Why are revlimiters uneven?

Commander Kinsey snipped-for-privacy@nospam.com wrote

No idea what you are on about this time.

Rather different technology. Many dimmers cant, they cut off abruptly at the lower light levels.

On cars? Yes. Cutting the engine abruptly when you hit the red line wouldn't be good so they simulate ignition breakup to give you a hint.

Never saw the original post. There is a holding current in a triac, and once the current drops below this a kind of Hysteresis occurs so you need to turn it up to get it conducting again. Brian

Brian Gaff (Sofa) snipped-for-privacy@blueyonder.co.uk> wrote

He f***ed up uk.d-I-y he had uk.diy

leds don’t dim using triacs but still cant be dimmed right down close to zero for use as a nightlight with the philips hues particularly.

OK what do they use then, All I remember from my brief experience trying to dim leds was that the most successful way of doing that was by duty cycle, ie on to off times with them driven by some kind of oscillator with variable mark space ratios. However it is obvious that even the briefest of ons and the longest offs tends to still be visible in most cases, and not terribly accurate if many leds are used as the load, there being a spread of linearity in any given number. Brian

PWM is the cheapest way as it needs no other component.

For a small improvement in efficiency then a constant current source would be best but obviously costs more.

Yes, I made a PWM dimmer circuit (for fun) based on circuits I found on the Internet and it works well, and dims to nothingness. (12VDC working a white LED strip.)

do it at high enough frequency and there is no visible flicker, and you can feed the LEDS via a small ferrite inductor with a parallel capacitor and reduce HF flicker too.

But at that point you've already gone over the peak of the power curve. I'd just make it drop the curve more quickly so it can never go too fast. The driver would still feel the lack of power.

Not possible to do that and still get the best performance.

Makes more sense to have a rev limiter and much easier to do too.

And you f***ed up uk.d-i-y in the newsgroup list, you have uk.diy

Had a rev limiter on my RS1600 with electronic ignition in 1975, it worked well at 6700 rpm.

You're getting f*ck all performance if you've gone over the power curve anyway.

No, the guy that made the group f***ed it up. How did you notice?

I did not have a rev limiter on my Renault Espace built in 1993. I broke the engine badly. I thought they all had them by then!

That depends entirely on how the engine is set up. A turbocharged angine for racing can easily be delivering nearly flat torque all the way to disitegration and will develop full power on the limiters edge

It's a primitive technique for an engine under-endowed with torque. When you change up, there is an inevitable drop in power so the idea of revving beyond peak power is to get further up the curve in the next gear.

Kinsey simply has no idea. In the old days you strangle the breathing so customers wouldn't over rev the engine as it was obvious that you would get more power at lower revs.

But even my MG midgets of 1970s vintage could produce more power beyond the orange and up into the red, by dint of having more carburation

Turbo and supercharging makes it even easier to get good torque at te top of the rev band

These days you cant get 'over the power curve'

It's obvious what I meant, the car warns the driver he's revving too high by making the engine jump on and off.

But if I buy a "dimmable LED" lightbulb, I can dim it with a dimmer that's on the wall, at 240V AC. Is there electronics inside the bulb sensing the PWM from the dimmer and using that to control its internal current limiter?

Please tell the car manufacturers how to do this so I don't have to watch flickering tail lights while driving behind someone, or flashing headlights on TV.