Ok, I finally cracked the lid on this KiCAD thing, and have a couple questions.
So far I made this:
Edit: This picture is more correct, there is an error in the above version:
Does it KNOW what size resistors those are? I put those numbers in "value" but is 3k3 mean 3,300 ohm to KiCAD?
What's a smart way to give these parts numbers? On the OEM circuit, it was nice, since there were two of these, every part corresponded. I.E. if the cap was C101, then on the other circuit it was C121. If the pullup resistor was R108 then it was R128 on the next, which I really like.
Anyway, just want to know what do to next. Guess I'll try layout next but not sure where to go.
If it looks good, we coule build it into the new boards.
Op-Amp buffered "hall" type sensor inputs
Op-Amp buffered "hall" type sensor inputs
Last edited by AbeFM on Mon Aug 25, 2008 1:20 am, edited 1 time in total.
Re: Op-Amp buffered "hall" type sensor inputs
Can you explain how it works/what it does, or is there a thread already about that? If there is, please link it here :-)
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Re: Op-Amp buffered "hall" type sensor inputs
Ok, I'll try to dig up some theory. Right now I'm just trying to get on top of the mechanics of KiCAD here.
Basically, though, the opperating pricible depends on the functioning of an op-amp. If the inverting input is more negative than the non-inverting, the output will go positive.
The output of this chip is put through a voltage divider so the "zero" point you're looking for actually moves dependant on the output.
The trick is that the response of the chip is not infinately fast. You get a small period of time (nanoseconds, I believe) which is constant, but where the output is fighting the input. Any noise spike that doesn't last longer than one gate-delay won't make it out the output enough to swing the output, and is entirely and digitally removed.
Not like a cap where it depends on rate, stregnth of the signal, etc, it's only a question of "is the time spent over the zero crossing point greater than X". If so it swings, if not, it doesn't. It's really very elegant.
I forget the numbers, I worked them out once and can likely find them, but it's something like if you're in the "high" state, you need to drop below ~1.7V for an entire gate delay, and if you're in a "low" state you need to stay above 3.8V for the same time. So you're 'doubley resistant".
(Unrelated, does anyone know where to get a dual row header/jumper block in KiCAD?)
Basically, though, the opperating pricible depends on the functioning of an op-amp. If the inverting input is more negative than the non-inverting, the output will go positive.
The output of this chip is put through a voltage divider so the "zero" point you're looking for actually moves dependant on the output.
The trick is that the response of the chip is not infinately fast. You get a small period of time (nanoseconds, I believe) which is constant, but where the output is fighting the input. Any noise spike that doesn't last longer than one gate-delay won't make it out the output enough to swing the output, and is entirely and digitally removed.
Not like a cap where it depends on rate, stregnth of the signal, etc, it's only a question of "is the time spent over the zero crossing point greater than X". If so it swings, if not, it doesn't. It's really very elegant.
I forget the numbers, I worked them out once and can likely find them, but it's something like if you're in the "high" state, you need to drop below ~1.7V for an entire gate delay, and if you're in a "low" state you need to stay above 3.8V for the same time. So you're 'doubley resistant".
(Unrelated, does anyone know where to get a dual row header/jumper block in KiCAD?)
Re: Op-Amp buffered "hall" type sensor inputs
Ooops, no wonder it doens't make sense. BRB, need to fix something.
Re: Op-Amp buffered "hall" type sensor inputs
Could just use a voltage follower configuration
http://en.wikipedia.org/wiki/Voltage_follower
About KICAD, you select the reference designators manually, or KICAD can place them automatically. KICAD can replace the "Ref" field from say R? to something like R128, but starts at 1 and moves up sequentially. You can't really link one side of circuit to another. Or at least I don't know how to do it.
Try the help file. KICAD is about the only help file I've found helpful. Check into schematic annotation.
If you get into the freeEMS 1.0 files, I can offer some tips that might make it much easier to work with.
http://en.wikipedia.org/wiki/Voltage_follower
About KICAD, you select the reference designators manually, or KICAD can place them automatically. KICAD can replace the "Ref" field from say R? to something like R128, but starts at 1 and moves up sequentially. You can't really link one side of circuit to another. Or at least I don't know how to do it.
Try the help file. KICAD is about the only help file I've found helpful. Check into schematic annotation.
If you get into the freeEMS 1.0 files, I can offer some tips that might make it much easier to work with.
Re: Op-Amp buffered "hall" type sensor inputs
I looked for a bit at the schematic you posted, and it made a lot more sense after a short time (learning how to open and close sheets).
Well, I named them by hand and it worked fine. I got a netlist, and now a layout. Only I don't know where to get a TL-082C (the dual op-amp package).
The motivation for using this circuit (and all the values) was seeing it in the OEM ecu.
Well, I named them by hand and it worked fine. I got a netlist, and now a layout. Only I don't know where to get a TL-082C (the dual op-amp package).
The motivation for using this circuit (and all the values) was seeing it in the OEM ecu.
Re: Op-Amp buffered "hall" type sensor inputs
I see you added labels to your second schematic, the bottom left anchor point, if attached to a wire will pull through the net list. General descriptions should use the text tool.
Oh also 3.3k doesn't really do much for KICAD simply follows through to the layout. SPICE has to be done with another software package at this point.
Sounds like your learning a lot by playing with it. Just keep playing and you'll catch most of it. I'm more then willing to help if something doesn't jump right out to you.
Also are you doing this on a linux / POSIX enviroment, or straight up windows?
Oh also 3.3k doesn't really do much for KICAD simply follows through to the layout. SPICE has to be done with another software package at this point.
Sounds like your learning a lot by playing with it. Just keep playing and you'll catch most of it. I'm more then willing to help if something doesn't jump right out to you.
Also are you doing this on a linux / POSIX enviroment, or straight up windows?
Re: Op-Amp buffered "hall" type sensor inputs
Ok, I was able to struggle through a single sided board.
It could be reoptimized, it would be nice to have my in vias and out vias by each other, in pairs, or taking advantage of a dual layer I'm sure it could be made smaller. And there's basically no labels on this.
So, if I wanted to try to put these circuits up as something you could use, how would I do it? I would think I'd want to design "one" and then you would build it in twice? I was thinking I would add it to your block with the LM18xx chip, but I guess it could be stand alone?
Edit:
OMG! How awesome! Who knew a bunch of free software could go from idea to 3-D model in half a day, without even knowing how to use it?
Re: Op-Amp buffered "hall" type sensor inputs
Free software kicks arse Abe! You will come to know this once I get a bit further in this firmware writing game :-)
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Re: Op-Amp buffered "hall" type sensor inputs
Don't get me wrong, it needs a lot of polish - many things you have to 'know' instead of having a nice interface. But still, I prefer free-with-problems to expensive-with-problems.
How do I get the attention of the main board design? I want to see this incorporated.
How do I get the attention of the main board design? I want to see this incorporated.