water jacket weave porting
 

i'm going to attempt to resurrect this seemingly dead art by giving it a new name!

since these days rotaries have been going onward and upward exponentially over the past few years i thought this would be a good time to attempt porting the water jackets and try to get a comprehensive result on if they help or do anything at all. keep in mind this will take some time to actually prove whether it is worthwhile or not at all.

https://img413.imageshack.us/img413/2439/dsc00155oc.jpg

the point of the modification is to slow water flow across the surface and increase the surface area in the hottest part of the engine(the combustion area durr!). by increasing the surface area and slowing water flow it increases heat exchange and increases cooling efficiency in that area. the spark plug bosses have a tendency to crack, meaning the engine is running far too hot in that area so cooling it more would be beneficial.

the other reason i see for doing this is to help ignition breakup and additional apex seal cooling through the housing, the hotter the housing gets the less the apex seal can cool itself and can flex(too much flex can cause it to catch the exhaust port opening). increased internal temperatures also increase the possibility of a detonation cycle to occur.

this modification is nothing new to rotaries, the last i heard about it was about 5 years ago and doing a recent search leads to that being the end of the discussion here on this forum.

now i don't foresee myself trying to sell this service because it takes a gross amount of labor hours(about 15 hours per housing..) to do a meticulous job, as i do it all by hand so as not to dig too deep into the rather thin walls of the spark plug castings. but if tendinitis permits i could probably do a pair of housings for around $500(both leading and trailing spark plug passages). i know other shops do it for less but they also don't put up pictures of how well they do the portwork or where the focus is made...

after the porting is finished and cleaned up i will treat the raw aluminum surface to help prevent electrolysis from eating away at the hot spots(around the spark plugs).

on another side note, mazda did attempt to cool the rotor housings more in this area on the S6(FD3s) rotor housings by cutting deeper towards the rotor around the spark plugs but they did not bother to recast the housings for more improvement beyond that. S4 and S5 housings had 0 improvements in this area that could be benefitted from.

these housings will be used in an S5 FC that will be making ~500-550WHP, it made 450WHP on the dyno with the old turbo @17psi and the engine lasted approximately 30k miles until it died due to injector fouling(i built the engine about 3 years ago and last tuned it about 1.5 years ago. test results showed the leading secondary injector output 15% low which resulted in a dead front rotor). it has always had temperature issues with the front mount IC.


anyways, i'm curious if anyone else has used this type of modification and have noticed beneficial or negative results. the key to progress is making sure you move forward, the more we can do to make more reliable power from the rotary engines the better to prove the others wrong.

It doesn't slow the water flow down at all... speeds it up quite a bit actually where it counts - the surface.

it may not slow it down......but the fact that it is a rougher area makes it less easy for the water to pass by (much like cooling fins on a radiator or IC)...which could possibly help cool the motor. I'm interested to see the outcome...

the important thing is that the surface area is increased, this allows the heat to be transfered more quickly, the same reason air cooled Harleys have fins on the cylinders, or like heat sinks on computer processors.

That's cool stuff Karack. I've been reading about Meth/Water injection as a way to enhance cooling also. I bet they could be used together as a double hitter.

The purpose is to increase turbulence, so that more of the water flowing through the water jacket will contact the walls.

Remember, the whole point is for the water to make contact with the surfaces. Fast, turbulent flow is the goal. You don't make radiators with three or four huge tubes, and for the same reason.

i figured that this car was a good test for this modification due to it's purpose. the only thing it has removed is the a/c, has full trim and more additional weight from upgrades and stuff in the bins(full trim weight is probably ~2800lbs dry) and is using ~500whp for sustained periods in canyon runs going uphill as well. since we are not going to the v-mount i wanted an additional safety buffer to aid in combustion cooling since it is going to be turned up to 20psi or so on pump fuel/AI.

sub

I never understood this mod.

The flow of the cooling medium is at 90* to those ridges. So I would think that when fluid flows over/through it, I would think the same thing would happen there as when water flows over a set of ridges on the beach - flow slows, water flow becomes turbulent and bubbles form.

When looking for performance in the intake tract, you replace accordion style intake piping for smooth piping. That makes sense to me. This does not but I'm not expert.

"Bubbles" can mean two things in a fluid: 1) air or other gas suspended in the fluid and 2) fluid is momentarily transformed into a vapor, aka cavitation.

There should be very little air circulating in the system. As well, cavitation requires areas of very low pressure. So cavitation tends to form where there is extremely high shear in the fluid and this only tends to happen around water pumps. Cavitation makes a sound and is erosive to the surrounding parts, so if this is going on one could see the wear after some time of operation.

The turbulence in and of itself helps cooling flow. Since the coolant at the walls of the passage are hottest, turbulence moves the hot coolant away from the hot wall and keeps the coolant in contact with the wall cooler on average. That means more cooling. Similarly, more wall in contact helps with the heat transfer too.

The downside to turbulence and increased surface area is drag. Depending on the speed and type of pump used, extra drag may result in slower coolant flow which means hotter coolant coming off the block and less cooling for components near the end of the coolant flow circuit.

Like all heat exchanger applications, there are many aspects that are a tradeoff. You almost need to build, instrument, and test to see if they will work. Like most aftermarket modifications this is probably not practical. If you're just adding turbulence in a few critical areas I suspect the net effect is positive.

David

Only a slightly related note: I wonder if it's possible to modify the cooling passages so that each rotor housing is cooled in parallel, rather than in series. You'd have to add ports and stuff but that might help solve the problem of excess rear rotor heat.

the only way i can see to cool in paralell would be to add block offs and external ports to the rotor housings but i think that is going overboard in thinking that the heat exchange is going to make such an aggressive difference which i don't think it will. the tension bolts just mess with the coolant flow paths to make it nearly impossible to do internally.

i'm sure it will work a bit better but we have seen plenty of front rotors blown to know that it isn't simply a rear rotor cooling issue(of course i would lean on the rear rotor detonating before the front if AFRs and timing were perfectly equal though). the most ideal thing would be to have the freshest coolant going across the combustion side of the engine, which is already true anyways.

I haven't had a chance to look at irons since I thought of this, but aren't there coolant freeze plugs on the front iron that could be replaced with a nipple that would lead to the leading plug coolant passages?

My idea is to take a Bosch auxiliary water pump off of a VW (I'm pretty sure I still have one around here) and take coolant from the lower radiator hose and push it through. This would bypass the thermostat with a small amount of cold water, and increase the flow rate through passage.

Have any idea of the flow rate on that pump/

there is actually 2 ports on the front iron, the front freeze plug leads to the trailing coolant passage and the lower block plug leads to the passage just below the leading plug passages. you could block the trailing plug passage inside the iron to keep the water pump coolant from disrupting the add-on pump flow, i really only see this working for the trailing plug which in theory should be effective as it is the hottest part of the motor aside from the exhaust port. the exhaust port cooling isn't as important as the combustion side cooling IMO. it's been documented that the trailing(top side) of the combustion chamber around the trailing plugs is the hottest area and most susceptible to where detonation occurs.

i thought this was something racingbeat does also?

they do, as well as another shop that i can't recall the name of. racing beat doesn't do it quite as aggressively as i did but they do it more towards the intake side of the housing versus just the combustion side.

You may find this article about the dimples in golf balls useful.

I'm not an expert either, but I think it applies here.

http://www.aerospaceweb.org/question...cs/q0215.shtml

I can't find any exact info on the VW pump, but similar Bosch pumps are used on Cobalt and Lightning A/W intercoolers and from what I see on those, it is around 2-5 gpm.

Karack, I had forgotten about the fact it is the trailing side that runs hotter. There would be no need to block off the flow from the main water pump, and in fact, I believe that would be a poor choice, since if the aux pump failed, then there'd be no flow through that region. Instead, you would allow the aux pump to assist in accelerating the flow of coolant through the crucial area. The coolant is also the coldest you could get if you route it from the outlet of the radiator.

Of course, then there is another pressing issue. The main drawback to a centrifugal pump such as the Bosch is that it is awesome at creating significant amounts of flow, but it lacks the ability to flow when it must push/pull against a resistance. The resistance in this case would be the thermostat. Therefore, you would need to bypass the thermostat with a sufficient volume so as not to restrict the auxiliary pump flow.

On an car that is going to be used without a heater core such as mine, this is quite simple. Use the heater core nipple on the rear iron (conveniently located at the coolant passage we need to cool), and route a hose to the water neck.

In this picture, blue is the freeze plug where I propose to pump cold water in, and red is the outlet for this auxiliary circuit. It would the be routed into a nipple on the upper water neck.
Attachment 722867

Any critique or errors I overlooked?

^^ You might want to search for a really long thread in which Barry Bordes partipated on a similar topic. There was a lot of good info in that one about both this weave porting stuff as well as hot spots in the trailing plug area and coolant re-routes.

Why not tap the water pump, weld a bung between the trailing plug and run hose. Kinda like what that 26b has.

gracer7-rx7- Yes theat thread NEEDS to be brought up at this moment. thats were ^^^^ idea came from. I used to have the thread in my fav. on my old laptop but it got broke. i been searching for it with no luck. Has SO much good information and will answer tons of questions. So many good ideas, cooling mods and such

Found it!! So full of OMGoodness information

https://www.rx7club.com/rotary-car-performance-77/why-apex-seals-fail-866513/page9/

Racing Beat charges $100 per housing for this service, which leads me to believe they have some sort of CNC program set up to crank these out. I notice they specify the housings shouldbe used for race engines, which leads me to wonder if they believe there might be some long-term issues with removing the material.

Hmm interesting...Well guess theres only 1 way to find out lol.

i can only assume they noticed some failures due to the raw aluminum not having been treated prior to assembly. the raw aluminum is more susceptible to the chemical reaction of gases being created during the cooling system boiling off on hot spots in the engine or minerals from hard water becoming statically charged and discharging on the surface with least resistance.

i have seen it in plenty of cases where there is heavy electrolysis around the spark plug bosses and exhaust port casting, the areas not affected still had the oxidized protective layer where the pitted areas looked more like the base metal. i will harden the surfaces before i plan on putting the engine back together to fix this. i however don't have plans to bypass the cooling system and add an external pump for these passages like some others have done at this time so i will only be testing the porting aspect of this modification for now. but the idea is better for getting fresh water to those areas.

my first rotary engine build in my personal car(first rotary rebuild ever for me) had the worst case of electrolysis i have ever seen to date and i dealt extensively in figuring out the cause. the water jackets were completely compromised in some areas and i still have the housings that i use for painting fixtures for the irons.

I can go with that theory, adding only that sacrificial anodes are very likely to be needed even on protected coolant jackets and using a good coolant mix.

the best way to fight it that i have noticed is to just make sure the coolant is always fresh and mixed with distilled water, changed yearly, with good engine grounding.

my car also had a brass radiator, one less sacrificial anode and one reason why mine was so abnormal compared to all the rest of the engines i have torn down.

i'm still finishing up the porting and then i will treat the surface in a few days and get pictures after i smooth the surfaces and harden them.

I had something similar done in a motor of mine and ended up with a cracked spark plug hole, and it wasn't a high HP application. Dunno if they were directly related, I didn't take too good a look at the busted housing, but it's possible that this sort of mod can cause stress raisers in the material that allow cracks to begin.

i suppose that it may still not be enough to keep the plug holes from cracking.

how many miles did you put on the engine before pulling it apart and were they used housings or brand new? the cracking is perfectly normal but something that should be avoided if possible by any means.

isnt there anything that can safely coat the water jackets to prevent any build up? like how they have paint for the v-8's to help with the oil...ill check my magazine to get the name in a few

the best heat transfer is a raw surface, adding any coating will just reduce efficiency, aluminum has a hardened protective surface when oxidized anyways. an example of this is anodized aluminums, but the cooling system can do this on it's own however it can also continue to anodize the surface indefinitely(like submerging an aluminum part into a battery and applying voltage it will eventually disintegrate down to the base metals leaving basically an aluminum goo like jello and copper particles), as referred to in the "sacrificial anode" discussion earlier. adding a buffer may be better but also counteracts the benefits made from the modification. oxidizing the surface to start with is much better than leaving it raw and starting the anodizing process on it's own, leaving the majority of the cooling system heat transfer properties intact.

the OEM cast rotor housings already have hardened oxidized surfaces in the passages, which mazda left alone i assume because it didn't greatly negatively impact the cooling system heat transfer properties(rather it protects them), if there was no drawback i assume they could have cleaned the passages prior to the final milling stages which would have taken minimal work, the external housing surfaces have a much more raw aluminum finish which leads me to believe that they oxidized the internal passages on purpose for the reasons i mentioned.

The thermal conductivity of oxidized aluminum is nearly an order of magnitude less than that of plain aluminum. If the coolant passages are, in fact, oxidized, any effort to remove the oxidized surface would be hugely beneficial for the cooling of the housing.

perhaps that is why racing beat leaves them raw and submits them only for racing applications but my purpose is to get the most out of the modification for street driving purposes which is what most cars now are built for. also depends how much you oxidize the surface, the more black/greyed it becomes the less of the aluminum is on the surface leaving the copper behind which conducts less heat transfer than the aluminum. the stock housings aren't oxidized much and hold up for long periods of time.

perhaps later on a non customer vehicle i can play with protective coatings.

Used housings on a rebuild. Maybe 15k miles I forget, but a lot of them were on the track. Only one housing cracked.

Do you know what I'm talking about when I say a stress riser? I don't know how much engineering background you have... :)

Do you know what I'm talking about when I say a stress riser? I don't know how much engineering background you have...

.. stress riser? You are talking about the cracking of a chrome surface on a stamped steel insert cast into an aluminum housing where the opposite side of the aluminum would be machined for this mod....

You want to look into the expansion rates of chrome, steel and aluminum to find the forces at work with the very common spark plug hole cracking.

it's common in more than just those cases, many automobile heads tend to crack between the intake and exhaust valves because the heat can't be evacuated fast enough so it heats the most, in most cast iron cases this is a bad point for expansion/contraction cracks to form. in bad enough cases they would crack beyond the hardened valve seats into the passages or into the coolant jackets in even piston engines which have much cooler combustion temps. the cracks always form at a 90* angle to the apex seals on the rotor housings, sometimes some chatter is noticed if the cracks are bad enough, which i assume is what you mean by "stress risers". i still haven't seen the cracks really affect the seals too much or cause premature failures but i still don't like them present.

I tried to read this whole thread, but only a couple mentioned any corrosion or electrolysis.

Also, the raw aluminum immediately builds an oxide that inhibits the heat exchange, but the increased surface area is a plus. I do all of mine, because it makes me feel that I'm going the extra mile. But then none of mine are streetable.

I believe that anything you can do to better the heat dissapation is a good deal. But there are allso a load of placebos out there in the Rotary world.

EDIT: Calculon mentioned the oxidation item. If you could coat the AL with something, it would help to reduce the oxidation.

GD

What I mean by stress risers is imperfections (such as machining marks) in the surface of a material, which can lead to cracks. Stress is at its greatest concentration at the surface of a part, so any scratch, sharp angle, etc is going to be stressed and can lead to the formation of a crack. Once the crack starts the amount of material at the spot is reduced, and the next time it is stressed it can crack more, etc...

Exactly, but have you ever seen a housing crack on the backside of the spark plug hole in the aluminum?

never myself, just the chrome face. i have seen a number of housings that electrolysis-ized through into the spark plug hole and exhaust port jacket. aluminum doesn't have a tendency to crack unless bent more than a 90* angle regardless of heat/contraction exchanges. cast irons/steel/chrome will crack when heated and cooled too rapidly in succession at weak points(imperfections in the surface, think of pouring hot water on a windshield with a tiny rock chip and think of the plug holes as that rock chip) aluminum will just laugh at you or melt/shatter if it gets hot/cold enough(cold usually not being an issue..).

i can think that the porting helps keep the surface cooler but i can also think that it also cools the surface more rapidly once the combustion heat is removed(engine off with some coolant still cycling through the system slowly). it may still effectively cool the surface too rapidly and not be beneficial to help from keeping the plug holes from cracking. but it should still be beneficial for keeping internal combustion temps lower and reduce the chance of hot spots inside the combustion chamber. basically plug hole cracking seems to be inevitable based on theory and feedback, there may be a cure for that issue but the question is, is it necessary? i haven't seen much ill effect from plug hole cracking so far even in high horsepower applications, although i don't like the thought of the exposed cracked edges being possible hot spots in the chamber.

Karack,
Consider increasing the flow on #5 below with a little porting. Mimic the factory racing housing water passeges.
Barry

Attachment 722870

How did your rerouted coolant experiment work?

Playing with 10v Audi coolant bleed modifications, I've been thinking of rerouting some coolant from the water pump outlet (engine inlet) directly to the spark plug area.

i did notice that the lower spark plug passage was a bit restrictive.

mazda has been doing that for years in some of their race cars. they inject coolant smack bang right into that area.

also, look into how the rx8 rotor housings are modified in that area.

http://www.atkinsrotary.com/images/store/Rx8Housing.jpg

vs this

http://www.atkinsrotary.com/images/store/Drummond.jpg

interesting ideas concerning more modifications at internal cooling, keep em coming!

too bad i've been slacking off at getting these rotor housings all finished in a timely manner but it will be soon.

The way I machine the water jacket housings is fast, simple and cheap.

I clamp a piece of plywood to my drill press to move the housing around on and use a dremel bit (with a spacer floating on the shaft to control cut depth) as a mill bit.

I change the depth control nuts on the drill press around so that it is locked at a certain depth, make the cut, then set the depth deeper for the next cut (which is easy to gauge using the depth rule).

https://www.rx7club.com/attachment.p...5&d=1134124294

it's crazy that i came across this now because i was working on a set of rotor housings last night and thought about starting a similar thread both to see if i could get a discussion started and to see if people with better experiences would chime in on techniques.

this is my second time doing it on an engine, however, i never got to test it. my first set of housings are still sitting in my garage right now and i have no idea exactly when i'll get to use them. the housings i'm working on right now should be in an assembled engine within the next 3 to 4 weeks.

not being at home to work in my own garage right makes me think my first job was done better than my current attempt, but i did my best with what i had to work with.

see? this is one of the things i was curious about. 30 hours per set? taking into account that you probably have a deeper technical understanding (metallurgy, physcial chemistry, etc.), i'm tempted to think that i'm doing something wrong. all in all, it took me about 5 hours to do a set of housings - tops! also, that number included a few breaks here and there.

as one of the handful of guys aound here that i respect (and admire) their work, i'm just curious ... how fast is fast?

Blue TII, Can you post a pic of your gigarig setup with a iron on the press. Me very curious:nod:

i'm just curious ... how fast is fast?

I did my first set of housings in an afternoon. Probably same as you- about 5 hours.

My second set went a little faster mainly because I wasn't so paranoid something was going to go wrong.

Blue TII, Can you post a pic of your gigarig setup with a iron on the press. Me very curious

Its just as simple (rigged?) as you are thinking. Drillpress, sheet of plywood, rotor housing, dremel bit... safety glasses!

i had thought about something similar but i figured it would be almost as difficult to get a bit in there without breaking something or stab myself in the eye with flying shrapnel. i may have to give it a try and see how it goes though with something similar to save my arms some cramping.

the motor has been back together for a few months and is just about finished with break in(almost 1k miles), seems to be doing a little better temp wise than before but i won't know until mid summer when it will reach it's peak temp wise and on the dyno pushing 20+psi. i did all the faces inside the ports however, making it slightly more difficult to maneuver inside those ports, i figure it will help the effect if it is uniforum throughout but it also turned into much more work than i bargained for and was cursing by the end of it.