Mazda MX3 for learning and Rally/Ice X seat time

Hello All! I've been lurking a while and thought it's time to introduce myself, name's Valentin, hail from Alberta in the Canada. I'm totally green to this rally thing you all do with your cars, but have been rather seduced by it all and the wealth of knowledge and good humour pooling here seems to be too good to pass on.

I'm a mechanically minded artist by backround, learning aircraft maintenance in the present. Have been in love with the idea of driving a shitbox car as fast as possible on loose surfaces for quite some time. Until now circumstance has not allowed such fun. Long story short, I've come to possess a 1994 Mazda MX3 (yes, just like that one in the buy/sell here) with an engine that has eaten itself due to oil transforming into tar. The engine is an odd 1.8L vee 6 K series thing, so in search of a cheap substitute I found a 2.5L vee KLG4 that bolts straight in to the chassis and the trans (with some minor electrical surgury of course). Weight is similar, but torque and power is obviously more handsome, also it was the only replacement I could find that wasn't a total wreck. The engine is similar to what said car in the buy/sell has under the bonnet, save for it being lower compression and some other things.

In any case I've been rebuilding the junk yard substitute in my tiny shed for the last three weeks and as usual I get a little carried away with modification... (I love the building aspect of things just as much as driving) especially the stuff that costs die grinder bits, sand paper and labour (I am a student after all). Of course the "real" purpose of the exercise is to verify the tolerances are within limits and to replace the wear parts. However I was hoping to get some advice on port shaping and combustion chamber shaping. I know I should probably just bolt it all together and go driving, but it's just so damn interesting to think about aerodynamic/fluid dynamics inside the head that I can't resist the opportunity to improve a little on the factory casting. I don't mean anything extreme in the sense of hogging out them ports to enormous size and reducing gas velocity, just to balance the ports to the size of the stock valves, work on the valve bowls a little and remove hot spots in the combustion chamber sort a thing. Of course I'll shave the head to compensate for lost compression and put her all back together. Aircraft mech shop helps in some tooling, but in in Aviation we don't modify anything at will otherwise people get unhappy if the planes don't come back to earth softly...

I've been reading and looking at examples of nearly everything I can get my hands on. There's a ton of porting jobs online that have very little bearing on fluid dynamics. There's an Aussie guy called Graham Bell (of whom I'm sure you know) that wrote some books on preparing 4 stroke engines. What's the consensus on what he preaches? From what i've read hes level headed and actually tests results on a flow bench. Also there is a mighty interesting post analysing V10 F1 head on a F1 tech forum HERE

Here is pictures of what I've done thus far. Basivally deshrouded the valves/ removed sharp edges and casting here and there. Valves are out now and I'm ready to work on the bowls and port. I have yet to measure the valve seat diameter, throats and ports.... accurately with some tee gauges. So no math yet.


Please school me.
Valentin



Edited 1 time(s). Last edit at 10/25/2016 10:44PM by tin.

Got excited and took some rough measurements of what I'm working with:

Intake valve seat dia: 28.5mm
Area of 2 ports at valves: 1275.9mm2
Intake throat dia: 28mm
Area of intake ports at manifold: 1128mm2

Exaust valve seat dia: 24.5mm
Area of 2 valves: 942.39mm2
Exhause throat dia : 24mm
Area of Exhaust ports at manifold: 984.10mm2

I'd like to increase flow at low to mid valve lift to boost torque some, my logic is that the valve spends most of the time in this area and the benefit would be felt. I know some grind the throat flush with the seat or close to it, others prefer to have a 90% size throat in proportion to the seat. Supposedly a lower diameter throat would induce a high velocity/low pressure zone to aid VE at lower gas velocity (low/mid valve lift), while a straight port gives better flow at a high gas velocity (mid/high valve lift). Am I understanding this right? If I want to improve VE in low to mid lift then should use this venturi thing that the manufacturer made. Bell says to smooth it for best result. I obviously need to measure more precisely what the port is like throughout, but would it be advantageous to accentuate the valve bowl to have a slightly more drastic venturi? Say by enlarging the bowl to a couple mm larger than the throat and blending towards the port flanges?

Also it looks like the intake port at the manifold has a smaller are .88 times the area of the valve seat diameter area. I have calculated valve lift area but would this pose a problem being smaller? would this reduce gas velocity as it enters the combustion chamber? Should I enlarge it a tad?



Edited 1 time(s). Last edit at 10/27/2016 11:47PM by tin.

This thread is useless without pictures, especially if you want someone to offer some advice on where to remove material. I haven't looked at one of those heads personally, so I can't really tell you where to spend your time.

I can tell you that you will find most of your low flow gains with the valve shape, valve seat profile, and work in the bowl area. But without pictures....

As far as valve seat ID does, you want to smooth this perfectly into the port. Some engines have really large bowls with small valve seats, this is probably the worst thing to have (Volvo 16v is notorious for this). If the bowl is smaller than the ID of the seat, hurray! I Would say that an ID Of 90% of (measured at the smaller of the two edges) is pretty conservative. Most high performance heads are in the 93-95% range. Take pictures...

To your last question, you need to take into account the valve circumference and lift to get a better idea of flow area. Measuring just the valve seat diameter doesn't really tell the whole story.
Also, an intake area of 12mm^2 is pretty small

Hi Robert, thanks for the reply. 12mm2 is indeed miniscule, hense why I need advice

Had some time today and I took some pictures of the ports, some more comprehensive measurements also. Even I have a much more clear understanding of the port design, let alone those who've never seen the thing...

It looks like the bowl on the intake is roughly 28mm throughout give or take 0.2mm. So that's pretty close to the seat size which is 28.5mm. The throat cut end is the narrowest point and dips to 27.9 or so. The dims are consistent in the mid section of the port aft the split/divider. The port is also pretty close to round.

The exhaust measures out to be 24.7mm in the seat, narrows to 24.13mm at the throat cut end, then expands to 24.9mm at the bowl. The port then takes on a square shape, which measures 22.5x25.2mm and 28.5 diagonally (this works out to roughly 550mm squared per port). The area of the seat is 478mm squared for comparison. I know I should go measure the lift and figure out the area of that, but this is all I've had time for now, as I'm writing midterm exams this week. The exhaust ports definitely have a more pronounced throat cut, especially in some areas. It looks like the bowls have been sanded some but its almost no material removed, just to get rid of carbon here.

Also (don't know the right name) the wall that divides and merges the two ports into one has a very blunt shape in the exhaust ports. It seems that this is done for strength, would a thinner taper help aerodynamics here? What are your thought's on knife edging these? I know from aerodynamics classes that blunt round radiused forms cut through flow smoothly and thinly tapered forms merge flow without turbulating or disrupting that boundary layer and laminar flow. Does this hold true to port design? It seems there are other design factors at play as well.

And... when you talk about work on the valve shape... do you mean things like a radiused valve face corner and back cutting on the other side? Unfortunately I don't have tooling to grind the seat profiles.

You may want to learn what the engine characteristics are before trying to change too much. Do you want to increase low end torque, mid range or peak power? What you do want to do with the car will help in determining what you want to do the engine. Have you used the car at an event and found it's shortfalls?

It may be best to learn to drive it well as a completely standard car before you make any "improvements". Doing work on the engine will not make anywhere close to an improvement in times as work on the driver.

A good driver can be quick in a slow car, a bad driver in a fast car will just crash quicker.

I know that isn't what you want to hear....

---

Yes, it's a Ranger.
Xr4Ti, it is rwd and was made in Germany.

Those ports are fantastic looking! The ONLY thing I would recommend doing is smoothing out the casting lines and the step just under the valve seat insert in the exhaust ports.

The factory did a nice job!

Hi Matt,

Oh no worries, I appreciate honesty. I agree with you 100%. I drove the car home from the seller and it's stood ever since. The current engine is a 1.8l 60* V6, and its very worn in the bottom end and cams. Oil changes were done maybe 15000km interval with "synthetic".... so I pulled an engine from a Mazda 626, a KLG4 2.5ll V6, because it was the only decent specimen I could find that bolts in. It was very dirty outside and so I'm doing a refresh of bearings, rings and seals to extend it's life. It looks like it had oil in the combustion and there's a lot of carbon in the combustion chamber, so rings valve stem seals and a properly working PCV system should do the trick. I'm not planning on spending more than a few weeks on the engine build, just want to iron out any wrinkles, so that I can daily drive it as well as go and learn on the lake one that freezes. Potentially enter the Ice X on ghost lake this year, if I can make it early enough and it happens.

The engine is actually quite nice out of the box. Supposedly it features a forged crankshaft, rods and pistons, the valves are shim on bucket with dedicated oil feed to the bucket's bore, oil jets under the pistons and a rigid split aluminum crank case, coil ignition and a variable plenum intake manifold. Should be fairly durable/zippy. Torque is pretty flat to 7000rpm redline. I'd like to balance the pistons, pins and rods, correct any pitfalls in ports and combustion chamber and skim the head a tad to balance out compression or maybe raise it a little 9.5:1. The heads have to be squared anyway.

What are the thoughts about raising compression. Obviously our Canuck 91 maybe 94 pump gas, but would it require a timing advance to make use of the extra squish?

Say Matt do you still need volunteers for the Powderface Rally? I applied on the interweb, but no reply yet. I'd be happy to help out. I've got a tow vehicle, but would be fine with whatever.

V.

I'll try to do the ports on both heads this weekend. What are your thought on the square exhaust ports and the casting where they merge? Would it be advantageous to slim that piece, its roughly 15mm and tapers down in a blunt fashion (I've drawn the top cross section).

Also here are a couple photos of the car if anyone is interested in what it is. Sex spec from the 90's, though unmodified and has its original 15" wheelies. Best part is that the gearbox has pretty short gears and it weights 1200kg and the struts are new (hope they'll last till spring), but has an open diff and a clutch pack diff is hard to find for these, should I just try it welded and stock up on axles? It has rotten sills that I will have to weld up and needs a full exhaust (the junk yard headers were cut with a shear)... I got a Chinese 6-2-1 manifold to replace the iron logs, all for 60 doll hairs, its a bit of a joke I know... but at least it has equal length primaries and secondaries, and the tube sizes actually arn't too big.

No, don't try to sharpen the divider on the exhaust side, likely doesn't help anything but makes a hot spot. The Mazda K series V6s are actually pretty good - Ford used a heavily reworked one destroked to 2L in British touring car in the late 90s. Not a heck of a lot out there in terms of racey parts though. If you don't have a flowbench or experience doing ports, I'd concentrate more on just deburring things and making sure there's a smooth transition between the seats and the port - nothing major.

Yeah that's what I thought too, probably end up cracking from the heat. There's so much misinformation on the interweb. Those F1 Heads have a blunt port merger thingie too



Here is the intake



I was very surprised that they use pneumatic valve train. Nevar herd o dat b fore...



Edited 1 time(s). Last edit at 10/28/2016 08:38PM by tin.

Quote
mekilljoydammit
No, don't try to sharpen the divider on the exhaust side, likely doesn't help anything but makes a hot spot. The Mazda K series V6s are actually pretty good - Ford used a heavily reworked one destroked to 2L in British touring car in the late 90s. .

And Ford were massacred. Crushed. Annihilated. Stomped. Kicked. Murdered.
That time frame is the only time i ever voluntarily looked at anything on asphalt circuits..
It was easy to understand why Ford's decision was doomed...
With the then max rev limit of 8500, there were zero advantages to the smaller and lighter components in a 6...The extra 2 pistons rods, the extra 8 valves, lobes, buckets all that was a LOT more drag---losses, than a 4...It is after all 50% more of all that stuff rubbing and sliding..

No advantage.....

Oddly it is what I frequently ask people to consider "What advantage does this (car, part, operation. technique) have over ________________"?

I don't want to get wrapped up in pointless debate but....what advantage will say "super marvelous" ports actually be to do what? Rev more? What good is that....without gearing to turn all those millions of revs into acceleration?
Is there alternate gears for the final drive? Or the box? Is there a diff?

---

John Vanlandingham
Sleezattle, WA, USA

Vive le Prole-le-ralliat

www.rallyrace.net/jvab
CALL +1 206 431-9696
Remember! Pacific Standard Time
is 3 hours behind Eastern Standard Time.

Quote
john vanlandingham
I don't want to get wrapped up in pointless debate but....what advantage will say "super marvelous" ports actually be to do what? Rev more? What good is that....without gearing to turn all those millions of revs into acceleration?
Is there alternate gears for the final drive? Or the box? Is there a diff?

Not the best but for stocker it isn't the worst by a long shot......200km/h top speed with the 1.8 V6 not sure how the 2.5 revs in comparison.

---

Power means nothing if you cannot control it!

John, I totally get what you're saying. I ended up with the car for next to nothing and the engine choice, while it could have been a 4cyl 2L with less moving parts, ended up being a v6 because it offered a bit more displacement at basically the same weight as the stock 1.8l. This is what the torque and power curves look for the 2.5:



The 2.5 revs to 7000 also, but falls flat after 5 ish. I believe the MX-6 and 626 came with a final drive of 4.11. What would be an example of ideal or good gearing?

So the advantage of the 6 over a 4 is (correct me if I'm wrong) more displacement, moar torque, same weight, same cost (junkyard prices them the same). Disadvantage is cost of rebuild, time of rebuild, reliability?. Speaking of reliability, the tolerances in my 200k km 2.5l engine are all withing service limit.

As for the diff, its open unfortunately. There is a LSD that came in the Protege gearboxes, it's planetary gear type, I'm pretty sure. I think Kaaz makes a clutch diff for the G series box as well. What about welding the stock diff for the short term? I don't have the $$ for a diff this year. What is it like driving a FWD with a welded diff anyway?

Quote
john vanlandingham

I don't want to get wrapped up in pointless debate but....what advantage will say "super marvelous" ports actually be to do what? Rev more? What good is that....without gearing to turn all those millions of revs into acceleration?
Is there alternate gears for the final drive? Or the box? Is there a diff?

Point I was tryna make was, hey, it's a not really bad engine as is, so don't bother messing with it too much.

Quote
tin
What would be an example of ideal or good gearing?

It depends a bit on the engine characteristics, tire size, etc but for naturally aspirated engines this will give you a good idea.
5 speed
1st: 2.337/1
2nd: 1.681/1
3rd: 1.358/1
4th: 1.150/1
5th: 1.000/1

4 speed
2.3/1
1.5/1
1.2/1
1.0/1

For rally these ratios will get paired with anything from a 4.3:1 rearend (turbo, lower revving also a turbo motor will a lot of times have wider gear ratio spacing) to a 5:85:1 maybe even shorter for a seriously high revving motor. Again, depends on a number of factors. -

The 5 speed ratios are copy/ pasted from a e30 m3 group a spec sheet. Same gearbox that a number of teams used (volvo, I think jag, and ford, among others). Granted these ratios are from the 80's but if you go shopping for a 5 speed close ratio gearset offered now the ratios will be more or less the same.

These are examples of ideal ratios. To get them in your car will take a few bucks.



Edited 1 time(s). Last edit at 10/30/2016 11:42AM by Eric Ewert.