I have the EV button installed with a scangauge II to tell me if I am in EV mode. The problem I am having is that when I press the button to go into EV mode, it will quickly go to "ON-OFF-ON-OFF". Its is not always in this order, but it will cycle a few times. What's frustrating is that a lot of times I'll press it and it will cycle and end off in the state I began in. I am not sure what is causing this but it seems to be temperature dependent. When I have been driving for 10+ mins, this issue will usually not occur. I don't recall seeing it in the Fall when I installed the button either. Any other ideas what might be causing it? Maybe a bad solder which only creates perfect contact when warmer due to thermal expansion?
tom1|21, Is your EV switch the Coastal one? -- Not that it really matters. Since the cycling you see happens early in the day/drive, it may be related to HSD imposed low HV battery temp EV mode restrictions. It could be that the battery is right on the border between EV mode being OK and not. I've got the Coastal EV and SG too. I've noticed that after an overnight cold soak, sometimes if I try to engage it immediately after the POWER button, it will go into EV mode and then after a second or so drop out. Then EV mode is locked out until the HV battery warms up some. Maybe your situation is similar.
Rokeby thanks for the reply. I installed the push button switch from the instructions provided by efusco online. When the button flickers/cycles, it does so within 0.25 seconds and doesn't beep at anytime denying EV mode. I know what you more specifically are speaking of as I have seen that as well. I think this might be a different issue but the battery temp is something I have considered. Usually I can enter EV mode ~100F with an ambient of ~35F. Sometimes I'll be at 150F and still won't be able to enter which I think is a battery temp issue (6 bars < 34mph). I might try re-soldering but its such a PITA so push the pin into the ECU.
I'm not sure about how you override a US Prius II to have an EV switch but I do know from the 2010 Prius I have that sometimes it will refuse to engage EV mode due to low battery even when the battery has 4 bars left in it - out of 9! Maybe that's what is happening on yours?
It may be just the way the SG is reporting it. If it is not beeping and not actually going in and out of EV, then it is probably just a display issue.
It may be because the switch you used is of inferior quality or its contacts are a bit corroded, producing a lot of contact bounces. Could you post a photo of the switch you used? I don't think it is because of bad soldering, especially not on the pin side. But anyway, could you also post a photo of your soldering job and insulation you used? I would try to change the switch in the first place. The ECU should filter some contact bouncing if designed properly, but your switch may be too much for it to handle. If the problem persists, you may try to install some ceramic capacitor, say 220 or 330nF, over the switch. If this simple solution doesn't work, try to add a simple first-order RC low-pass filter between the switch and the ECU.
The switch I used is the one recommended by efusco in his guide. Its a push button switch from radio shack. I might switch it out for one of the "clicking" switches. Not sure if that will work better since its a clear "on/off" as opposed to the normal switch. The connectors were pretty clean to me. I'll see if I can post some pics in the meantime.
I didn't mean the connectors, but the internals of the switch. The "on/off" switch is not convenient for this purpose, I would try to obtain something more suitable. Do you have or could you borrow an (digital) oscilloscope, able to catch non-periodic signals, so you could check how the voltage transition when you push the switch looks like? This would tell you the story.
The switch I was referring to was not the "lightswitch" type but a type that makes a "click on" and "click off" sound. Regardless, I think I will buy another one of the same switches since its cheaper than buying any instrumentation (don't know of anyone that has one). I am interested though what is going on. When I press the switch, the circuit is completed and broken several times in a fraction of a second. Perhaps the connection isn't flush. This is the only reasonable hypothesis I think.
Well, sounds like a micro-switch. That should be suitable. Exactly. This is a normal feature of all mechanical switches, but some of them might have especially harsh transition. This condition gets even worse when the contacts of the switch are corroded (even a little). It is therefore necessary to somehow filter the unwanted, multiple short transitions. The ECU should be designed for that, but maybe the switch you use is too much for it to handle . So, you could use a better switch, or try to mitigate it's behavior by adding some type of analog filter I was proposing in my first post. Or, the switch is not to be blamed at all and your problem is caused by something else - but contact bouncing seems likely to me. I know it very well from my own circuit design experience .
I am glad it helped . Well, not necessarily a bad lot, but the quality of individual switches, even from the same lot, may differ and still remain within the design limits. Also, the contacts and springs may (and likely will) degrade over the time. If you want to have it bulletproof, add at least a capacitor parallel to the switch, as I said in one of my previous posts. I think the original "Toyota" switch most likely has at least that .
Thanks again for the help. I think I'll keep the switch the way it is for now, unless more problems arise. How exactly would I connect the capacitor and what exactly does it do from an EE standpoint? I'm a civil but am very interested in EE as well.
If done right a resistor and cap would provide a debouncing circuit. Debouncing is required when you have a switch/relay that controls a digital circuit. When ever you push the switch or the relay makes contact the small leads that make contact will bounce. When controlling a digital circuit these very small very fast bounces look like really fast on/off presses. I believe that the Prius has a debounce circuit in the ECU not in the EV button. I could be wrong because I don't remember what the EV button looked like from the inside. Your old switch may have had very lose contacts or very close contacts that when the car vibrates the contacts in the switch make contact. This is the only thing I can think of. If your switch was a Click on click off switch that would be the wrong kind. You need a momentary push button switch. Some of these clicky switches can be converted to a momentary switch by removing one of the parts inside the switch.
The capacitor should be connected parallel to the switch, as shown in attached scheme. I assume that there is a pull-up resistor (denoted R1 in my scheme) in the ECU, which would conveniently form a simple debouncing circuit along with the capacitor C1. The pin interface circuitry in the ECU is surely much more complex than what is shown in my simplified scheme, there is surely some kind of protection against shorts/voltage spikes etc., but a pull-up is definitely present. In the steady state, when the switch is released, the capacitor is charged to the Vcc. When the switch is pressed, the capacitor is shorted and quickly discharged through the switch. When the switch is released again, the cap would charge through the pull-up resistor R1, filtering the bounces (now the R1 and C1 form a first-order low-pass RC filter). The time constant will be ~3ms (assuming the R1 is ~10K and C1 is 330nF), which is relatively short time but should be enough to filter the rebounds. I agree with The Force and also assume that there is some kind of debouncing protection already implemented in the ECU (most likely in the firmware, not in the circuitry), but evidently some switches are bad enough to overcome it .
I agree, but a cap parallel to the switch cannot do any harm . I think there might be a tiny SMD cap in the original switch, but I may be wrong - I am just guessing. I think that by "click on/off" switch Tom meant a micro-switch, not an arrested switch (see our conversation above). Like the one you have in your mouse (most likely) - clicking when pressed, but not arrested in the "on" position. That would be a good switch. But he didn't use it, was just considering using it if I understood correctly.
Regarding to the battery temperature, you can activate the EV switch on between 0C(32F) and 45C(113F). Ken@Japan
Yes, that is correct, good thinking. I tried one of these switches, but it again had the bounce. I went back to theoriginal one, but red! Looks better and no bounce! Thanks Ken, I appreciate it.
Rat Shack switches are perennial garbage. If you can find something decent from C&K or Cherry you'll be doing much better. A "clicking" momentary gives nice tactile feedback, and helps the thing to not bounce even after it's been in the car for 4 years. . _H*
Thanks for the heads up hobbit. I'll give those places a shot if I have more issues. I was wondering, is using the EV switch to glide when below S4 going to give any gains in FE? I use it for this among other things such as moving around parking lots etc. I seem to notice better FE in the first 5 mins but have read from others that there is no difference and it can even lower FE this way.
