This post has been de-listed
It is no longer included in search results and normal feeds (front page, hot posts, subreddit posts, etc). It remains visible only via the author's post history.
I'm looking to build a dyno so I can learn to tune at home, I'm deep in the research phase, and I can't work out if steady state tuning on an engine dyno is going to give the same result as it would on a chassis dyno. I don't see why it wouldn't, in my eyes, it shouldn't matter how the load on the engine gets to it, just as long as it does. If I were to build an engine dyno that allows for RPM hold the way a chassis dyno does.
Would steady state tuning an engine that way yield any differences to the tune vs steady state tuning on a chassis dyno with the engine in the car?
Plan would be on the engine dyno to run the exhaust, intake, everything that would be in the car on it for the tuning process.
Reason is it'll be a bit simpler to build an engine dyno vs a chassis dyno, and the space needed for an engine dyno would be a lot less. Boxers are small motors and I want to make sure I can hear the knock if it occurs, much easier if I'm in a shed than out on the road. I have a few spare motors, including two with over 300K on them, if they pop I'm not going to cry because one of them is well on it's way out and the other's got to be close.
Just so it's not a question to be answered, at this stage I'm tossing up a truck torque converter with the output side locked or braked, and a water pump as the load. I can measure using a load cell and could probably even feed that info directly into the motec to log it there for tuning purposes. Otherwise, I'm not beyond building an arduino based system to measure RPM and load cell load.
Cheers.
Subreddit
Post Details
- Posted
- 11 months ago
- Reddit URL
- View post on reddit.com
- External URL
- reddit.com/r/ECU_Tuning/...