bahurd

iTrader: (8)

Years ago I did some work on heat shields for some furnace units. We found a double layer with an air space to create a convection flow was effective. <br /><br />I had an MP62 setup so know it doesn't have a lot of room over the header but you might think of that approach. <br /><br />Initial layer + 1/4" air space + top layer. Use standoffs to separate the layers. <br /><br />Keep the bottom and top open to get the convective flow going. <br /><br />Just an idea.

huesmann

iTrader: (1)

Why not a double layer heat shield: two layers of your stainless, with a sheet of silicone sandwiched between?

y8s

iTrader: (8)

That's a triple layer.

And what Bahurd said is: convection

plus I think silicone would get unhappy that close to exhaust heat.

anyway, let the hotness in the middle "layer" blow away instead of being stuck there and staying hot (like silicone, fiberglass, unobtainumonite, etc.)

patsmx5

iTrader: (16)

Good to know. I was hopeful when I bought it the "unfinished" side would be decent and it is.




Regarding doing a double layer with an air gap- after making this shield, I don't think I have the patience or skill to pull that off. I could do it on the top and side that are mostly flat, but the curved spots would be.... A huge pain.

huesmann- the metal gets a bit over 300*F with a raised idle and retarded timing to increase EGTs. I would guess at full load, it would get a lot hotter. Not sure if silicone would hold up to that heat. Fiberglass or an air gap would probably be better.

But actually, I am going to do a double layer with an air gap. Only how Dustin mentioned doing it. One layer on header (as I have) and one layer on the SC base (today's job). That will put an air gap between them that varies between about 1/4" to 5/8".

patsmx5

iTrader: (16)

That's what I'm hoping happens. Header shield and another layer on SC base, and let airflow through the engine bay help to carry away the heat between the layers.

bahurd

iTrader: (8)

Don't over think it. Why does the curved part need to be curved? Mock it up in cardboard or something and bring it to a tin shop. Top and bottom can be identical and use a oversized nut as the standoffs. Then bolt through. <br /><br />Are you using factory header?

patsmx5

iTrader: (16)

It's an 01 stock header, modified a bit to improve flow. I already build a shield, so not taking it anywhere. It was curved in a few spots because the header curves in those areas, and packaging between the primaries for cylinders 1&2 is really tight relative to the SC.

Stealth97

iTrader: (5)

Watching this closely. Definitely putting this on the to-do list for underhood temp reduction. Impressive and detailed test results.

patsmx5

iTrader: (16)

Thanks! I'm making good progress. See below!

Ok so as I mentioned I put the reflective stuff on the SC base and tested that today. It works fantastic. In fact it seems now that the majority of the heat is actually coming from the 2 mounting tabs that connect to the header studs, not from radiation. This is a good problem to have, no radiant heat is practically not an issue anymore, and it very much was before.

Pics and temp data below. The temp data, I measured the temp in multiple spots at 30min, 45min, and 60min. Then I used a blower I plan to install to cool the SC (active cooling of course) to blow on the SC base, and the temps at the hottest spot on the base (right next to the mounting tabs dropped down to 118*F in 5min! 92*F during testing.







Little insulators I made for the header studs


Base installed for testing


Pic of reflective surfaces with air gap!


Fan mentioned


Temp data:


Overall I'm happy with it. The temps are down a lot, and with the fan they fall close to ambient fast.

huesmann

iTrader: (1)

I would think the way to go would be to get 2 sheets of your sheet metal and sandwich something pretty solid between it, maybe a thicker sheet of copper, then beat them to shape around a wood (or metal) form (I know this metal forming technique has a name but it escapes me). Then remove the middle sheet and bolt/rivet the two pieces of metal "bread" together with spacers to keep them apart.

patsmx5

iTrader: (16)

Thanks, I hadn't thought about using another layer as a spacer that gets removed to reveil an air gap. That's not a bad idea! This is why I like having threads like this, it helps me learn and get new ideas for better ways to do things.

However I believe my heat shield is finished. The testing I have done shows that heat into the SC base via radiation is practically nothing. At this point, conduction via the 2 mounting tabs to the SC header studs are by far the dominant heat source, and even that took is a small amount of heat. I finished my SC fan setup yesterday. It blows air directly on the gear case of the SC to help keep the oil temps from getting too high, but it also directs about 10-20% of the airflow to the rear by the SC mounting tabs. Of course will test all of this with and without fan soon enough.

EDIT: Fan setup

huesmann

iTrader: (1)

Indeed. Now you've got me thinking. Does an air gap need to be completely sealed for it to be useful?

JasonC SBB

You effectively have a large air gap if you have a heat shield near the heat source (header), and another near the s/c.

JasonC SBB

This fan setup strikes me as almost worthless. There's very little surface area for the s/c to shed heat.

I presume your setup is intercooled?

y8s

iTrader: (8)

In a situation where you've got flowing air (even slowly) around the heat shields, it would make it worse to seal in that air. It has mass and will hold heat. If you leave it open, you'll get lazy convection pulling even more heat out of the space between the shields. This is a good thing.

patsmx5

iTrader: (16)

Nope, y8s answered it but a little air flow between the layers will actually help since it will cool off the shields, dropping their surface temps and thus how much heat they can now transfer

Fan setup is to be tested. It's purpose is to help cool the oil in the SC (just the oil case on the front). The SC lost an oil seal after swapping pulleys to spin it to 20,000. It was fine at 17,000. Oil was very nasty/dark with only 1,500 miles on it and Whipple said it got too hot. Whipple doesn't offer an oil cooling solution for their SC's. Thus the fan, it can only help. Yes it's intercooled.

Agreed.

Dustin1824

iTrader: (2)

I know nothing about superchargers, but I'm sure it's possible to DIY a oil cooler.

Could you use a heat exchanger and electric pump? Maybe use the fill and drain plugs as the return and feed? Is there anywhere that would be a decent place to drill and tap?

Of course, that's going to be pretty expensive, but it would work. What's the oil capacity on the whipple?

patsmx5

iTrader: (16)

There is a drain and fill plug, though they are tiny. In theory they could be used though.

Biggest problem with that idea is cost. I need a pump and heat exchanger, fittings, hoses. The SC oil capacity is 5.5oz. This oil is 26 dollars for a 4oz bottle! Adding even a tiny heat exchanger pump could add a few hundred in oil capacity to the system. You would need to run tiny diameter lines to keep the oil used down.

It may come to that though. For now, I'm going to test a simple (and cheap) fan and see what that does. My thoughts now are that the SC was being partially heated by the header. Now that is very well shielded, and I'll put a 20mph breeze of ambient air on the oil case and test. Hopefully this gets the oil case to an acceptable temperature. If not, then a pump/hoses/heat exchanger setup will probably happen. If the fan setup works, it's lighter and cheaper, and less parts to leak or fail.

Dustin1824

iTrader: (2)

I was wondering why you decided to run that fan, now I see why.

Remind me to never buy a Whipple lol. Seems like you are trying to solve an issue that needs to be a design change to the supercharger, they didn't leave you with many options and it's strange that they don't really have a common fix.

That's just a tiny amount of oil for something that's consuming a lot of power...why are you running a whipple again? Car looks very fast from the video on youtube, but I think your power goals may be beyond this thing's capacity.

Does your car have a vented hood? Maybe a hood vent directly over the supercharger would help?

Is there any way to increase oil capacity, such as cutting open the case and welding on a larger reservoir? Doubling the capacity would probably help a bunch. Once again, I have no idea how this supercharger unit is configured inside, this may not be possible.

Does anyone know of an equivalent oil to this $$$$ oil?

patsmx5

iTrader: (16)

Well I talked to whipple, and they have a solution. They told me "buy a spare, and when it breaks install the spare and ship back the broken one to be rebuilt. That's what our customers who race do." Not my idea of a "solution". In their defense they did say to move the SC to the cold side of the engine to help with heat which would help some, but I think the shielding I made will prove adequate on that point.

Kenne Bell does offer a cooler for the oil case on their blowers as an option for high boost applications. Whipple doesn't. Kenne bell says their cooler helps keep the oil happy for high RPM/high boost setups.

Basically 99.9% of whipple's customers are not pushing their SC to 28 PSI and trying to daily drive it. I've proved the SC can make the boost, now I just got to make it stay cool enough to not destroy itself.

I'm running the whipple because I wanted a 350whp miata with no lag but the capability to make more power down the road (either more boost or a compound setup). This blower should be very capable of moving that much air and actually a lot more. My goal is 350whp, but I'm betting I could hit 400whp now that I've got E85. First it was belt slip, got that fixed, now it's keeping the SC cool enough to not destroy itself. Literally every SC failure I've been able to find online was the result of something in the SC getting too hot. Rotors too hot and touch, case too hot and rotors hit case, bearings too hot and fail, oil seal too hot and fail. There's a pattern to these failures...

Not running a vented hood right now. Trying to keep the stock hood/stock body but if everything fails, a hood vent may happen.

Modding the case wouldn't package very well on my setup since the P/S system is right below it, belt right in front. Motor to the passengers side 1/4" away. Only way you could extend it is to the drivers side. Would be easier to add an external oil cooler as you mentioned than to do this.

I did some basic heat calcs on the gear case and the fan should be able to dump 1-2hp worth of heat from the gear case with ease based on 100*F day, 200*F oil, and I think 200 CFM of airflow and I don't remember what effectiveness. It should be plenty on paper, but I will test it and verify. I can find the math if anyone wants a look!