I think what he means is that the swirl is needed to keep the flow .... flowing. That a smooth surface actually impedes the flow, which is why you aren't supposed to polish the intake side of the head. In fact, in the intake ports you aren't supposed to use anything finer than 60 or 80 grit sandpaper. So, apparently that includes the backs of the valves.

My problem is that I'm using a Dremel that takes 1/8" shanked tools but the Roloc disks I've seen are 1/4". That would require me to use my die grinder. Maybe I can get it to fit in my new tool post grinder adapter on the lathe. Or, maybe I can find a rubber disk to back a sandpaper disk with and put it in the Dremel.

 

Gary,
Any surface roughness that is circumferential (like I see) will cause a little ridge where the air molecules have to 'jump off' on a microscopic scale.
With swirl polishing they tend to follow along the ridge on the back surface of the valve.

Travers Tool has 1/8" shank flap wheels good to Dremel speeds.
http://travers.com

 

Subsequent to posting the pictures I tried actually polishing the valves, both back and front. That worked pretty well and took out much of the circumferential grooves. Would that be satisfactory or is the swirl still preferable to smooth? I'm not sure I fully understand the smooth vs textured issue.

But, as I think about it, those grooves are caused by the Dremel being "on center" with the lathe chuck, meaning that the grinding drum is going straight down at 90 degrees to the stem of the valve. I wonder what pattern I'd get if I moved the Dremel up or down with respect to the stem awa the lathe chuck? (Luckily that's easily done since I built the holder for the Dremel to fit in the quick-change tool post of the lathe, which has a provision for height.)

 

Moving the dremel off center would give the swirl.

Flow is never going to be laminar except on a flowbench where the valve is held open to set lifts to see how much airflow there is, but in reality, the airflow starts and stops every time the valve opens and closes.

As such, we need to take the most advantage of flow at low lifts, IOW where the the head of the valve is just coming off the seat and when it is coming back closed. With a smooth surface on the back of the valve, a smooth(no sharp edges/turns) port, and unshrouded valve head, we can get better flow. This low lift flow is what we want for torque. It helps get the new charge into the cylinder and exhaust the old. Some of this is overlap caused by the cam and is a whole nother subject.....

One more note. The more angles you can get in the valve job, the better. A 5 angle flows better than a standard 3 angle due to reducing the angle of each step. And again, the difference is at low lift, which is most advantageous.

BTW, I have heard of some Nascar testing to make the sonic pulses of the intake ports help flow. I don't remember it all but something about sizing the port volume, timing the port velocity to the valve opening to create more flow.

 

Stang - Thanks. Didn't get to do any more work on the valves today. Responded to a craigslist plea for help with bolts, brackets, etc for a 302. Responded and found a family that sold their '81 F100 Flareside years ago and just bought an '82 Flareside with a disassembled front of the engine. No bolts for the timing cover nor water pump, no fan nor clutch, etc. They came over and spent time finding what they needed for the first pass, with many more to come. Bet we'll get some new members very soon.

Also responded to another CL post about a 9" rear that was supposed to be limited slip. Got the price down to $50 but then we both realized it wasn't LS. But, in the interim I discovered this guy has a Corvette restoration business and has 3 of them in the Hall Of Fame. Anyway, we talked heads and valves and he thinks the one I polished is ready to go.

So, that leads me back to the question - is a swirl-polished/ground/sanded finish better than polished? I think I can do either, and if moving the Dremel off-center will do the swirl that is easier/faster. But, which is best?

 

Lots of racers take advantage of Helmholtz theory to tune the volume and length of both the intake and exhaust.
This works, as I said, for a particular cylinder volume and rpm.
F1 cars were the first to have computer controlled velocity stacks to lengthen and shorten the length of the intake tract and broaden the powerband of their extremely high strung engines.
Two strokes have expansion chamber exhausts to take advantage of these resonant pulses.
Much the same as the difference between long tube and shorty headers.

A quick Google found this page; Intake Runner and Peak Torque Calculator

And this one; Optimum Intake Runner Predictor Program

 

FWIW, Chrysler did a whole lot of work in the 60s on ram tuned intakes. The first was the "Sonaramic" intakes on the 413 engines in 60 or 61, followed by the "short ram" intake in 62. The first one had the carburetors on the outside of the engine, the second had the carburetors on opposite sides of a large intake manifold. The cross ram design was used on a number of different engines, not just Chrysler products. Even AMC had one, Fords one I can place was the "Cross Boss" running one or two Autolite inline 4 barrels.

The tuned length intake is widely used on EFI engines, now frequently with variable length runners.

 

Looks like raising the grinder above center did the trick. Here are a couple of before and after pics on intake valves:



Slightly closer:

One thing to note is that the original valve has a decided lip of about .015" just above the lapped area, but that is gone on after the grinding. I believe that lip would cause the incoming air/fuel mix to veer back toward the seat rather than go through the valve/seat opening in low-lift conditions if not in full lift.

In fact, in a way it is like getting another .015" of free lift, which is a bit over 3% of the actual lift. And, it isn't there redirecting the flow of the incoming charge in all conditions. However, since the valve is in a low-lift condition for quite a while as it opens and closes, this is probably most significant in that condition.

Thoughts? Am I "there" on the finish to put on the valves? And, do the exhausts need a swirl or a true polish? I can do either.

 

I'm guessing you guys are getting weary of this since there have been no responses to my last post. However, many of you've said you like it, so here's an update.

I have all the valves back-cut and swirl polished, both intake and exhaust. I then went back to the heads themselves and didn't like what I saw and am doing a lot more work on them. Looks to me like some very careful use of a sanding drum on the Dremel will smooth things much better than I'd previously left them w/o taking off much material.

Also, I've read that smoothing the combustion chambers will help as it will reduce the heat that is lost into the cooling system, and that will increase efficiency. However, I'll be removing some material to do so and that reduces the CR unless I have the heads cut more, so I'm wondering if it is worth it.

Meanwhile I've been using both DesktopDyno and Comp Cam's CamQuest to model what I should expect in HP and torque. As it turns out, they are pretty comparable in results, with CamQuest giving slightly better #'s, but those #'s are so close it is within the margin of error. And, that might be explainable since CamQuest has a intake manifold pick for an Edelbrock Performer while DD has but one 2-plane intake.

So, how good are the #'s? How about 275 horse @ 4500 and 390 ft-lbs @ 2000? Too good to be true? That's what I thought, so delved into all the behind-the-scenes parameters, and especially the flow #'s for the heads. Then I did a lot of searching and reading on the internet regarding my heads - D8OE's. The most kindly thing I found regarding them is "door stop", although a couple of guys ported them and posted before and after flow #'s. As it turns out, their before #'s are almost identical, so it looks like they didn't copy each other but did flow the same heads. However, comparing those #'s to DD and CQ #'s says neither program thought anybody would use D8OE's since their #'s for stock heads are significantly better. In fact, one guy's ported #'s are not as good as either of the two programs "pocket ported" numbers - the first step above stock.

So, am I chasing shadows? Wasting my time? I may find out numerically as my nephew just informed me that a friend of mine in my home town has a flow bench. I'll call him tomorrow to see if we can test these heads and find out.

 

Can't wait for the results!

 

Today I re-worked the valve pockets on one of the heads. I did more blending of pockets with the runners and tapered the valve guides. Here's the intake:


On the exhaust I cleaned up the valve guide and took more material off of it:


And, I gasket-matched the ports on that head. Here's the before:


And, here's the after:


The thing to note on the ports is that the runner narrows down about .200" on the right side just inside the head due to the holes for the head bolts. While the walls for the bolts appear to be .225" thick I wasn't comfortable taking more then .100" off, but that took quite a bit of the bend out of the runner. And, by raising the floor and ceiling to match the gasket it looks to me like there's as much area in the head at the bend as there is in the intake manifold.

So, tomorrow I'll be doing the same things to the other head. Once that's done I'll start polishing the combustion chambers - unless you think I shouldn't. For instance, if the additional efficiencies gained from that is less than what I'll lose in compression ration. Thoughts?

As for the friend w/the flow bench, he's not called back so that's iffy. I'll keep you posted.

 

Yes, please do! Even though I don't say much to this project; I'm more learning than anything....

 

Knock any sharp edges off in the combustion chambers.
Polish until you feel like it's done.
Have you CC'd them as they stand now?
How far in the hole are your pistons?

 

Chris - I'm learning as well. Was beginning to think my quest has been in vain, that I had started with the worst possible choice of heads like most say on the web. And, that was in spite of having found two guys saying they've ported these heads, although one of them quoted what seemed like impossible flow numbers for his heads. But, I just read about Thumper Performance's heads, which are ported E7TE's and are pretty much identical to the D8OE's I'm working on. He gets almost the same kind of flow #'s that the other guy quoted. Further, he confirmed what I'd read - that small valves are the ticket for low RPM torque, and these heads have the smallest Ford put on a 302/351W.

Jim - Haven't cc'd yet. And, haven't measured the piston height as they are still in the box, not the block. The plan has been to do all the dirty work, like grinding heads and polishing valves, then clean the shop prior to starting to assemble things. However, since the weather has been so nice I've been doing the porting work outside so there'd be less mess to clean up.

Oh, and then there's the fact I don't have a burette (?). Guess I'm going to have to find one. Got the plexiglass and plenty of ATF.

Having said all that, I'll take your silence on the head work to mean you don't see a problem or big chance for improvement. Hope I'm right.

 

Gary,

It's all looking good!
I don't know a thing about these particular heads. So I'm sure as heck not going to make anything but generalizations.

You have a lot of effort into these. I certainly hope the results are exactly what you hope for.