CB7Tuner Forums

Go Back   CB7Tuner Forums > Beginner forum -- New members post here! > Beginner Technical/Performance

Reply
 
Thread Tools Rating: Thread Rating: 2 votes, 4.50 average. Display Modes
Old 09-19-2017, 02:44 AM   #21
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
deevergote,

Just recently I have stumbled upon a racing team called "KPO motorsports" and their all motor CRX. Turns out they are also running F22A and on their way to follow Bisi's footsteps. Too bad they are not Americans (at least so it would seem), which means that there is almost no (for me) readable information on internet other than a few videos on youtube with 1/4 mile runs and pictures on facebook. Looks like they have already reached 360 whp range, I will make sure to follow their progress.

sonikaccord,

Actually I am using another program for this very purpose. It is using CFD or flow-bench data in order to calculate the effects of runner lengths and almost any other spec of four stroke engine. I can run the calcs if you like, but it will be of most benefit if you could list up the specifications of your engine. For example:
- head mods
- cam specs
- current runner lengths and diameters (intake + exhaust)
- bore x stroke
- rod length
- static compression ratio

The result of all this will depend on the accuracy of the data that you provide. And of course let me know your desired peak power RPM. From my experience tapered intake runner will give you more area below the curve at the expense of peak power, and cylindrical runner, or one that does not change the cross-sectional area over length will do the opposite (provides the most ramming effect).

As for the CFD,

I am getting a little worried that with 36 mm valves, this thing might flow a little too well for any displacement H or F series engines are capable of running at reasonable RPMs, meaning that the flow velocity could be improved. This might also be one of the reasons why PT3 head is known to respond to mods not as good as other engines (at least what I have read). You would have to go all out to take advantage of the design, which is what Bisi did and KPO are doing now.
I plan on doing some more experiments with flow simulations, but this time I will focus on not removing the material, but adding it. We can always use epoxy for this kind of work. I will be expecting still astronomical CFMs with an overall smaller port, meaning that average flow velocity has been increased, which should in theory make quite a big difference in performance.
Stay in touch!
KBA is offline   Reply With Quote
Old 09-19-2017, 04:16 PM   #22
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
Slowly initiating some progress here. I am now in the process of building a CAD model that allows me to easily control the shape of the port by adjusting these projection splines. I have another set of those projected on the top plane to control the width. It will then update the surfaces to adapt to the new shape.
Based on my experience from previous CFD + CNC porting operations, this combination of longer short side radius (floor) and shorter long side radius (roof) should yield me better CFM readouts with a smaller port volume. I guess we will find out when the model is complete and sims finished.

KBA is offline   Reply With Quote
Old 09-21-2017, 03:11 PM   #23
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
I still have not finished the new port design or run the sims, but it is in process. Just taking a lot more time to complete than I expected.
In the mean time I have something else to show you all. I have finally pieced together all the data necessary to re-create a stock F22A6 dyno chart in my laptop. In essence - I have now got a virtual F22A6 that I can dyno at any time, with any mods.

- Short block: specs by "The official H/F piston/rod/crank/block specs and CR thread"
- Head: scanned & measured by me
- Compression ratio: by zealautowerks.com
- Intake system: thanks to aventari
- Cam / valvetrain: scanned & measured by me
- Exhaust system: a rough estimate (maybe somebody will respond with precise measurements of stock F22A6 header tubes - length and diameter)

Fuel: 93 octane pump gas
Ambient conditions: according to SAE



According to advertised numbers, this should be quite close. Wait till we start dropping in some virtual parts

Last edited by KBA; 09-21-2017 at 03:18 PM.
KBA is offline   Reply With Quote
Old 09-21-2017, 05:14 PM   #24
Rilas
Gold Member
 
Join Date: Apr 2010
Location: Missoula, MT
Posts: 430
Rilas seems to have made some friends!
I'm so stoked to see what actually comes of this project that you are working on. This is some very cool stuff that you are doing for our community. All of this information is going to be a goldmine in the future. I commend your effort and time that you have put into this project for all of us.
__________________
MRT: 1993 Honda Accord SE Coupe (Lola)
Rilas is online now   Reply With Quote
Old 09-22-2017, 07:54 AM   #25
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
Thanks Rilas!

Today I found this pic on google. It is said to show H22A in cross section. I found it interesting to introduce this image in CAD software and overlay with my scanned PT3 (F22A) file.
I must say, there are quite a few resemblances, yet the intake angles on the PT3 seem slightly different. The combustion chamber is not as deep which positions the F22 valves slightly lower, allowing a greater floor radius. The intake runner openings match perfectly, and the exhaust port is smaller and placed differently when compared to H22.
Let the image speak for itself!

KBA is offline   Reply With Quote
Old 09-26-2017, 07:19 AM   #26
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
While I keep you all waiting for the progress of port mod CFD, I just thought this might seem interesting to some of you.
Since I am working in exhaust manufacturing company, I have to fill my free moments somehow every now and then, and I've been doing that by designing a killer header for the F22. It incorporates stepped primaries and should hit pretty hard @ 7800 RPM. Nevertheless, it is still in design phases and some of the things that must still be done are primary length matching to a higher degree and addition of secondary steps. Then it's 4:1 collector, which I already have designed to a certain degree and I can start talking with the guys at shop floor to fabricate the parts.




KBA is offline   Reply With Quote
Old 09-26-2017, 12:40 PM   #27
Jon
Contributing Member
 
Join Date: Jan 2007
Location: Treasure Valley Idaho
Posts: 1,754
Jon is very helpful Jon is very helpful
Thank you for everything you are doing. This sort of research on a nearly 30 year old engine is just awesome to see and I'm sure it will help many at some point in the future.

I'm definitely interested in the header design you're working on. I was one of the many who was suckered into buying a Bisimoto header to not only be plagued with leaks, alignment, and fitment issues, but lack of performance (even though he says each one is tailored to a specific spec sheet provided by the end user, they aren't).

Thanks again and keep up the awesome research!
Jon is offline   Reply With Quote
Old 09-26-2017, 12:56 PM   #28
Bunta
CB7tuner Member
 
Join Date: Feb 2015
Location: Texas,Arizona and Massachusetts
Posts: 198
Bunta is cool... so far.
Looks like that header is mounted on an H22 though (not that us Hseries owners would mind another option). Do you have something for the Fseries worked up or is the picture of it mounted just using a generic engine picture as an overlay?
Bunta is offline   Reply With Quote
Old 09-26-2017, 02:47 PM   #29
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
Jon, thanks!

Bunta, the engine picture is just an overlay to give some idea about the proportions. I used H22A since I was able to find this detailed split on google and these engines are virtually brothers to each other. The CAD prototype you saw was designed specifically for PT3 F22A head geometry using my scanner session files.

About the header: I just ran some quick 1D engine simulations of how this header might perform on stock F22A6 when it is complete. One thing that is definitely a must is free flowing exhaust. I used only estimates (150 CFM for stock exhaust system and 300 CFM for opened up system). If one was to use that header on a bone stock setup, he would lose 5 - 7 hp.

You can see how it performs similarly on the low RPM and starts to hit near the redline. The design itself is good to go up to 8000+ RPM, it is the stock cam that is seriously holding it back.



Meanwhile, the port design continues...

KBA is offline   Reply With Quote
Old 10-08-2017, 05:48 AM   #30
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
...and once again I have faced delays with this project. As it turns out, I went after a PhD in mechanical engineering and I will have to work on other projects in parallel. I will be developing a throttle body of unseen design and concept for my CRF450R supermoto bike with hopes that it could one day see some business. So.. sorry for that, but anyhow, the sims are now running with the new port shape, 36mm valves and cut guides. Here's a screenie

The results should be posted later today...

KBA is offline   Reply With Quote
Old 10-09-2017, 02:53 AM   #31
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
The calculations are finally done, and I am a bit surprised. The flow rate went down the slope with the raised floor, shorter roof and 36 mm valves - 330 CFM. This leaves me curious about the port volume, if it is significantly smaller, then the overall efficiency might still be improved, but if the volume has not changed much, then I might suspect that the engineers that designed the PT3 head knew exactly what they were doing. Time to regroup and find another way at it. Maybe the characteristic of the Honda fourbangers really does prefer a longer roof like it is stated on some CNC porting articles. I guess we shall see!
KBA is offline   Reply With Quote
Old 10-09-2017, 07:55 PM   #32
aventari
CB7tuner Member
 
Join Date: Nov 2010
Location: San Diego, CA
Posts: 135
aventari is cool... so far.
What's your opinion on knife edging the port walls? I did that to my head last month when I did a mild port/polish job for my Lemons race car.





I can't say for sure that the porting helped, but with the new head and port job and new rings the car definitely pulled harder on the top end and I was passing cars on the high speed straights that should have a lot more power in theory. I'm sticking with just the A6 cam though as reliability is paramount in endurance racing, and keeping the RPMs down is how you do it.

Other than knife edging, I basically just cleaned up the casting and gasket matched the intake

Last edited by aventari; 10-09-2017 at 07:57 PM.
aventari is offline   Reply With Quote
Old 10-10-2017, 07:51 PM   #33
sonikaccord
Contributing Member
 
Join Date: Nov 2005
Location: In Traffic
Posts: 4,086
sonikaccord seems to have made some friends!
Man, maybe I do need to get my master's degree! You are doing some really interesting things.
__________________
<Click for story - Welcome back!
sonikaccord is offline   Reply With Quote
Old 10-12-2017, 05:56 AM   #34
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
Aventari,

knife edging always helps! Sure it's not astronomical, but cleaning up the casting debris is a slight gain. None the less, it seems that these ports are already heaven from stock, just like Bisi said, the engine will only need a cam and compression. The only thing keeping the int. ports from going to the moon with CFMs are the narrow valve seats that score a flow rate equivalent to many great heads running bigger valves. Remember - I picked up almost 100 CFM @ .550 lift on my CFD analysis by swapping to +2mm valves and enlarging the seat throats. And on the exhaust side, I can tell you that exhaust ports are generally overrated. The thing that people ignore is that you have only so much of pressure differential on the intake (significantly below 14 PSI) but the exhaust sees thousands of PSIs when the valves open, making the exh. port CFM that much more insignificant than intake. Sure, it helps but nowhere near as drastically as intake. Besides, you don't want your gasses to escape the chamber too quickly, because it might screw up your tuning events. I remember once I did some experiments with increasing the exhaust cam lift on my bike, and believe it or not, but I actually lost power. The bottom line is: it all depends on your setup so bigger is not always better.
I also built my bike for reliability because I cannot feed enough money into it if it has to be rebuilt every season. The one trick I did for that - I called WEB and asked them to list down the available profiles out of which I picked the ones that had the most duration while retaining low lift. The engine now runs a lot quieter than it did, and produces gobs more power. Sure, not as much as it could with high lift, but not very far from that. Low lift and long duration is key to happy valvetrain.

Sonikaccord, I can only encourage you. Those studies sometimes open up a whole lot of possibilities

Last edited by KBA; 10-12-2017 at 05:59 AM.
KBA is offline   Reply With Quote
Old 02-07-2018, 06:10 AM   #35
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
Just so you guys wouldn't think I'm dead - still working on the PT3. I have done several more CFD sims, and the final results are as follows:

Intake:

Stock (cleaned) port with 36 mm valves (C32B), blended valve seats - 310 CFM @ 0.500 in @ 28 in H2O (with filled injector cavity, can probably score another 50+ CFM with this cavity present, but it does not function when the runners are installed. Bare in mind that CFM reading you see online are with injector cavity functioning as part of port entrance.)

Exhaust:

Stock port with 30 mm valves, blended valve seats and bigger entrance - 210 CFM @ 0.500 in @ 28 in H2O

I think there is no need of "properly porting" this head. Just some basic operations because it is already awesome from the factory. The small valves are what chokes it.

Right now I am hunting for some AMPCO 45 billets to machine my guides and seats, and then I will be shipping my cam to WEB for a custom hardweld + regrind.

Meanwhile, I have mostly finished the CAD for my header. Check it out:

ID42mm x L155mm + ID47mm x L300-500mm + ID42mm x L50mm + 4:1 collector + 67mm main pipe. The steps seem wierd, but in fact, they should extend the torque curve to the top of RPM.

KBA is offline   Reply With Quote
Old 02-07-2018, 07:03 AM   #36
NolsAccord_CC1
Newcomer
 
Join Date: Aug 2017
Location: Europe
Posts: 7
NolsAccord_CC1 is cool... so far.
I really admire your work! but wouldn't that choke the exhaust? Wouldn't it be better for torque to just use the 42mm throughout?
NolsAccord_CC1 is offline   Reply With Quote
Old 02-07-2018, 07:23 AM   #37
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
Quote:
Originally Posted by NolsAccord_CC1 View Post
I really admire your work! but wouldn't that choke the exhaust? Wouldn't it be better for torque to just use the 42mm throughout?
Well that's just the thing... I too once thought that it is important to keep the exhaust header uniform to maximize the speed of a greater volume of gas. But the deal with exhaust is never that simple. What is really important for the intake, is less important for exhaust and vice versa. While you want an unobstructed high speed flow at the intake side, exhaust couldn't care less. It cares more about the back-and-forth propagation of sonic (pressure) waves in the tubes which is determined by the changes in cross-sectional area A.K.A. tube diameter. By creating a stepped path, we can create wave reflection points. Stepping to a larger diameter reflects a negative wave, stepping to a smaller diameter reflects a positive wave. By adding these steps at the correct points, we can cancel out the waves when they are not wanted, and amplify them when they are of benefit. From what I have found, the greatest exhaust "sucking" effect can be found with longest possible tubes for the given RPM. This will have a downside however, it will virtually kill the torque in any other RPM. Using shorter tubes will broaden the torque curve, but it will not reach as high. Using steps in combination with longer tubes will give you the best of both worlds if done right. In this case, the larger section in the middle is used more as a damper or "resonator" to induce the wave interaction. My expectations are that torque will start mild, gradually increase along with RPM and die off very close to the power peak. Similar to what you would see with a centrifugal supercharger.

Sorry for such a long write up but I can't think of another way to explain this..
KBA is offline   Reply With Quote
Old 02-07-2018, 09:16 PM   #38
NolsAccord_CC1
Newcomer
 
Join Date: Aug 2017
Location: Europe
Posts: 7
NolsAccord_CC1 is cool... so far.
Smile

No, thank you for your long write up One can never know too much.
For how much you think you can build this header?
NolsAccord_CC1 is offline   Reply With Quote
Old 02-07-2018, 10:11 PM   #39
Rilas
Gold Member
 
Join Date: Apr 2010
Location: Missoula, MT
Posts: 430
Rilas seems to have made some friends!
KBA, I'm curious as to how far out, you might be on actually making this header, and dyno'ing your setup?

I love everything that your doing for the community as a whole. I'm seriously interested in the results, due to everyone thinking the F22A/PT3 platform isn't worth even mentioning.

What are your thoughts about cam options? Are you just going to get a hardweld with high lift and such for the high rpm, or something that has a better torque curve?

I'm sure you have more than enough knowledge gained and forgotten in the engine building department than most of us will ever have even gained. You may want to read a previous members ride thread, as he pioneered a lot on the F22A/PT3 platform. Here is the link jdm92_accorn.
__________________
MRT: 1993 Honda Accord SE Coupe (Lola)
Rilas is online now   Reply With Quote
Old 02-08-2018, 01:32 AM   #40
KBA
CB7tuner Member
 
Join Date: Jul 2017
Location: Latvia
Posts: 35
KBA is cool... so far.
NolsAccord_CC1,

Well, I am not sure really.. The laser cutting could cost about 20-30 Eur including material. The pipe bending could reach as much as 100-150 Eur including material as there are 4 pots X 5 bends = 20 bent pieces (each with about 50% wasted material). The collector is sheet metal so the rolling might cost another 20 Eur. And then there is TIG welding. (TIG because I like my pipes to look good) This could be another 50 - 100 eur.

Add everything up and we are looking at 200 - 300 Eur as just cost-price. It could be done cheaper I guess but that would require higher quantities. The biggest problem are the bends. We have a zillion companies over here offering different pipe bends but they are all for construction purposes so the sizes are way off from what I need. The only option is to get them done at my employer but he is a pain in the... when it comes to anything other than serial production bringing big bucks. We shall see.

Rilas,

It is always nice to hear such kind words, thanks! Generally I do not like to think of myself as very much "experienced", I am simply an enthusiast with an open mind and endless curiosity.

The cam in my opinion is the reason why no one is considering this setup worth mentioning. It is virtually non-existent when compared to any other engines that are called worth-while, like H22. I usually get very mad when researching this stuff online, I see 99% of people judging engine's potential based on the number of cams and presence or absence of VTEC. The major thing that keeps the F22A back from cranking H22A like numbers is the cam. And it is not because there is only one, it is because the profiles are very short. There is no overlap. It barely lifts the valve in comparison. On the other hand, the base circle is bigger than H22A and the rockers have more leverage, this is again something that Bisi found. The valvetrain is setup the same way the head is. The potential is there but the small valves and cam are whats keeping it back. The regrinds that you can find are decent, but still not enough to really open that thing up. Personally I do not like regrinds because they cut down on reliability by decreasing the base circle and compromising the hardness.

About the torque curve, after having a closer look at the overall design, I think there really are better heads for reasonable torque curves, like the F23A. It has a smaller port that flows better at low lift which is ideal for strong low and mid torque. The F22A head, however, provides the opposite: large port that outflows anything at higher lifts. This is ideal for mid to top end torque and in my opinion one more reason why the engine lacks performance per displacement - it is set up for low RPM from the factory. So overall what you have is a professional sprinter put to weightlifting, a very good power-plant used for the wrong purpose. That's describing the F22A in metaphor. Personally I will be aiming my build to whatever the stock F23 block can safely withstand with lighter pistons, better bearings, ported oil pump and perhaps a block guard. Could be 7500-8000 tops, so I will be aiming at 6500 RPM peak torque and 7500 RPM peak power.
At the moment in cam specs that would be about 240/268 deg @0.050 and around 13/12 mm lift. The cam is practically made for such specs..

About the header, I think I will get it done, the question is when.. Since my hands are full at the moment I cannot pay full attention to anything I am doing, only here and then.. We shall see.
Reading through jdm92_accorn's post was the first thing I did when coming here. The same goes to Ghost and his blog

Last edited by KBA; 02-08-2018 at 04:23 AM.
KBA is offline   Reply With Quote
Reply

Tags
cfm , f22a , porting , pt3 , valvetrain

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 03:23 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.