MS2 vs MS3X idle duty
 

Since switching from MS2 to MS3X, I had to raise my duty cycle from 32ish to 70ish for a stable idle. Needless to say this is not the preferred range of the valve.
Anyone else notice this? I haven't fiddled with the idle screw yet, so it might just be a matter of adjusting, but still, this is a big difference.

using the same freq. for the valve?

Same duty cycle here, although the idle valve buzzed a little more on the MS3 than on the enhanced MS2.

I've got my valve set to 153HZ I had to raise my idle open duty from 63 to 90 in order for it to hit the target warmup rpm after I changed from the diypnp to the MS3x. Looking closer I think it may be because of my cranking duty table. Duty% when cold is 84.4. With the open duty of 63% closed loop wasn't allowing the idle valve to open enough.

Well, the freq has changed. Before I was using Jean's PWM board (can't remember the exact freq, 480ish Hz), now I have it at 390Hz. Changed to 780Hz and had the same high duty so I don't think it's that freq dependent. Probably more because the x is using a different transistor. I wouldn't expect that big a difference though.
Weird that you have the same duty cycle Dimitris. Is this with the x or plain ms3 (iow, tip120)?
The valve likes to operate in the 30-35 region so I'll see if I can get it down with the idle screw.

djp0623, 90% seems very high, I'd try lowering it as well.

Just to make sure we are on same page. Show me a screen cap of idle duty you are talking about.

and flyback diode, yes.

When I went from MS2 to MS3x, I continued to se the same TIP on my mainboard, not the MS3x. My scale for my 1.6L valve is between 25-60%.

Mine is 67-74% from fully closed to fully open, clearly not enough resolution.
djp, I'm sure were talking about the same duty. The duty needed to idle. The valve is most lineair in the 30-35% region (and it's were the oem ecu has it).

I just ran the idle valve test and adjusted each direction until the rpm stopped responding. Then added a little more in each direction. 67-74 seems like a very tight window. I may run it again to see. I think my min duty is 18 (response seemed to stop around 25) and my mas duty is set to 90 ( I believe response stopped around 83)

Brain: that matches closely with what I found on my 1.6 with my diypnp. Is it worth it to add a little buffer around this number? What are you seeing on your ms3x for the 1.6 valve min and max duty values? By closing the delta between these values, isn't that going to inhibit how much room the Closed loop idle has to work???

i typical hover around 29-30% at idle, 42% with a/c, and high 40s on a cold morning. never in the 50s-60s, that would drive it up to 2-3k

I couldn't get the closed loop idle to hit target until I raised the max duty. I just raised it until CL idles was able to hit and maintain target. At 63% I could barely get the car to idle above 700rpm.

https://www.miataturbo.net/picture.p...pictureid=1982

(Sent from phone via capitalism)

Just use whatever works. Your idle valve is running off the msex. The min and max should still be a good wide range...its possible to alter the flyyback on the ms3x...ask yates about it will pprobably drive down the duty.

I've got a 99 Throttle body on my car, and running ms3x at 390 or 780 have the same issues as you... It seems to act ok but its almost always around the 70% mark, which traditionally the MS manuals have said is bad. I just kinda assumed it was the idle valve, nice to know that i'm not alone at any rate.

Tried upping the timing at idle?

Dann

the value of the zener in the xpander flyback circuit affects duty. i went from a 32v to a 22v and it lowered the duty values a bunch. the schematic does say 36 but it aint.

i'm not sure if my 12V wire also impacts this but it might.

So is the general recommendation for anyone building an MS3X is to go ahead and build the normal idle circuit and use that rather the MS3X idle?

why? ms3x idle is fine.

Just questioned after reading through this thread and seeing that by using the standard MS3X idle you get a fairly low resolution. You mentioned replacing the zener in the flyback circuit (D13) with another diode, did you just put a 1N4748A in its place?

you dont need to replace it. I only did as an experiment but! if you plan to run VVT, the amount of flyback will toast the flyback circuit and you'll have to add a wire from the cathode of the ms3x flyback diodes to the S12 pad on the main board.

resolution is about 1/3% and it works well even with poor PID tuning.

oeps, wrong thread