Discuss all things tuning in this section. News, products, problems and results.
I had the experience of tuning my car on the dyno a few weeks ago, supercharged 1.7L and threw in some E85 at the end to crank it up a notch. I'm using a Vipec V88 wired to my 92' Miata, and using an audio amplifier in conjunction with musician in-ear monitors (have -27dB NRR) with high quality ear muffs overtop (-31dB NRR)
One thing I noticed was that the car ran much better with knock control turned off. I found that the sensor was very easily triggered under false knock due to a very noisy engine. To confirm this, my audible knock detection would get pretty noisy during high RPM but it did not crackle and pop. I am fairly confident that the engine is tuned safety and conservatively after only tuning with audible knock detection and doing multiple ramp runs with increased run time back to back.
I have a few questions regarding my experience:
Should I look into adjusting the knock levels or trying to pick a low pass threshold for the knock detection to work properly? For example, on my logs the engine noise levels will typically be around at 70 and then jump to about 200 over 6000rpm. This can be quite annoying when I feel a sudden drop in power when false knock gets triggered. Would I be able to set the knock targets on the street or would it be better adjusted on the dyno where I can hold the load a little easier at a particular cell area? See the attached screenshot from one of my logs at a track day
My other question is regarding my audible knock detection setup - albeit a very basic setup, I feel that I can really hear details very well with a knock sensor connected to an audio amplifier. With noisy valvetrain such as mine, do higher end audio knock detection setups filter out the high level harmonics? Will they be able to differentiate between valvetrain and knock or would I be facing the same issues since the fundamental problem is a noisy valvetrain?
The dyno I was using was an older Dyno Dynamics unit with analog controller, one thing I found very difficult or rather took some learning to adjust was the load itself using the small dial. I did find that by adjusting the ramp run rate that the power output shown would change slightly as well, is this a result of using an older dyno that perhaps cannot consistently calculate the power output if one of the variables is changed?
Although the max power output is not a major concern for me tuning wise, it does play an effect on the time trials classes I occasionally compete in which require a power-to-weight ratio calculation, I have found that it has been very hard to adjust the ramp rate to accurately reflect what "other" dynos may read for power.
I must say that the Learn to Tune webinars and tutorials have definitely helped speed up the learning curve. Once on the dyno it helped give me an advantage to focus more on learning how to operate the dyno and effectively used the time I had booked. Thanks Andre and Ben!
I can't help for most of your questions, But I am using a Phormula KS Pro Audible Knock setup. And I am still get a fair amount of valve train noise. I am new to tuning, I find my self having to concentrate extremely hard when monitoring! Maybe this is inexperience, Or maybe I don't have any more filtering than your knock monitoring setup!?
For my Knock control i set the limit on the street without any problems, I knew took out a few degrees of timing, Took a bunch of logs and then set the threshold around 10% above here the logs were showing the engine noise level. It does seem to be quite exponential though as the revs increase.
Even a quality audio knock detection system will inevitably allow some background engine noise to be heard. The key advantage of a product such as the Plex Knock monitor for example is that you can choose the centre frequency that knock occurs at, and the processor will implement a filter that allows this particular frequency to be heard much more clearly.
When tuning the closed loop knock system, you need to confirm that the ECU can accurately detect knock in the first place. I confirm this using audio knock detection so that I can audibly hear the knock and make sure that the ECU is picking it up. I do this initially by holding the engine in steady state and advancing the timing until the engine knocks. Obviously this needs to be approached with some care though.
Once you have the closed loop system accurately detecting knock, you can progress to performing ramp runs. Using the audio knock detection equipment I make sure the engine isn't suffering knock, and then tune the knock threshold to suit.
From your explanation, it does sound like the knock system is possibly picking up false knock. While it isn't exactly applicable to the V88, check out this webinar on knock control we shot using the Motec M1 - https://www.hpacademy.com/previous-webinars/motec-m1-knock-control-example/