Q&A Forum
I am in the process of setting up my dyno - when I look at a run without any filters applied, I am seeing a lot of noise on the load cell data to the point that when I apply the filters I am shocked it can make any sense of the data. I have made some changes to the load cell mounting method and this has improved the data, however, I do not know what good looks like for unfiltered load cell data - please can somebody share a run with unfiltered data to help me understand if I actually have a problem or if this is typical? Thanks.
The dyno is a prony brake style with the engine driving an axle and controlling the hydraulic brakes in a similar way to controlling a water brake. I think it's fair to say there is very little inertia in the system? Applying the 'standard' 3, 3, 0 filter does indeed filter it out and make it look like a reasonable set of data, it just seemed so noisy. There are a few improvements planned to the axle mounts which will likely improve any mechanical/vibration induced noise, but you have given me more confidence that it's not out of the ordinary. Do you have a typical water brake run where you could share the unfiltered data?
It is quite high but not out of the ordinary. What type of dyno and how much inertia is it? It is quite easy to filter out, maybe except around 3000 RPM.
This is an example of the unfiltered data, please let me know what you think. Thanks.
Hi Jostein, thanks for the reply - I can't see an option to post any images, do I need to reach a minimum number of posts before this option is available? Thanks.
Hi, the noise varies quite a bit from one dyno to the next. Low inertia, direct connected, single cylinder typically gives a lot of noise while high inertia soft coupled are smooth.
Results should look good with filtering between 3 and 6. 3 is default, and probably 80% of the setups are ok at 3. Post your unfiltered graph and describe the dyno system and I can evaluate.