EMPI Head casting flash removal

[Outlaw6]

I put on the new EMPI heads from Sonex last year and am struggling with high CHTs. It's a turbo Aerovee with top mount oil cooler so I recently taped off the lower cowl opening to see if that would help by increasing the ratio of exit/inlet opening size. It seemed to help the CHTs some but not significant. But it did bring my oil temps up about 50 deg so I'm close to 200 deg now (88 deg OAT) which should be better for the air/oil separator.

So, as I'm scouring around for answers I find info about removing the casting flash in the heads. I knew about casting flash in theory when I got the heads but if you don't know what you're looking for and the flash "fills" the hole and "looks like" it should be there I kinda hesitate to just drill through and hope it's flash.

At the bottom of the first page of this thread you can clearly see a difference in the air passages after removing casting flash.
http://www.sonexbuilders.net/viewtopic.php?f=3&t=2804&hilit=casting+flash

What I've found so far is that there are a buch of different casting designs made by several different companies and manufactured in multiple countries, so basically enough to know that I don't know enough about it.

Does anyone have a picture of what casting flash can/should be removed on the Sonex EMPI heads?

[sonex1566]

Hi there,
I've got the Mofoco heads on mine, the casting around the valve guides caused me a different sort of issues. However the cooling issues I went in another direction. The coldest temperature that I have ever flown in is 15 degrees centigrade and it was bloody freezing! We don't do cold where I live generally. Because of some of the horror stories I'd read plus some overheating issues I'd had with a LSA Jabiru I enlarged the recommended size of the outlet in my lower cowling by 50% than what is showing on the plans. As the hole is also taken up with exhaust pipes and the lower part of the engine mount I figured the worst that could happen is it might go a little slower. Initially I didn't have any fresh air going into the lower cowling at all just relying on the top mounted oil cooler, however after 10 mins or so the oil temps were hitting 240 degrees F and I started sweating even though outside it was only 60 dgrees F. I cut a cooling hole in front of the oil pump at first. A 2" square hole lowered my temps by 25 odd degrees instantly. The important thing here is that my CHT'S are pretty stable around the 240 - 280 mark. So after 60 hours of flight time my oil temps are around the 170 - 180 all the time and my cruise is about 118kts true at 3200 ish rpm. So my vote after all that drivel is to make the outlet bigger from the get go. I think I've sacrficed a couple of knots but I save on buying fresh underwear.

[Outlaw6]

Well my thought was if there is casting flash blocking the cooling paths I'll just remove it, then open the cowl exit.
Do you lean at all? Fuel flow?

[GraemeSmith]

With the AeroVee that came on my plane (bottom oil cooler) I was having to step climb a lot to keep CHT's in norms.

I switched to the EMPI heads from what ever original horrors were supplied to the builder (he didn't like them either!). I did almost no cleaning up of the cooling fins which were well cast with well matched molds so the fins came out smooth and even with no discontinuity where the molds joined. There were a couple of small squares in the fin structure at ether end of the heads (like 5mm x 5mm) that were flash filled and I drilled and filed them clear. I ported and polished the heads to get rid of any sand and slag contamination inside the heads and smooth edges.

After a careful read of the cowl instructions - I enlarged the bottom outlet by nearly 50% over what was built to get to the recommended number of square inches of opening - taking account of the blocking by exhaust pipes.

I also cut the opening for the oil cooler air inlet larger - to the full width that is implied by the baffle setup. That made a significant difference to keeping oil temps under control. However in the winter the opening is too large and getting the oil to come up to temp is difficult. I made a couple of sliding baffles that can be fitted to the inlet and adjusted for cold days (variation on the Cessna oil cooler shroud for winter ops), It also seems to have helped the CHT's - I guess it is allowing more airflow through the whole system and improving scavenging on the low pressure side.

Hottest day I've now flown this setup was 31C OAT on the ground. I can now confidently shortfield at Vx without worrying about CHT's. Though to keep from overdoing it do clear the obstacle asap and then nose over for Vy. And as long as I keep at least 70KIAS CHT problems are gone. No more step climbing.

Combination of everything seems to have got things under control.

[achesos]

After a careful read of the cowl instructions - I enlarged the bottom outlet by nearly 50%

Hi Graeme,

After some frustration and research, I need to revive this post.

What is your cylinder baffle configuration? My heads are outside the Sonex "baffle box" setup, so I'm hesitant to start cutting into the cowl since it doesn't seem like they are getting any forced cooling anyway.
I wish there was an easy way to figure the statistical % of Model A Sonex with AeroVee engines that modified (increased) the outlet cross-section?
I'm chasing a solution to only balance CHTs (right rear is hottest of course) but decrease them all.

Sean

[Bryan Cotton]

Hi Graeme,

After some frustration and research, I need to revive this post.

What is your cylinder baffle configuration? My heads are outside the Sonex "baffle box" setup, so I'm hesitant to start cutting into the cowl since it doesn't seem like they are getting any forced cooling anyway.
I wish there was an easy way to figure the statistical % of Model A Sonex with AeroVee engines that modified (increased) the outlet cross-section?
I'm chasing a solution to only balance CHTs (right rear is hottest of course) but decrease them all.

Sean

Graeme had his Sonex for sale and hasn't been on the forum since July. Maybe a PM would get his attention, or there may be contact info on his website.
http://graemejwsmith.com/

[achesos]

Thanks Bryan - I guess I only read the lines sometimes and not between them - in one eye and out the other? Or I simply did read it and totally forgot.

So even if Graeme has left the fold, maybe someone else would provide some feedback regarding any adjustments made to cowl opening cross-sections for improved cooling? Seems pretty drastic to start chopping fiberglass if there are more obvious solutions available (to some others - apparently not me).
My baffle setup cools the cylinders - not the heads. Pardon the pun, but 'baffling' how the increase of air flow through the cowl might help the CHTs. I read about machining out some flash in the heads to improve air flow, but with my baffles I suspect very limited air flow through the cowl around the heads anyway?
I'm leaning towards expanding the baffles to extend out and include the heads in the internal airflow after doing some more fin machining in the exhaust port areas... maybe there is sufficient airflow in the cowl now? (fingers crossed)

"Experimental" for this type of reason I guess.

[Bryan Cotton]

When you say heads, do you mean valve covers? My heads are inside the baffle and the valve covers are out.

[Bryan Cotton]

So on my first set of heads I was not too diligent about removing all the flashing. Also the cooling fins around the outside did not line up well. Same thing this time, but I've got Religion about ensuring airflow past the exhaust side of the head. Here is one side after some rework, mainly going after the tight spot and filing down to the bottom of the cooling fins:

Doing this with a needle file really sucks. So I made a sanding tool in the spirit of David's oops fitting tool.

It's going a lot quicker with the tool than with a needle file.

[Bryan Cotton]

Here is one of my old heads. The quality of the castings was worse on the old heads, though the new ones still need work. View is looking in the direction of airflow.

Bad cooling fin alignment.png (2.61 MiB) Viewed 739 times

I really should have done this on the old heads. It is a pain in the ...