Mazda 16 valve head query.

[4v6]

Good morning Guy, i hope your vast experience may help with this particular cylinder head.

I have been tasked with improving a cylinder head from the mazda stable, as far as im aware its from one of the many 323 variants.
I recently worked on an almost identical cylinder head for a friend and made some respectable gains in both bare port flows and with valve in flows.

This cylinder head is almost identical to that previous one, however it differs in that the left port of the pair (intake) has a huge intrusion/deflector that pretty much covers half of that left side runner off.

My flowtests of the previous head showed that bare port flow of that head was 191cfm@28" wheras unsurprisingly, this one is somewhat lower at 180cfm@28".

A cotton trace applied to the port bore out what i was thinking was happening, with air being deflected around the "hump" and then creating turbulence ( lots) before it entered the valve throat.
The opposing port remains pretty stable at all points except at the splitter where it exhibits some disruption.

I then applied both traces to the port and blew air down it so i could observe the exit of the traces ( into the cylinder as it would be) to see what the traces were doing.
The obstructed side exhibited a rather violent clockwise and intermittant rotation and had a very disrupted flow, the opposite side also showed some rotation anti clockwise but appeared very much more stable.

My thoughts regarding this "hump" are to attempt to induce swirl/tumble maybe but its clear that flow suffers as a result.
My question therefore would be; How best to proceed on modifications?
Should i lessen the obstruction/remove it and gain better overall flow or am i likely to do more harm than good do you think?
Ive attached some photos for you to see for yourself the offending "hump" and any help you can offer as always would be gratefully appreciated, apologies for the poor quality of them, my digital cameras developed some kind of weird fault where it places lines across almost all photos.

Tony.

Obstuction/deflector and closing off of port.

S8000211.jpg (79.88 KiB) Viewed 7204 times

Looking backwards up the intake from the chamberside.

S8000212.jpg (97.21 KiB) Viewed 7204 times

Chamber. Carbon traces in port behind the hump at the throat.

S8000216.jpg (136.53 KiB) Viewed 7204 times

[Guy Croft]

Good burn or high flow?

Now that is a real conundrum!

GC

[4v6]

It is indeed!

I would hypothesize that since Mazda produced an almost identical head minus the "hump" (not sure whether it's a year related alteration) that it (the non hump head) has adequate burn characteristics, so assuming no breakthrough into water jackets happens ( it appears clear to proceed) maybe I can either eliminate it entirely or reduce its disruption via a modest profile change.
I think the latter will probably be the way to go initially, especially if I can approach the previous heads flow figures.

Maybe I can get the best of both worlds?

[Guy Croft]

OK - decision: If you can increase the CR substantially I would remove it entirely - if you can.

GC

[4v6]

Many thanks Guy, ill do some digging into what the current Cr is, although raising it could be an issue due to the owner wanting to utilise it in a turbo charged application, so maybe the burn aspect wont be quite so prevalent a consideration.

Once again, many thanks for your time on this.

Tony.

[Guy Croft]

Sorry, did not realise it was a turbo, low CR and take it off then!

G

[4v6]

Apologies for not responding sooner, my computer has been misbehaving itself.

Re: Mazda head- Sorry Guy, my fault for not clearly stating the intended usage.

Intake ports have now been modified by removal of the "hump", bare port flow rises from 180.1cfm@28" to 207cfm@28", so thats in line with what i noted on the turbo head. No breakthrough into coolant galleries and i had removed a core plug in the head just to confirm material thicknesses for peace of mind.
Interestingly, the flow with the valves in is better with the "humped" head as stock than the turbo head in stock condition, however after modification it loses out especially from 5mm lift onwards.
E/I ratios for the stock head are a very high 95% (172cfm exhaust vs 180cfm intake).
I will post some photos up when i get my PC install back up and running.

Thanks again, Tony.

[4v6]

Reason for better flow than the previous head with its valves in appears to be due to a larger pair of intake valves, 33mm vs the turbo heads 31mm.
A couple of photos to show the rather large contrast between stock intake port and modified.

Stock intake port featuring a "deflector".

S8000233 (816 x 612).jpg (61.95 KiB) Viewed 6991 times

The same port but with the deflector removed.

S8000237 (816 x 612).jpg (96.84 KiB) Viewed 6991 times

[4v6]

The flowtest i carried out shows a marked increase over the standard ports, tested at 28" depression setting on the flow tester.

Modified port shows rather marked improvement over stock items.

Mazda BP26 stock vs modified intake..jpg (83.54 KiB) Viewed 6840 times

[Guy Croft]

very real gains there, well done Tony,

G

[4v6]

Thank you Guy, I didn't expect the improvement to be quite so marked.
Have to say, I'm happy with it though!