|
Title: Timing oscillations ME7.5 Post by: mitkoenchev on December 04, 2014, 07:06:56 PM Hi everyone, I have some trouble identifying the source of some timing oscillations in my logs. I have it no matter if I modify IOP/IRL/WOP/WOP2 or I use the stock maps, I even have timing oscillations with the stock file. In my latest logs it seems zwout is following zwsol and is capped by zwbas. As I read on some threads here, it seems it is torque intervention, but I can't find a way to fix it. Could it be ARMD?
I attach some logs, but only the last one has zwbas and zwsol. I also don't have mibas, milsolv and mimax defined in me7logger. Do I need to log anything else to help me solve that? Thanks Title: Re: Timing oscillations ME7.5 Post by: mitkoenchev on December 07, 2014, 06:58:51 PM I tried TT225 IRL/IOP/WOP/WOP2 maps and touched the last row of KFDMDARO from 10 to 50. It didn't fix the timing oscillations, and now I even have a requested load drop below 2500rpm :'( The only other thing I did was lower the derivative component (forgot the abbreviation) of the N75 PID below 3000rpm to fix the undershoot I used to get, but I don't think that's the problem for the lower load request. I don't know what went wrong...
I would appreciate any help. Thanks. Title: Re: Timing oscillations ME7.5 Post by: littco on December 08, 2014, 02:19:04 AM I tried TT225 IRL/IOP/WOP/WOP2 maps and touched the last row of KFDMDARO from 10 to 50. It didn't fix the timing oscillations, and now I even have a requested load drop below 2500rpm :'( The only other thing I did was lower the derivative component (forgot the abbreviation) of the N75 PID below 3000rpm to fix the undershoot I used to get, but I don't think that's the problem for the lower load request. I don't know what went wrong... I would appreciate any help. Thanks. Have to say 4 degree swings isn't that bad ! You'll never get a smooth line and best you'll se is 2-3 even with it all turned off, what makes yours look bad is the scale is quite small on the timing access. Mine normally goes -10 to 35ish, so it looks smooth even with 2-3 degrees swings. Title: Re: Timing oscillations ME7.5 Post by: bk56190 on December 08, 2014, 06:38:27 AM Hi,
It's torque intervention, too much load vs requested load; You actual boost is too high vs desired boost; Title: Re: Timing oscillations ME7.5 Post by: ddillenger on December 08, 2014, 07:00:13 AM Hi, It's torque intervention, too much load vs requested load; You actual boost is too high vs desired boost; What do you base this on? Title: Re: Timing oscillations ME7.5 Post by: mitkoenchev on December 08, 2014, 08:09:50 AM Have to say 4 degree swings isn't that bad ! You'll never get a smooth line and best you'll se is 2-3 even with it all turned off, what makes yours look bad is the scale is quite small on the timing access. Mine normally goes -10 to 35ish, so it looks smooth even with 2-3 degrees swings. Thanks, I never looked at that scale-wise... So it leaves just my other issue - req. load drop which I guess comes from the IRL/IOP/WOP/WOP2 maps. Is it better to have the last column of IOP with just a bit lower values than it would perfectly match IRL? Would that make the torque monitor a little more tolerant for slight overboost cases, for example, or should I alter IZUOF/FIL maps to achieve this? Title: Re: Timing oscillations ME7.5 Post by: littco on December 08, 2014, 09:44:43 AM Thanks, I never looked at that scale-wise... So it leaves just my other issue - req. load drop which I guess comes from the IRL/IOP/WOP/WOP2 maps. Is it better to have the last column of IOP with just a bit lower values than it would perfectly match IRL? Would that make the torque monitor a little more tolerant for slight overboost cases, for example, or should I alter IZUOF/FIL maps to achieve this? What's ldrxn? That's likely to be the requested load issue. Title: Re: Timing oscillations ME7.5 Post by: mitkoenchev on December 08, 2014, 12:49:22 PM What's ldrxn? That's likely to be the requested load issue. I haven't touched LDRXN a long time ago. AFAIK specified load follows LDRXN at WOT, but the log shows that requested is way below specified up to 2500rpm. After that it's following specified load. I changed my IOP/IRL maps and there's no problem now. I'm just curious why did that happen and I'll investigate why. Obviously there's something I'm missing in the torque model. Attached are LDRXN (I upped the values at 2250&2500rpm by +2 today) and today's log with fixed IRL/IOP. I'm still "figting" with PID tuning. I just can't get it to react nice at different IATs... |