degree bracket on the bottom, they won’t fit in any of the traditional sleeves. Made up some quick 3D printed sleeves to keep them where I want them.

692FDB09-9899-4A13-8FF4-01570B510AC3.jpeg

27E6AB1F-EEB4-4498-ADE8-C66258E2DE3A.jpg

58B31FB2-416F-42D1-A0B4-35196D79DBED.jpg

[aquillen]

You've just about got me talked into adding a 3D printer to my hoard. Been resisting for sooo long.

[Ajzride]

There's so much stuff I don't put on here too, it's invaluable. I made a quick bracket to mount the AC Drier today, took less than 15 minutes to CAD and 20 minutes to print. I couldn't do another car without it.

[Rob T]

I misunderstood the part you were talking about. I successfully jack the car at that point, as did Chad. You can get both wheels off the ground at the same time. The lift point is right at the place where all of those parts come together. I am not qualified to make any comments about the impact on structural integrity or stiffness.

[Ajzride]

I had hoped to finish the AC lines and pull a vacuum + charge this weekend, but alas it was not meant to be. I had made a trip to the junk yard a few weeks ago to pick up some factory hoses that connect to the compressor because it appears to be a proprietary connection, and I had ordered some bead-lock to bulkhead fittings so that I could crimp them on and then connect to the rest of the AC system with standard o-ring connectors. But the actual rubber hose on the factory lines is some oddball size, at least the bigger one is. The smaller one was a #8 reduced barrier, but the larger line was bigger than a #10 and smaller than a #12 in reduced barrier, and bigger than a #8 and smaller than a #10 on standard barrier hoses. I wound up leaving them with the hose shop and they are going to cut the hard pipe and weld on a #8 and #10 nipple for me so that I can connect them to the rest of the system.

Today I focused on the cooling system, I went through a bunch of trial and error stuff, but in the end I drilled six holes in my thermostat and it is working, although at the moment my heater core is bypassed, so I still have to hook that back up and rebleed the system. I've ordered some high temp clear silicone hose that I'm going to put on the inlet and outlet of the block for the heater loop so that I can always monitor the flow, it should be here by next weekend.

Here is half of the holes I drilled in the thermostat:

Holy Thermostat.jpg

Turns out I probably would have been good with only 3 holes, but due to getting myself confused and not doing things in proper order, I convinced myself I needed more, so doubled it. with six holes it takes 16 minutes at idle to warm up enough for the fan to come on.

While I was messing with all this, I noticed my valve covers are getting grimy, it looks like oil mist, not a real leak. Anyone have any pointers on where to look, or is this just normal for a subaru?

Dirty Girl.jpg

[Ajzride]

I've tried not to post a bunch of mundane stuff in the build thread, a lot of the stuff I'm doing has been covered by others over the years, so I've tried to stick to the stuff where I go off script, or where I'm needing help. Today I got the AC system charged after I got the welded fittings back from the hydraulic shop. Nothing very fancy about it, it just works and I'm glad to be very close to moving into the next phase of the build. I still have to finish the suspension modifications of trimming the passengers upper control arm and installing the camber bolts, perhaps I'll get to that tomorrow, but then I should be mostly complete with mechanicals and starting on the litany of custom stuff I'm going to be working on. First up will be finishing the electronics for the electric parking brake, that is actually almost done, I just need to 3D print a case, add some terminal strips and get it mounted and do a big write up on it. Then it will be mounting the factory dash, which I will probably be narrowing so that I don't have to trim the door panel. After that I will work on the custom center console before starting on the body mods. I'm looking for a little advice here on which order to attack things in.

I'm thinking I should start with the side panels, get the scoops enlarged and get those mounted, and then moving on to the doors. For the doors I'm going to be either ordering the FFR coupe door frame and window kit, or trying to retrofit some glass and regulators from a porsche boxster. I'm also planning to flip the latches around so that the striker is on the door and the latch mechanism is on the body, this way I can have all of the electronics for the door poppers mounted in the body and not swinging in the door. If you have watched FastThings GTM build on youtube, this is how he tackled it and i think it worked out well. However this is an 818 and not a GTM so if anyone thinks this is a bad idea please let me know. Once the doors are modified with glass and poppers, then I can mount the windshield frame and start working on the hood/fenders/nose, which is where 90% of the customization will be at.

Thoughts?

[Ajzride]

This is the point where frustration builds, dejection sets in, and the build gets stuck in a storage unit and not touched for years. Ive now been fighting overheating for 105 days. Yesterday I made some modifications to the cooling system to try and help with filling the block/radiator/HVAC all at once with the thermostat installed. My change was to put a fill neck right where the coolant exits the block. Other than the degas tank, this is the highest point in the system, it has an unimpeded path to the radiator and the block. I then tapped that fill neck (on the bottom) and installed a 5/8 nipple which I used a TEE to connect to the heater supply line. In that line between the new filler cap and the TEE, I installed a shutoff valve. So I can fill the system at this new spot, from the highest point, and it connects directly to all 3 coolant reservoirs. Once the system is filled I shut off the valve feeding the heater line TEE, and the system functions as normal. After all of that, the heater still wasn't working, and I relized there is an issue with the heater valve not opening. I had thought it was open because when I pulled a vacuum on the system the hoses would collapse, but apparently that just means the valve is water tight but not air tight.

Jump back to today, I just want to be able to go-kart the car for the memorial day weekend, so I put the U-hose back on the heater inlet/out and completely bypass the heater core, and vacuum/fill the block and radiator. I also removed the thermostat during my troubleshooting and have not reinstalled it yet, so no reason I should have any trouble keeping the car cool. However, I'm still seeing bizzare behavior. The car takes a long time to warm up (expected with the thermostat removed) and you can hear water running through the radiator (sounds like a garden fountain), but once the car gets hot, the water seems to stop moving. After you hit about 200, the temperature starts to take off really fast and if the car is left at idle(750 - 800 RPM) it will overheat pretty quick. If I rev the car up to about 1500 RPM, the temperatures will drop like a rock from 235 back down to 165 in a matter of 30 seconds. It's like the water is just sitting in the radiator getting cold until you rev it up and then it dumps all the cold water into the block.. It's really strange because it was obviously flowing really well until it got hot, there was a nice even temperature profile around the entire system, and it took a long time to heat up.

I'm completely stumped, is this is a symptom of still not burped, or could this be a symptom of a bad water pump that won't push flow at low RPMs once the system starts to build pressure?

[Bob_n_Cincy]

Alan,
give me a call, I'll help you get the cooling system fixed.
With the mods I did, I just fill it at the degas tank, rev it up a couple of seconds 3 times. Then top it off, and I am golden.
Bob

[Ajzride]

Alan,
give me a call, I'll help you get the cooling system fixed.
With the mods I did, I just fill it at the degas tank, rev it up a couple of seconds 3 times. Then top it off, and I am golden.
Bob

Thanks for the offer Bob, I promised the family I wouldn't work on the car over the holiday, so I'll touch base next week.

[Frank818]

Oh man AJ I understand so damn much how you feel. I know Bob will fix your problem, he's a coolant expert, so you can sleep well over the Holiday.

[grabera7]

Alan,
give me a call, I'll help you get the cooling system fixed.
With the mods I did, I just fill it at the degas tank, rev it up a couple of seconds 3 times. Then top it off, and I am golden.
Bob

Bob, can you link to a post for your coolant mods?

[Bob_n_Cincy]

Bob, can you link to a post for your coolant mods?

The basics
A> radiator should be coolant in high and out low.
B> I blocked of the fill cap on the top of my radiator. Some Subaru radiators don't even have a cap. https://www.rockauto.com/en/moreinfo...rUAnkAzV47ibjr
C> On the top of my radiator, on the discharge side ,I have a 5/16 bleed hose connected at the highest point..
D> On the discharge hose off the top of the engine, I have a 5/16 bleed hose connected at the highest point.
E> in hose "D" above I have a pill ,with 1/8" hole, in the hose to limit flow.
F> Hose "C' and "D" connect together with a "Y" just before going into the top of the degas tank.
G> Try not to have any of the coolant system hoses have a high spot that will accumulate air bubbles.
H> A small 1/8" hole in the thermostat will help clear any air in the radiator hose going to the engine. I did not do this.
I> running without a thermostat may cause the system above not work to clear the air in the radiator.



The pressure in the system is highest right after the pump. This setup will have a small flow off the top of the engine and radiator back to the degas tank. So any air in the system accumulates in the degas tank. You should check and top of the degas tank after some heat cycles. The degas tank will self refill on every heat cycle as long as you have fluid in the overflow bottle. you overflow bottle will tell you what happening. If it goes down on a daily basis, you are getting more air out of the system. If it goes up. more air is accumulating in the system, and you are getting into trouble.

My system is based of the Pantera system (mid engine car) without a swirl tank

pantera.jpg

Here is my thread when I was testing my system.
https://thefactoryfiveforum.com/show...l=1#post228056

On a side note. If you go through exotic measures to bleed air from your car. You are asking for trouble down the road. Small air bubbles will travel through you system. The will eventually accumulate in the radiator. Slowly defeating the radiator cooling effectiveness and causing you to overheat.

[Ajzride]

Bob

Did you mean to include a picture that has hose C and D on it, or were you referring to my diagram?


NVM: I get you were referring to you bullet list. DOH!!

So a few questions.

C> The bleed hose on the discharge is at the top of the radiator, or the top of the discharge hose?
D> It seems like this is very close to where you put Wayne's mod. I already have Wayne's mod, could I just Y that into the bleed from the radiator, or do I need both? Any pictures of how you are adding a 5/16 tap into a coolant hose would be nice. I didn't see any reading your thread from the link you gave us, forwards about 4 pages.
E > Not sure what a pill is, google was not helpful either, probably because I don't know how to phrase my search.

[Bob_n_Cincy]

C> The bleed hose on the discharge is at the top of the radiator, or the top of the discharge hose?
D> It seems like this is very close to where you put Wayne's mod. I already have Wayne's mod, could I just Y that into the bleed from the radiator, or do I need both? Any pictures of how you are adding a 5/16 tap into a coolant hose would be nice. I didn't see any reading your thread from the link you gave us, forwards about 4 pages.
E > Not sure what a pill is, google was not helpful either, probably because I don't know how to phrase my search.

Hose "C" needs to be at the highest point of the radiator, So pressure from the water pump pushes any air in the top of the radiator back to the degas tank.
wrx.jpg

Wayne's mod is OK. It would be better if it was on the top and the highest point. I would use something like this right after the top of the engine. https://www.summitracing.com/parts/uro-1j0121087b

I made m own "pill" out of a piece of 5/16 rod 1/2" long. With a 1/8" drilled the long way. Put this in your hoses and air will go through pretty fast, but it will really restrict fluid flow.

[Ajzride]

Thanks Bob, here are my thoughts:

A) Putting a 0 PSI radiator cap on the radiator, and use the overflow nipple for the vent, it's definitely the highest point of the radiator, run that back to the degas tank.
B) For the vent on the rear, I installed an inline radiator filler neck (see below) even with the top line of the engine, I can put 0 PSI cap there and use the vent nipple back to Y into the radiator vent and take back to the degas tank. I think this gives me the same result as the Yipe you linked.
C) Since I would be using the vent nipples, they are smaller than your 5/16 line, I might not need a pill?

https://smile.amazon.com/gp/product/...?ie=UTF8&psc=1

[Bob_n_Cincy]

Thanks Bob, here are my thoughts:

A) Putting a 0 PSI radiator cap on the radiator, and use the overflow nipple for the vent, it's definitely the highest point of the radiator, run that back to the degas tank.
B) For the vent on the rear, I installed an inline radiator filler neck (see below) even with the top line of the engine, I can put 0 PSI cap there and use the vent nipple back to Y into the radiator vent and take back to the degas tank. I think this gives me the same result as the Yipe you linked.
C) Since I would be using the vent nipples, they are smaller than your 5/16 line, I might not need a pill?

https://smile.amazon.com/gp/product/...?ie=UTF8&psc=1

I would still put a pill in it. With a pill, both engine and radiator will be cleared of any air.

[Ajzride]

So this week I worked through the following modifications:

>> Put a 0-PSI cap on the radiator and used the vent nipple to run a 5/16 line from the radiator to the expansion tank. Added a pill to the line on the radiator end
>> Put a 0-PSI cap on the filler neck that I added right where the return line exits the block and used the vent nipple to run a 5/16 line to to the expansion tank. Added a pill to the line near the nipple
>> Added tee to the heater core supply line right where it exits the block. Used the tee to add a 5/16 line with a pill in it running to the expansion tank.
>> Installed the overflow tank and connected it to the expansion tank. Previously the expansion tank vent nipple had just been blocked off.

I had never installed the overflow tank because I didn't realize it was actually integral to the cooling system. On my last build (65 mustang), you just had a puke bag, and all of the expansion, contraction, and burping happens at the radiator. Having installed it and watched the system work, I can now it is very important to the Subaru setup. You can watch it while you are warming the car up and see all of the air bubbling through it as the system bleeds and burps.

I spent a long time yesterday chasing down perhaps a phantom leak. I could not get the vacuum system to pull a vacuum, it would max out at 5 inHg. However when I pressured the system up to 25psi, it held and didn't leak, so I'm kind of confused. Perhaps there was just a lot of water in the system still that was boiling off.

The car is heating and cooling well now, but I still don't have any flow through my heater core, I think it is still full of air. I have a leaking AN fitting at the bulkhead on the inside so I have to pull that anyway, I may try to pre-fill the heater core with coolant while I have that line loose.

thanks to Bob very much for his advice on setting up a vent system, and to Pete for keying me in to how important the overflow tank is. Also big shout out to KC for making up some pills on the lathe for me, it would have take half a day to make those with only a drill press and bench belt sander.

Going to try and figure out the heater core today, and once I know have it all working, then I have to drain the whole thing and fill it with real coolant. I've filled and drained so many times in the last 3 months I started using straight water so I wasn't putting so much ethelyne gycol down into the storm drains.

[Ajzride]

I'm pretty sure I have a bad water valve. It's a 4-way valve that is supposed to handle recirculation for the motor when the heater core is closed off, but I took it off the car and tested it with a water hose (5/8" just like a heater hose), and any back pressure at all, even just turning the valve on its end so that that gravity is fighting against it, and the water just recirculates and does not take the path for the heater core, no matter what position the valve is in. I think the flapper is loose on the shaft or something. I can't get it to flow through the heater lines. I know they are full of water and free of air because I filled and purged them manually. I'm going to look into how to make my own bypass and use the 2-way ball valve, it will seal much better.

[Ajzride]

The 4-way valve I bought from restomod air was labeled as a "Coyote" valve, because the Ford 5.0 requires the same constant cooling loop that the Subaru does. So I headed over to the Cobra section of the forum to see what those guys were using to accomplish this.

A lot of them use a 1996 Ford Ranger 4-way valve. This is controlled by a vacuum canister that can either be rigged up to a vacuum solenoid, or remove the canister and hook up a cable. That solution doesn't really appeal to me because I want to keep it integrated with the logic for the rest of the HVAC system. The other option they were using was an H pipe with a crossover that is smaller than the supply and return. The theory is that the 5/8 line that goes to the heater core and back should be less restrictive than the smaller crossover pipe, and therefore flow should choose that path when the valve is open, but still use the crossover to recirculate when the valve is closed.

https://www.summitracing.com/parts/v...BoCbCEQAvD_BwE

vta-344470_xl.jpg

I'm not really in the mood to wait 10 days for something to show up, and also I'd hate to waste another $40 if it didn't work. However I figured I could probably find enough stuff laying around plus a cheap trip to Ace to make something up and test the theory. What I wound up with was two 5/8 x 5/8 x 3/8 Tees which I fashioned into an H-pipe, and then installed in-between the block and the 2-way heater valve (The tee in the top line is the bleed line with a pill in it):

Redneck H-pipe.jpg

This proved as fruitless as everything else I have tried. At this point I just don't understand why I can't get hot water to flow to the heater core. I can trace the hose from the block to the heater core, and it starts out nice and hot, and cools off as you move along it until you to about halfway down the passenger side, then it is ambient temperature. I know the heater core works because while I was doing some troubleshooting I blocked of the recirculation line of the 4-way, which forced the water to go through the heater core, and the hoses got hot and warm air started blowing.

I really am completely stumped as to why I can't get the flow to choose the heater core over the recirculation line. I would think that maybe it wasn't taking the recirculation line either, except that I was able to force it through the heater core by blocking that off.

At this point, I'm exhausted, my hands are raw from wrestling hoses and hose clamps all day, and I'm ready to just sell the damn thing for $500. I use the stepless clamps because I hate the worm gears, but I blew through 40 of them today between moving the heater valve twice and then replacing it with the two way. Also, make sure you have all of your work along the firewall done before you install the windshield surround. I put it on about a week ago to check how the 2010 dash fit against it, and it made working on the heater valve and bulkheads a huge nightmare.

[Ajzride]

A 1996 ranger valve is only $12. I ordered one to see if I can make the actuator I have work with it.

[lsfourwheeler]

I'm running a Vintage Air H-pipe on my car and am not having any issues with heat. Looks identical to the Summit picture you posted. Once the car is up to temp, hot coolant flows through the heater core just fine. You sure you've gotten all the air out of your system?

[Ajzride]

I'm running a Vintage Air H-pipe on my car and am not having any issues with heat. Looks identical to the Summit picture you posted. Once the car is up to temp, hot coolant flows through the heater core just fine. You sure you've gotten all the air out of your system?

No I’m not, but I have no clue how to get it out if it is still there.

[Ajzride]

The Ranger valve came in today, It's identical to the Restomod air valve except it has a vacuum actuator on it instead of an electrical one. I tested it with the water hose and it sealed perfectly. So I started looking at how to move the electric actuator to the Ranger valve, and when I took it off, I found that Restomod air had just cut the metal arm from a vacuum actuated valve off short, made a small adapter from metal and ran a set screw through it. A little playing around I realized there was nothing wrong with the restomod air valve, it sealed just fine with the actuator disconnected and turning the shaft by hand. It appears the issue is getting the actuator which has a 90 degree movement lined up exactly with the valve that also has a 90 degree movement. Any misalignment and the actuator will over travel, sense high amps, and back off. It's incredibly difficult to line the actuator up with the valve perfectly because the set-screw can only be accessed to tighten at certain positions, and full open or closed are not in that range of motion. After fighting with it for about 4 hours, I've given up on trying to align it using the set screw, I've put a little JB Weld on the shaft and stuck it together at full open, and once that sets up I will rotate it and tighten the set screw. Maybe that will fix the valve sealing.

Still not convinced that is all of my problems though, because under the current circumstances, I will only be able to have full heating or full cooling, If the valve is anywhere in between, all of the water will go through the recirculation line and nothing to the heater core. Very confused about that, and why the H-pipe doesn't work. I still think I have a bad water pump that doesn't make sufficient head.

[Ajzride]

I just can't get the 4 way valve to work with the electronic actuator, and I'm not sure it ever will work right based on how an electric actuator works and how the valve is designed to be held tight with a spring. I did however make some more progress today and might actually be getting close to a resolution.

It appears that part of my issue was not following Bob's directions to T, and trying to expand on them. I placed a bleed line on the outlet of the block going to the heater core so that it would bleed off any air leaving the block and heading to the heater core. I figured this would actually catch a ton of air because when the engine is cold, this is the primary circulation path for the coolant until the thermostat opens up. Well I tested the entire heater circuit today with the water hose, and it appears that a 5/16th line with a 1/8 pill in it has less resistance than a heater core. The water flowing up out of the bleed line instead of going through the heater core. I was able to solve this by moving the tee for the bleed line to the return circuit of the heater core, although now any air that is in the system will have to make it's way through the heater core before it can bleed off. I'm not overly concerned about this because that is exactly what would happen if there was no bleed line there at all.

The next experiment I worked on was using the 2-way water valve with no h-pipe. The theory behind the h-pipe is to allow coolant to recirculate when you close off the heater core so that thermostat can still get hot water and open up. Since I drilled two 15/64" holes in my thermostat, I postulated that I might have enough coolant flow even with the heater valve closed to warm up the thermostat. This proved successful today (85 degrees and sunny). I will want to try this at least two more times to verify that there are no issues. Once it was warmed up I opened the valve and got plenty of hot water all the way through the heater core and very warm air from the unit. If I can indeed use the two way heater valve with no h-pipe and count on the holes in the thermostat for circulation, I will be at the end of my misery for the coolant system. If this does not prove to be 100% reliable to open the thermostat, then I will be back to figuring out why I can't get an h-pipe to work. My best guess at the moment is that the heater core from restomod air as a lot of small tubes versus fewer big tubes (compared to the vintage air unit or the Subaru unit) and it is requiring much more pressure to push through the core.

[aquillen]

Every time I read through this I get the feeling your heater core is partially plugged shut by a defect in the core. No heater core I can imagine should have more flow resistance than a 1/8" pill hole. You said the 1/8 pill flows more than the core which just can't be right to me. Your successful testing could also come out just as you say in post 306 with a blockage going on in that core, but it won't heat like it should in real use due to low flow = low thermal transfer capacity. Even if it was constructed with a lot of small tubes inside, it would be built that way with a header on each side of the tubes and in the end would flow very freely due to the multitude of tubes in parallel. If it is a single multi-loop tube it may be kinked or otherwise bad somewhere inside there. Such a blockage in the heater would also inhibit the H-pipe setup from working as intended.

From guesswork only, I'd think a typical heater core would pass a strong flow if connected to a 20# to 40# water hose running from your tap water and certainly more than a 1/8" pill blocked hose. Can you get some cover off that heater and inspect the tubing/assy from the outside at least?

If you cannot get access to the heater inside to look over the tubing, I can suggest a rather crude but probably realistic test. The internal length of the core's piping is probably an effective length of about 10 feet. Get that length of cheap hardware store plastic tubing in the same ID. Connect each to tap water supply (if well water - check what pressure you are using during each test). Run water into a bucket for X seconds and compare the performance of each. Should be in the same ballpark. Without two cores to compare this is pretty crude but I would expect the flow rates to be maybe 50% apart or less.

[Ajzride]

Every time I read through this I get the feeling your heater core is partially plugged shut by a defect in the core.

That thought had not occurred to me since it is a brand new unit, but would make more sense than a water pump that is not generating enough head (but still manages to cool the car just fine). I think I should probably call the vendor and ask them if they have a prescribed method of testing it, I wouldn't want to do anything that would void my warranty, which I should still be under. There is certainly no way to test it without dumping my freon charge, although that wouldn't be the end of the world.

If the rain blows out of here this afternoon I'll make a video showing the test with a hose.

[Ajzride]

Recreated the test on heater core flow and videoed it.

[Jetfuel]

Are you holding the nozzle with some force to keep it from blowing back?
Restriction to flow makes pressure

[Ajzride]

Yes, I'm holding tight. When I did the initial experiment I wasn't using the spray gun, I cut the hose and stuck it on the barb with a clamp, so no extra pressure. That had the same result. I just didn't feel like cutting my hose and replacing the tip again (it needed to be done yesterday).

[aquillen]

Good test in your video. But that valve you have in there may be complicating the results. I suggest you remove the valve and connect your source water directly to the heater core line just past the valve. No vent line or pill at all in the next test. See if you get large flow (you should) or still get a trickle out of the core (= core/hose problem).

In your video, either the heater core or that valve is restricting the flow which is why you are getting very little flow through the core. This is also why you get nothing out of the vent line in the last part of your test where you went "backward". In other words, feeding in the normal direction, the restriction in the heater section causes plenty of pressure at the vent - pill line and you see high flow there. Whatever it is, it is that same restriction in the other direction that drops pressure before it gets to the pill so there is not enough left to push any water out that line. You are showing that you have a restriction in the main flow path - valve or core. And that does not look like a ball valve (may be a globe valve). Ball valve would be a much better choice if it is not one now and you need to have it in there.

[jforand]

I think I would put my vote in for the bad heater core. There should be no restriction to flow as aquillen suggests, especially at the flow rate your hose was producing. I have chased this on other cars and spent days trying to dissolve out deposits and then of course deposits that were flaking off all piled up in the heater core as the lowest and most remote part of the system.

I know you say your core is new, but I think it is defective. Likely excess solder that has clogged a bunch of the flow. I am thinking you have little choice but to pull it and test the flow directly. Again, they are designed to introduce near zero pressure drop at the nominal flows they see, which is not dissimilar to your hose.

[aquillen]

jforand has it. But I would say as long as you ONLY have a correct size hose into and back out of the heater core you can test it for good flow using those lines, just don't have other things in the path like that valve, the T to the pill and so on. If you still have low flow in that straight forward path in and out of the heater then pull it and plead your case with mfgr.

[Ajzride]

I can pull the valve but I don't think it is the issue. I'm thinking what I will do tomorrow is perform the same test at the bulkhead, eliminating all hoses, valves, etc (everything except the heater core).

[jforand]

If going to the bulkhead cuts everything out except a straight hose path, as aquillen suggests, you will get the answer on the core.

Good luck, hope it locks things in for you.

[Ajzride]

Well I certainly didn't get the results I expected. The summary is that there is no flow drop through the heater core, and a moderate drop through the entire circuit at a high flow rate (4GPM) but virtually no drop through the entire circuit at the flow rates the water pump produces (1.15GPM).

I'm not sure if I should be chasing the flow drop I'm seeing, or marking it up as acceptable and focusing on the overall low flow rate, or just saying it is all good and moving on with my life.

Here is a chart summarizing my results:

FlowResults.png

I started out with what i thought was a reasonable flow (3 GPM) and tested the flow drop through the core itself, which was nothing. Next I started chasing flow drop through the circuit thinking all of the 90s I had added to keep the hose routing clean might be causing an issue. There are 3 on each line so a total of 6 on a round trip. While this did add up to a 21% loss, it didn't feel like it should be enough to justify what I was seeing with the water wanting to jump through the air bleed line instead of the heater core circuit. I decided to check how much water the circuit actually flowed while at idle, and it came out to roughly 1.15 GPM (albeit inconsistent). So I rechecked the entire circuit at 1.15 and only came up with a 4% loss. This actually makes sense to me based on the 1 fluid mechanics class I had in college, because pressure drop through fittings is exponential with flow rate, especially if you cross over from laminar to turbulent flow.

I'm utterly confused and have no answers. Since the heater is blowing hot air (126 F) and the system is cooling, I could just walk away from it, but I'm afraid it might come back to bite me in the rear later since I know it's not functioning as designed.

Certainly open to observations I might have missed or suggestions.

here is a video of all the testing if you are inclined to watch and ensure I didn't do anything stupid.

[jforand]

Hmm, it might be time for me to step out of the conversation as I have not covered any of this ground yet. Just mounted my water pump to my engine rebuild

My final thoughts:

I will say that the water pump is pretty small in comparison to many other engines. I am actually quite surprised that it works well at all for our 818's. You mentioned the 90s which are certainly a source of head loss, but the line is as well. In the OEM world everything was up front and very short tubing runs. Here we have the engine behind the driver and the radiator and heater core very far forward. There is tons of tubing (and fittings) added to the system over what the OEM design was. The pump will reduce flow as the head loss in the system increases as you noted. Most people think that to deadhead a pump you must cap off the tube, but in reality you can dead head a pump if you hook a long enough (and/or small enough) tube to it. When the system losses exceed the pump's head you get no flow. One way to get a quick demonstration of the phenomenon is to take one of your long brake lines and try to manually blow out the tube from one end (something you do when cutting and flaring). You will quickly find it is pretty impossible and you need the additional pressure a compressor provides.

Not sure of the size tubing you are running, but as you go longer you would need to go larger to avoid adding head loss and of course each of the fittings is a sharp change in direction coupled with an inner diameter restriction.

I am thinking that you don't have enough pump to push all of the plumbing as it exists. The part that bugs me here is that others are saying they have no issues. Up'ing the RPMs increases pump pressure and will force more water flow. The heater core I talked about flushing was in a BMW Z3 and it clogged. Long story short, there was a bunch of scale in the system from lack of maintenance, when I did maintenance (new coolant and distilled water) the scale started to soften and break free. It basically all ended up in the heater core. The symptom was diminished heating. Eventually, the heat was gone at idle (sitting at a school bus stop) and would return if you rev'd the engine. We needed the excess pump pressure to push through and around all the gunk and debris. I know your system is clean so it is not a scale issue, but I think the head losses and pumping pressure is very much what you are dealing with.

[lance corsi]

I’m wondering if the addition of an external water pump would help. On my 818, I used a Stewart in-line 40 gpm pump but I removed the impeller from the stock pump before installing, thus rendering the pump to a pass-thru housing. Also, I removed the thermostat and am controlling the pump via a signal from the coolant temp sensor located on the upper crossover plenum.

[Ajzride]

This barb on the upper cross-over pipe is what used to feed the degas tank in the stock setup. That barb gets plugged when adding Wayne's mod.
Attachment 124632

Turns out my crossover pipe doesn't look exactly like that, but it does have a barb (smaller and further right) that leaves the crossover pipe and connects to the hard pipe that returns the heater loop to the thermostat. So that line should be heating up my thermostat even if the heater doesn't flow.