As most know by now I build Ducati race engines and have been trying to find the best developed Ducati head.
My developed Ducati 4v head with a 42mm valve (all testing done at 10") depression supports 150hp.
My head has poor low lift flow but takes off rapidly from 6mm upwards. Over the other side of the world a friend has done the same head but with a 41mm valve. We both have the same cross section at the splitters but mine has 75 degree valve insert throat angles where he has radiused 80 x 60 x 45 x 37 angles with his throats being 35.5mm wide where mine are 37mm wide.
My friend tests at 28inch and after converting his flows to 10" depression (for comparison) his smaller valved modified head is very very good in the low and mid lifts and equals mine in the high. He also offset the head 1mm on the barrell to gain extra flow from being shrouded by the bore walls.
His valves have back cuts, mine don't - even the Ducati race engines don't.
Anyhow he set the motor up as I recommend, took it to the dyno and the results were unexpected. His power figures were quite low compared to mine even though his heads out flowed mine with a smaller valve.
What's going on ? If most of the flow on these heads goes across the combustion chamber and if I were to bowl out the longside radius I wonder if I could direct more flow on top of the valve away from the exausts...
ideas welcome!
Thanks,
Chris
Ducati heads - low lift flow and power results
-
superbike
- Posts: 91
- Joined: February 13th, 2007, 4:19 pm
- Location: south glos
-
Guy Croft
- Site Admin
- Posts: 5039
- Joined: June 18th, 2006, 9:31 am
- Location: Bedford, UK
- Contact:
Chris, hi
Comparisons of that type are hard to do and unless a high degree of commonality exists between the engines it's easy to read too much into them.
For starters I need to know that both engines are identical other than differences described above. In other words, cubic capacity, cams, cam timing, ignition timing, CR, intake system layout, filters, jetting, ex headers and silencer type and anything else you can think of.
In addition how was the power measured (bench or rolling road dyno) and were the bhp results properly corrected for test cell atmospheric pressure, temperature and humidity?
If there are any differences engine-to-engine please ID them here so I can evaluate this. Power curves would be a big help.
GC
Comparisons of that type are hard to do and unless a high degree of commonality exists between the engines it's easy to read too much into them.
For starters I need to know that both engines are identical other than differences described above. In other words, cubic capacity, cams, cam timing, ignition timing, CR, intake system layout, filters, jetting, ex headers and silencer type and anything else you can think of.
In addition how was the power measured (bench or rolling road dyno) and were the bhp results properly corrected for test cell atmospheric pressure, temperature and humidity?
If there are any differences engine-to-engine please ID them here so I can evaluate this. Power curves would be a big help.
GC
-
superbike
- Posts: 91
- Joined: February 13th, 2007, 4:19 pm
- Location: south glos
-
Guy Croft
- Site Admin
- Posts: 5039
- Joined: June 18th, 2006, 9:31 am
- Location: Bedford, UK
- Contact:
-
Guy Croft
- Site Admin
- Posts: 5039
- Joined: June 18th, 2006, 9:31 am
- Location: Bedford, UK
- Contact:
Flow test results from two flowbenches of different capacity and the Ducati motorcycle heads tested on them, the differences between the heads are identified above.
Superflow tables enable the flow figures from tests at say, 10" to be compared with tests run at higher test depression, but as the graphs of standard heads tested at two different regimes show (on heads that are 'ostensibly' identical) there can be a significant difference between them, and this is more likely due to the behaviour of the airstream when subjected to a higher pressure ratio across the valve rather than any 'experimental' error.
Yes, there is a difference in gains between one head and another, but it's NOT massively better (if actually at all) because the apparent gains may simply be due to the effect of higher depression rather than 'superior' reworking. It emphasises that direct comparison of 'before' and 'after' on a head has to be done at the same test depression.
Superflow tables enable the flow figures from tests at say, 10" to be compared with tests run at higher test depression, but as the graphs of standard heads tested at two different regimes show (on heads that are 'ostensibly' identical) there can be a significant difference between them, and this is more likely due to the behaviour of the airstream when subjected to a higher pressure ratio across the valve rather than any 'experimental' error.
Yes, there is a difference in gains between one head and another, but it's NOT massively better (if actually at all) because the apparent gains may simply be due to the effect of higher depression rather than 'superior' reworking. It emphasises that direct comparison of 'before' and 'after' on a head has to be done at the same test depression.
- Attachments
-
- Ducati inlet flows 41 vs 42mm valve.GIF (27.07 KiB) Viewed 8552 times
-
ADsim
- Posts: 3
- Joined: September 8th, 2007, 5:12 pm
- Location: Arpajon, France (A)
- Contact:
Hello ,
I am not a good mechanic (there is a lot here) and I prefer playing with simulators and I am told sometimes that they are not very accurate. It's true because many time we cannot load the right inputs.
So I read a lot of SAE papers from guys who spent their life (and are paid for!) to try to understand why.
SAE paper 2000-01-1894 (***/***) speak about two 748 Ducati heads having same duct and port profiles , same CR and squish. The only difference is cuts in head pent-roof done around inlet and exhaust valves :
- one where cuts junction are smoothed by a big radii (chamber A) (like yours superbike)
- this other have sharp transition (chamber B)
Chamber A have better flow coefficient by 2.2% at high lift. Tumble ratio (air entering in cylinder and rolling like a water wave with horizontal axis - not a vortex that have vertical axis) is also better for chamber A by 22%.
On bench engine A was superior to B (same ignition timing and best AFR for each) due to better to higher Cd and higher tumble level especially in low rpm range.
Chamber A accepted richer AFR at high rpm due to higher flame speed to help give a 2.2% Hp more.
In another paper (2005-01-1530) (***/***) with same heads , it is said that at low lift , bigger valve become closer and give "a negative flow interaction" that become positive at high lift but without great details...
SAE paper 2002-01-0244 (***/****) is on same subject. Petronas tested to very similar head of a F1 engine having two different tumble levels.
Flow coefficient test give very similar results for each head. Twice the tumble level remain the mean difference.
Heads are "drilled" to install 12 flame wires detectors (1.1mm copper wire !) :
- 4 around plug
- others around bore
Single cylinder engines were tested at 14 and 17000rpm for better performance and to acquire flame speed during operation.
Here , low tumble engine won against too high tumble one. It made less power with near similar advance but delay and combustion was longer by 2‚° crank (tim=49.9/del=38.2/dur=68.7‚°).
Too much tumble is not adequate also and the only way to model this is to set correctly combustion duration that is something that most of us cannot quantify ...
I don't say that this is your problem but it could explain partially.
(***/***) is my rating of papers from 1 to 5 stars
Regards
Alain
I am not a good mechanic (there is a lot here) and I prefer playing with simulators and I am told sometimes that they are not very accurate. It's true because many time we cannot load the right inputs.
So I read a lot of SAE papers from guys who spent their life (and are paid for!) to try to understand why.
SAE paper 2000-01-1894 (***/***) speak about two 748 Ducati heads having same duct and port profiles , same CR and squish. The only difference is cuts in head pent-roof done around inlet and exhaust valves :
- one where cuts junction are smoothed by a big radii (chamber A) (like yours superbike)
- this other have sharp transition (chamber B)
Chamber A have better flow coefficient by 2.2% at high lift. Tumble ratio (air entering in cylinder and rolling like a water wave with horizontal axis - not a vortex that have vertical axis) is also better for chamber A by 22%.
On bench engine A was superior to B (same ignition timing and best AFR for each) due to better to higher Cd and higher tumble level especially in low rpm range.
Chamber A accepted richer AFR at high rpm due to higher flame speed to help give a 2.2% Hp more.
In another paper (2005-01-1530) (***/***) with same heads , it is said that at low lift , bigger valve become closer and give "a negative flow interaction" that become positive at high lift but without great details...
SAE paper 2002-01-0244 (***/****) is on same subject. Petronas tested to very similar head of a F1 engine having two different tumble levels.
Flow coefficient test give very similar results for each head. Twice the tumble level remain the mean difference.
Heads are "drilled" to install 12 flame wires detectors (1.1mm copper wire !) :
- 4 around plug
- others around bore
Single cylinder engines were tested at 14 and 17000rpm for better performance and to acquire flame speed during operation.
Here , low tumble engine won against too high tumble one. It made less power with near similar advance but delay and combustion was longer by 2‚° crank (tim=49.9/del=38.2/dur=68.7‚°).
Too much tumble is not adequate also and the only way to model this is to set correctly combustion duration that is something that most of us cannot quantify ...
I don't say that this is your problem but it could explain partially.
(***/***) is my rating of papers from 1 to 5 stars
Regards
Alain
Last edited by ADsim on October 29th, 2007, 5:44 pm, edited 1 time in total.
2T4T
-
ADsim
- Posts: 3
- Joined: September 8th, 2007, 5:12 pm
- Location: Arpajon, France (A)
- Contact:
Hello superbike ,
You have already a high level and I don't think you will learn a lot about porting. Each paper is $12 (9¢‚¬Å¡‚¬) but generally only 8 pages including title and bibliography : quiet short and cannot replace a book.
I would suggest the Hyundai paper for beginners as simple and formula included.
For future , I will try to rate them for beginner/expert (3 to 5 stars) in my posts.
Regards
Alain
You have already a high level and I don't think you will learn a lot about porting. Each paper is $12 (9¢‚¬Å¡‚¬) but generally only 8 pages including title and bibliography : quiet short and cannot replace a book.
I would suggest the Hyundai paper for beginners as simple and formula included.
For future , I will try to rate them for beginner/expert (3 to 5 stars) in my posts.
Regards
Alain
2T4T
-
superbike
- Posts: 91
- Joined: February 13th, 2007, 4:19 pm
- Location: south glos
I still dont know when i will recieve the other chaps dyno graph to finish this post . I can say my engine with the poor low lift flows made a whopping 153 hp and the identical bike the other side of the world with very good low lift flows struggled at 142.superbike wrote:Hi guys
There will be a follow up to this thread quite soon with the dyno results. Im just waiting for the data..
As soon as his data arrives it will be posted
Who is online
Users browsing this forum: No registered users and 3 guests