Page 5 of 7

Re: Power control of the EMS system

Posted: Sat Aug 16, 2008 9:13 am
by Fred
Thanks Jared, indeed I was posting in the wrong place and entirely my own fault :-)

You are right about the math, but I don't know how many AH you or bob or xyz has so a rough calc is good enough in this case :-) I typically use a battery around twice that big.

Fred.

Re: Power control of the EMS system

Posted: Sat Aug 16, 2008 12:32 pm
by jharvey
I should find out what a normal car battery does have for A/hr rating. I picked 150A/hr because it's a common gell cell we use at work.

I'm looking at the power regulator 1R resistor. The purpose for that part is current limiting, and we don't expect to every really need it, it's a fault protection. Perhaps we should have that as a fuse?

I ask because I'm trying to figure out footprints, that one probably won't be the normal R4-SM0805 footprint, or perhaps it should use the fuse footprint. I'm can't really focus on it now, I'm assinging footprints to other components, I should have resistors and diodes done shortly.

Re: Power control of the EMS system

Posted: Sat Aug 16, 2008 8:13 pm
by Fred
jharvey wrote:I'm looking at the power regulator 1R resistor. The purpose for that part is current limiting
Actually it *was* dual purpose, now it's just single purpose and that purpose is in conjunction with some significant capacitors on both sides of the reg to form a first order low pass filter and provide very clean power to the CPU and sensors for each respective circuit.

I had written this post here, but decided it was best in the other thread :

http://www.diyefi.org/forum/viewtopic.p ... 4153#p4153

Fred.

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 10:03 am
by Fred
I had an idea in the shower last night. Instead of having a separate clean feed and reg circuit for those things we want off during key off, we could use the dirty feed to trip a relay that splits those things away from the CPU feed and have one setup. That brings the complexity in board and joins sensor faults to the CPU feed again, but it could be worth considering. I think I like the dual setup better as it is more simple and clean. This way just saves one external wire, fuse and relay being setup. Probably not a good idea, but I thought I'd post it up anyway.

Fred.

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 10:40 am
by jharvey
I can't see the tree from the forest here. You would still need to route the key signal to the CPU wouldn't you. So it would still have similar wiring complexity. I kind of feel like if you split the feeds at the battery rather then the ECU, you'll get a much cleaner supply.

Perhaps you're seeing a different way of connecting the leg bone to the head bone?

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 10:52 am
by Fred
Sorry, currently the plan is :

1 clean supply for cpu that is always on
1 clean supply for sensors and 12v ref that is relay switched outside the box by the key
1 dirty supply for flyback and high side stuff that is relay switched outside the box by the key

I was suggesting to cut that two this :

1 clean supply for cpu that is always on that feeds sensors and 12v ref that is relay switched INside the box by the dirty supply
1 dirty supply for flyback and high side stuff that is relay switched outside the box by the key

Is that more clear?

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 10:54 am
by Brian
One of the three 12V feeds, the one that gives battery voltage info to the cpu, does it need to ever be turned off?

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 11:19 am
by Fred
That depends on the size of your batteries. If the accessories aren't being powered down at key off you have a divider network for the voltage reference drawing current, the temp sensors and map sensor drawing current, and any other accessory ICs drawiing current, none of which need it. I'd prefer an absolutely minimal subset of devices were left powered up at key off.

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 11:24 am
by jharvey
Ahhh, I see the tree now. I like it.

Does it need to be relayed, or could we get by with a FET. I guess a NC relay would be handy for high side driving allowing the default signals to float to ground. This would help prevent fuse blowing, but relays are so noisy. Perhaps we can use a fet as a high side driver? I'll have to stew that over a bit. Right now I'm still focusing on getting foot prints for components, so I'll have to stew that over later.

Re: Power control of the EMS system

Posted: Sun Aug 17, 2008 11:41 am
by Fred
I thought about a high side FET too. The relay would only be noisy at switch time, but it would draw power the whole time, but thats a non issue. If you go relay, it has to be a common footprint. Also, they are quite bulky (even the small ones). Actually, there is another snag too. The relay would have to be double pole to switch raw clean 12v to the div network for reference and 5v to the sensors at the same time.

It was just an idea. I'm not convinced I like it (even though I thought of it) but I think it's worth discussing..

Fred.