DFH - Defacto FreeEMS Hardware in KICAD

Jared's unmaintained and never-used TA based "Defacto FreeEMS Hardware" design.
User avatar
AbeFM
Post Whore!
Posts: 629
Joined: Sat Feb 16, 2008 12:11 am
Location: Sunny San Diego
Contact:

Re: freeEMS_1.0 rev A KICAD

Post by AbeFM »

Jean does a lot of nice stuff, I borrowed (on his suggestion) just idea of a jumper block, and found it to be more flexible even than I got from his initial advice. It both allows tinkering with signals as well as simple routing.

The more I look into it, the more I like the stacked approach - sure, you could spin a single board for almost any application, but, this will allow you to easily make any fully isolated board, really handles the modular approach (I would probably have a peak-and-hold card as something in my first build - I'm not driving/testing without it, it interests me on it's own, and I see it as being crucial to the widespread acceptance of the project). I like the idea of making cards without it being a total hack - the more MS "develops" the more of a joke it appears to me. The code is as hacked and tacked together as the hardware, and I'm embarased to see a board which sits in place of a board which is meant to simulate some old chip that no one would ever use and jumper cables here and there.... Make a P&H board on the bus and you know everything is good. And if you really have a power board and a non-power board, I think you could have one power supply for all the "controls" stuff, and only resort to a second/clean supply for the big things.

I already own Jean's board, and I'm not that hot to use it. I would if I had to, but ... I'd want to at minimum do more digging into some of the existing hardware solutions. More and more smart IC's are out there for automotive applications.

Then again, I'm building 7 card stacks for a living. :-P We might have to talk connectors later, but for this pass, it's a slam dunk.

Oh yeah - the windows and linux versions keep stuff in different places. It's pretty gosh darned annoying.
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: freeEMS_1.0 rev A KICAD

Post by jharvey »

nitrousnrg wrote:Should we add a jumper to get access to the ind_pd_* signals?
Yes the current plan is to do that, but not at INDuctive Power Driver net, that would be done at RP1 instead.
User avatar
nitrousnrg
LQFP144 - On Top Of The Game
Posts: 468
Joined: Tue Jun 24, 2008 5:31 pm

Re: freeEMS_1.0 rev A KICAD

Post by nitrousnrg »

Yes the current plan is to do that, but not at INDuctive Power Driver net, that would be done at RP1 instead.
Okay.

Another thing: in RPM_input*.sch, I don't see the output of the LM1815 (pin 10) connected to anything but the pulldown resistor.
Marcos
User avatar
nitrousnrg
LQFP144 - On Top Of The Game
Posts: 468
Joined: Tue Jun 24, 2008 5:31 pm

Re: freeEMS_1.0 rev A KICAD

Post by nitrousnrg »

Nevermind, that isn't the output pin :-)
Marcos
User avatar
nitrousnrg
LQFP144 - On Top Of The Game
Posts: 468
Joined: Tue Jun 24, 2008 5:31 pm

Re: freeEMS_1.0 rev A KICAD

Post by nitrousnrg »

Hi Jared,
you and Fred were wondering how to protect the AN pins against -0.2v. Did you consider using a schottky diode? Germanium diodes have 0.3v of forward voltage drop, and the schottkys have 0.15 or 0.2v.

In the microcontrollers i've used, the ratings were -0.6 and vcc+0.6, wich is usually a typical juncture voltage drop. The Freescale shouldn't be different, I think they use a schottky instead of a normal diode inside the micro.

This is just a random schottky, check the voltage vs current:
http://search.digikey.com/scripts/DkSea ... 0520LCT-ND
Marcos
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: freeEMS_1.0 rev A KICAD

Post by jharvey »

I'm not familiar with a Schottky that conducts like that. For example the BAS70 found at this link http://datasheet.octopart.com/BAS70-07% ... 118513.pdf notes at 1ma and room temp it conducts at about .3 volts. That increase with temperature and current to about .7v and on a real big spike .8v volts. So .3v is about as low as I would expect in the best situation.

Do you have a specific diode you might recommend? I've seen some funky TVS diodes, perhaps there is one there that will conduct at a lower voltage.
User avatar
nitrousnrg
LQFP144 - On Top Of The Game
Posts: 468
Joined: Tue Jun 24, 2008 5:31 pm

Re: freeEMS_1.0 rev A KICAD

Post by nitrousnrg »

I should ask a professor, he knows a lot about this. I just remembered what he said, found a schottky, and looked at the tables.

This diodes have a large reverse current, but at 25°C its Vfwd for 1mA is 0.15v and gets better with higher temperatures. In order to get more than .3v you need to be in really low temperatures and 100mA+ of fwd current.

You won't get much more than 1mA if there is a 1k resistor between the diodes and the sensor. Supposing an huge peak, able to generate 10mA, you'll have 0.18v @25°C. (I'm just looking this http://www.fairchildsemi.com/ds/MB/MBR0520L.pdf)
Marcos
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: freeEMS_1.0 rev A KICAD

Post by jharvey »

Sold to the man in the red had (any chance your wearing a red hat?)

I just added that part number to my todo list.
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: freeEMS_1.0 rev A KICAD

Post by jharvey »

Any chance you know of an equivalent with thru hole mounting? One of the goals for 1.0 was to make it friendly for the blind who don't like SMT or are scared of SMT. I've drawn this up with hybrid foot prints, such that you can do either smt or thru hole.
User avatar
nitrousnrg
LQFP144 - On Top Of The Game
Posts: 468
Joined: Tue Jun 24, 2008 5:31 pm

Re: freeEMS_1.0 rev A KICAD

Post by nitrousnrg »

nope, never tried a red hat.

To be fair, the diode I shown you was the first that came up in my search. I was expecting that every schottky should be that good. Well, I was wrong.

I found this:
http://www.vishay.com/doc?93419
http://www.futureelectronics.com/en/tec ... -E3-1.aspx

In digikey it cost $3 each, huge difference.
Marcos
Post Reply