PitBull

I got some time with the car yesterday and installed firmware one release back from the current and the problems seem to have gone away. Thank you for your assistance with this issue.

I haven't yet got my car out of storage to try this, but I was wondering -- do I need to roll back to 5.5.7 (two releases back)? Or do you think rolling back to 5.6.4 would be sufficient (one release back)?

Ok, thanks for the info. Apparently winter isn't over yet here in the Northeast US, so the car is parked again for a while and I haven't had a chance to try your suggestion yet. Once I do I'll let you know the result.

You've got mail

I recently upgraded the firmware in my Link Storm (old style) and I am experiencing two issues since then that I think are fuel-related. I didn't take note of the version it was at before but I think it was from mid-2016. So the first problem is with starting. Prior to the update, the car started fine under any condition hot or cold engine, hot or cold weather. After the update, the car starts ok when the weather is cold and engine is cold. But if the weather is warm or if the weather is cold but the engine is warm, it will stall about 1-2 seconds after firing. If I turn the key off and try again, the same thing will happen. If I keep turning the key off and then trying to start, the result is the same. However, if after the first try and stall, I leave the key ON and then crank it again to start, it usually starts and runs. The second issue has to do with the Dual fuel table. I have the Dual Fuel table mode set to "Overlay". This table is active when DI1 is ON. However, when I turn on DI1, the engine stalls almost immediately and from what I can tell I think it's because the fuel is cut way down. This table worked fine before the firmware update. As an experiment, I put all zeros in Table 2, which should effectively do nothing since it's an overlay table. However, even with zeros, the engine still stalls. As a final test, I disabled Table 2, and with it disabled, the engine does not stall when DI1 is ON. I'm not sure what steps to take next to try and troubleshoot. Can the firmware be downgraded or is that a bad idea?

Ok, I understand better now about the difference between using GP output and injection. I'm pretty certain of the answer already (no), but can an Aux output be utilized in group/staged injection?

Thank you for clearing that up for me

With CTC turned off, is there some form of temperature compensation going on in the fuel equation and the IAT trim and Warmup enrichment are tweaking that? Or do those tables represent the only compensation when CTC is off? So if they are zero, there is essentially no compensation for charge temperature going on?

Okay that's good information to have. My Storm only has 4 injector drivers so I have to use a GP output. If I upgrade someday I'd definitely switch it over to use an injector driver...

Yes, I have an Aquamist HSF4 system which varies the methanol flow using a "fast acting" solenoid valve and constant pump pressure. But it gets fed a PWM signal from an Aux output on Link ECU, rather than tapping into one of the fuel injectors. This allows me to control methanol flow independent of the fuel flow. I have the Link set up so if it hits MAP limit, it turns off the PWM signal to the Aquamist, which causes it to quit injecting pretty quickly. Much faster than if it was just relying on variable pump pressure to control flow.

Okay, I understand that. Can you clarify for me -- is CTC a better way for the Link to have a Charge Temperature value for it's fueling equation? As in, if it's turned off, it uses another less-accurate method (i.e. the IAT value)? Or if CTC is turned off, is there no charge temp correction going on except for what's in the Warmup Enrichment and IAT Trim tables?

The nozzles are Aquamist water/meth nozzles, but it has a fast-acting valve just before the nozzles that cuts the flow off very quickly (though not as quickly as a fuel injector of course). So it would mainly be the wetting of the intake pipe that would be providing the methanol once the cut hit.

I was doing some initial experimentation with the Charge Temp Correction (CTC) feature on my (Blue) Storm. First, when I turned CTC "on", I noticed there were values in the Charge Temp Approximation Table which looked like a sensible starting point. Then, on a graph, I logged the ECT, IAT, and Charge Temp. However, what I noticed is that the logged Charge Temp with CTC "on" was the same as with it "off". Turning it on or off made no difference in how the Charge Temp value responded to RPM and MGP. I thought maybe with CTC turned off, it used some kind of default response. So with CTC turned on, I modified some values in the Approximation table. As expected, changing these values caused the Charge Temp to change as well. However, when I turned CTC "off" the Charge Temp response stayed the same as if it were on. So what I am experiencing is that it seems like CTC is always active, whether it's turned on or off. But when it's on, I can edit the approximation table. when it's off, the table is blank. Am I misinterpreting what I'm seeing, or is there something wrong?

Hello, sorry I didn't get back to this until now... Adam: What I meant is that if I use fuel cut, the fuel injectors will shut off and the gasoline will be cut nearly instantaneously. However, since the methanol nozzles are probably about 2 feet before the throttle body, even shutting off the nozzles upon hitting MAP limit will leave a certain amount of methanol in the intake tract. I don't know if that amount would be enough to cause a lean condition and possibly detonation or preignition. I do realize that if it is an extremely lean mixture, it would just misfire, rather than cause detonation. At maximum injection, the methanol is accounting for approximately 30% of the total fuel, so if the gas were shut off but that amount of methanol remained, perhaps that would not be enough to cause any issues? The more I think about it, maybe it would still be better for me to use fuel cut, rather than ignition cut. I definitely won't try the Rotary setting after your warning, Scott.

I'm running methanol injection on my engine so I currently have the MAP limit mode set to "Ignition" since I figure if I use fuel cut, with the methanol keeps flowing the engine may experience a dangerously lean condition. However, I have on occasion hit the MAP limit in 1st or 2nd gear and in one instance it resulted in a very loud bang out the exhaust, presumably because the unburned fuel ignited somewhere in the exhaust. It was loud enough that it worries me that it could damage something so I'd like to prevent it from happening again. I notice the MAP limit mode "Rotary" says that it limits fuel and ignition. Is there any reason this mode shouldn't be used with a piston engine? It seems that cutting ignition AND fuel would be best for my situation. I am using sequential fuel injection mode.

Odd...my help file reads slightly differently than yours CamB. Brad, yes my WG DC tables are identical too, which is why I was hoping this feature interpolates between the target tables. I have RPM on my X axis and boost target on Y.

I want to interpolate between two boost tables. The help file states: "When set to Interpolate Between the Boost Table Ratio Table appears. The Boost Table Ratio Table is used to control the interpolation between Wastegate % DC Table 1 and Wastegate % DC Table 2." Does it also interpolate between the Boost Target tables 1 and 2? It seems like that would be the desired effect (and is what I'm looking for) but the help file doesn't state that explicitly so I wanted to be sure. Thanks PB

I'm pretty sure I had power-cycled the car/ECU two or three times between turning on those features and experiencing this phenomenon. I'd also add that both the ECU downloaded log and the PC datalog showed the same missing knock data. I'll post if it does it again. So far it's been fine... Thanks

I had a weird thing happen two days ago. When I got home from work and took a look at my logs, there was very little signal registered from the knock sensor. As in, the values from all 4 cylinders were mostly 0 with occasional blips of maybe 10. The entire log shows this from when the engine was first started until I got home -- about a 25 minute drive. So it's not like it happened in the middle or was intermittent. I thought maybe my knock sensor died, but I have since driven the car numerous times and logged the ECU and the knock signal has been normal. It makes wonder if it was a firmware glitch and rebooting the ECU cleared it up. I've been running the Link Storm for almost a year now and I do PC data logging just about every time I drive the car and I've never seen this happen before. Recently I did turn on ECU logging for the first time (about a day before this happened), though I still have it turned on now and everything seems fine. I also started using two Virtual Aux outputs and a Timer about two days before this happened. Those are the only recent functional changes I've made, though I make small tweaks here and there regularly. Thanks PB

Thank you for the update.

Hello, any info on the status of this? Thanks

The Aquamist failsafe output (at least the one I have: HFS4 v3) can be set to either go to ground or go high (+5v) on activation. The characteristic is selected by a jumper setting on the Aquamist control module. There are several conditions that activate the failsafe: 1. Low fluid detection 2. Flow sensor output is below lower threshold (i.e. pump failure, nozzle clog, meth line break prior to flow sensor, FAV fails closed) 3. Flow sensor output is above upper threshold (meth line break after flow sensor, FAV fails open) 4. Injection system is deactivated by the switch on the flow gauge If you set the failsafe to go to ground on activation, you could then add a toggle switch onto that wire to short it to ground when you want to trigger the failsafe maps manually. BUT, you'll want your switch to connect ONLY the Link's DI to ground, while at the same time breaking the connection to the Aquamist failsafe output. Shorting the AqM failsafe to ground would be a bad idea. I have my Aquamist failsafe connected to a DI on my Link and I have it set to switch between high and low Boost Solenoid PWM DC tables. I do not use it to switch fuel or timing maps. Rather, I use 4D fuel and ignition tables and have the Aquamist's flow meter output set as one of the axis and RPM as the other. So as the AqM's flow meter output rises (i.e. more meth being injected), the 4D tables remove fuel and add timing. I am still in the process of tuning this but it seems to be working fairly well. But I am running 100% methanol only and don't need to switch between 100 or 50/50 like you want to. You'd need to use fuel and ignition map switching for your purposes. The one issue I am having to work around is that the AFRs tend to get progressively richer as the meth sprays, even when other variables stay constant (RPM/load/meth duty cycle). My assumption is that since I have the nozzles just after the intercooler exit and there's about 2~2.5 feet of intake piping between the nozzles and the throttle body, some amount of the meth will wet the intake piping, but once it is wet, then the majority of it flows into the cylinders. I don't see it as a big problem, however, as the reading I've done on methanol says you can run it richer than gas. If I can get the AFRs to start out around 11.5 and drop into the 10's during the pull I think it'll be good.

Does logging every possible parameter reduce the # of samples logged, in a given time period, for each parameter? i.e. can it saturate the bandwidth and start dropping data points if I log too many parameters?

It'd be really handy if PC link would remember the last directory you loaded or saved a file into. For each individual file type. So if I load tunes from a network share and logs from a thumbdrive it'd remember that, rather than always starting out in the PCLink install directory. Thank you PB

Hello... Did this get fixed? I tried switching to a 3D MAP limit and it appears the MAP limit value in the cells are still displayed in kPa even when Units are set to Imperial Thank you