IcemanG17

worf928

Interesting. My 'bet' above was based upon a freshly-rebuilt MAS, known-good o2 sensor and no intake leaks.

It has been a very long time since I played-around with this. I have no stock 928 that isn't in pieces at the moment.

Most interesting. I've cleaned and bench-tested dozens of these and never noticed that. I have a couple lying around (because all the stock 928s in the shop right now are disassembled) that I can test. I need to see this for myself. They have to be cleaned anyway.

... which is more-or-less at 50% at idle. Thus, assuming linear operation, the shuttle would be about 1/2 through its travel.

As long as you understand that it is a pulse-width modulated signal from the LH. The ISV doesn't get a DC signal, as far as I know, so, the shuttle is never at the full-open stop. The PWM signal 'holds' the aperture open at a specific rotation.

Again the ISV's control algorithm is very stupid. You can verify this by creating an intake leak and watch the idle RPM rise. The ISV's logic is mostly about keeping the RPM 'up.'

The PWM signal (https://en.wikipedia.org/wiki/Pulse-width_modulation) is generated by switching the ground to the ISV. It always gets 12v. Thus 'more voltage' is a function of the RMS voltage of the PWM signal. (An average-over-time voltage, if you will...)

FredR

Dave,

Excellent posts! The empirical test of dumping voltage on the thing only drives the shuttle fully open but [most probably] proves the unit is viable.

The PWM provides the control element and thus [in theory at least] a mechanism to control rpm's above the set point. As we know, a vac leak will cause the motor to race but it is an uncontrolled event so the system cannot be expected to cope with much of a leak given it is false air with no fuel.

In the case of my currently installed ISV I am seeing a surging idle and my perception is that for whatever reason the shuttle is now a bit sticky so it has to apply an over correction to get it to move and then counter correct when it over reacts thus the cycling of the rpm's. My ISV more or less stopped operating some 17 years ago so did my first inlet manifold pull to find the shuttle full of a greyish dry powder. Cleaned it out, sprayed a splash of WD40 into the port and the unit is still there but probably past its sell by date.

the flyin' scotsman

given the idle speed needs controlling under many variables the PWM signal increases or decreases the ISV operation until the TPS is off idle?

Im curious what the PMW signal and ISV operation are when off idle.

great thread Fred/Dave

FredR

Malcolm,

My perception is that once the TPS idle switch senses the throttle being opened the ISV voltage is dumped [no volts] and the shuttle thus returns to the default spring return position until such time as the idle switch contact triggers and the thing duly springs back to life. I suspect the earlier ISV's do much the same thing just that they do it in a different [inferior] way. That it takes an oscilloscope to monitor what is going on is no surprise and presumably if one wants to simulate operate one needs some kind of signal generator.

At least I now feel as though I better understand how the ISV works. Now I have to decide whether to fit my spare item and see if it works or just bite the bullet and get new one. At least I have the benefit of knowing that I can get the inlet manifold off and back on again in 4 hours or so having done it quite number of times. I have a new pair of knocks sensors to go in that have been gathering dust for some time - the old ones still work somehow but the connectors partially fell apart so I had to jury rig the connections. All part of my annual worklist for this year that I will be starting shortly.

TheoJ

yes, the ISV manages the idle rpm using a duty cycle modulated square wave. The ISV has a small opening when not powered (possibly meant to keep the engine running even when the ISV is not operated). And it opens fully when put on well over 50% duty cycle. Testing can be done easily on the LH connector, or on the ISV itself (requires intake off). You can also do some basic testing over the diag bus using the actuator routines of the LH.
This is what it should look like: http://jenniskens.livedsl.nl/Technic.../MyTip1150.htm (activate the flash movie)
Often the ISV is sticky of has sticky spots due to dirt. It helps a lot to clean them properly. This should really be done outside the car. Spraying WD40 in the hoses may help if you are lucky.
regards
Theo Jenniskens. 1992 928 GTS

Captain_Slow

Nice to see you here again, Theo. I have the intake off and will test and clean the ISV today.

FredR

Finished the inlet manifold work day before yesterday. The new ISV seems to be working well- the idle is much more stable, not perfect but quite acceptable and way better than it was with the old ISV. I have a feeling my breather mods might be recycling more gases than it is comfortable with and need to test that theory by isolating the breather return pipes which is easy to do with my setup.

I need to take a look at the item I removed to see whether there are signs of accumulated crud.in there.

javiherdepa

As I can get to the isv to spray with wd40, the suction occurs in the admission with what I think this will go to the body of the intake butterfly and not to the isv, thank you very much