DFH - Defacto FreeEMS Hardware in KICAD

[SleepyKeys]

Is that a real issue that some board manufacturers will try to pass you some 1oz copper boards when you ordered 2oz? Or would that be to check a batch from some other uncertain source?

If it's the former then I'd like to know which manufacturers are suspect so I make sure to avoid them. I would think that any respectable board manufacturer could be trusted or they wouldn't stay in business for very long.

Jean

I have leared to basically trust NO one, thats why I really like the "burn" test

[Fred]

The "leave an area blank and another copper clad and micrometer one board from the batch to confirm relative thickness before posting them out or sending back to get replacements" method might be more appropriate, don't ya think?

[jharvey]

Kind of expensive isn't it. An entire board for something that would take up a an area smaller than your thumb nail.

There is a good chance it could also be included in an existing trace, perhaps the text etched freeems 1.0 text could include it. I don't see any harm in adding it.

[Fred]

It shouldn't cost you a cent. If you ask for and pay for 2oz and get and measure 1oz, they will replace them for you or get their building bombed :-) Seems straight forward to me :-)

I agree that if you wanted to you could just run a loop around the name and hang two terminals from it. No stress either way :-)

[jharvey]

I had almost forgot that ground trace around the board for noise reduction. So I'm working on adding it, I'm also making it duel featured by allowing it to function as a copper thickness check. I'm aiming for the narrowest trace I can, howevere, it appears 6/6 is a very common standard, so the min width is 6 mills. The trace will have a resistance of about .5 to .7 ohms. If it's 1 oz, it would be around 1 to 1.4 ohms. A bit harder to measure with an off the shelf meter, but should work well for most applications.

So I also need to get another release out there, I was hoping for the updated pinout, but I think I'll have to simply give it a go anyhow. I've added most of the BOM, as well as sub boms. The sub boms allow you go simply copy and paste into Newark's quick paste feature. Making ordering parts a painless task. I think the only part that can't be obtained from Newark is the PIP3104. I'm almost done with checking the BOM's, and I don't think it will be long before I have that ground trace done.

Any how, just an update.

[Fred]

I can't work on the pins during the week, so I'll keep on with the comms stuff on the train etc, but this weekend I AM going to do the research and at least give you guys an update. I would like some feedback and suggestions on that rough hardware interface doc too, though I have a lot to do to it before I'm out of ideas/things to add myself.

Attached is a pdf for you ;-)

Fred.

[jharvey]

Awesome and thanks for the PDF. I'll try to make some contributions to the document, but right now I'm trying to get the BOMs wrapped up and do another release with the couple of changes that have been made.

Some things I need to learn more about are the differences between staged vs normal injection. What a PIT timer is and how it works, as well as some of the other misc hardware in that little black box.

[Fred]

Some things I need to learn more about are the differences between staged vs normal injection. What a PIT timer is and how it works, as well as some of the other misc hardware in that little black box.

Staged is where you have a full second set of injectors. On a 4 cylinder it means 8 injectors. When the first set run out of flow, the second set kick in to supplement it. Most people won't use staged injection. I need more fuel, but am tending towards a single larger set. There is an epic thread somewhere with me and Abe arguing to the death about it :-)

Learning about PIT... ME TOO!! Not quite there yet. It's a hardware timer without any hardware pin access. I *should* be able to make it work to control ign and staged, but I'm not there yet.

Fred.

[jharvey]

I would like some feedback and suggestions on that rough hardware interface doc too, though I have a lot to do to it before I'm out of ideas/things to add myself.

Here's at least a 10 second feed back.

Under analog inputs, I suggest we filter as low as possible, perhaps a pinch lower than required as the roll off probably won't be much of a problem. So on the temp sensor where the input responce time is several seconds if not minutes, I suggest we use a large cap and a cut off of say around 10 hz. With the TPS we should have a roll off of around a 200 to 300 hz.

Under Extra Stuff, I suggest we specify a default pwm pin, but do it via small table so it can change if required for different layouts. (Jared ducks to avoid pie plate tossed by Fred)

Any how, back to work.

[Fred]

Under analog inputs, I suggest we filter as low as possible, perhaps a pinch lower than required as the roll off probably won't be much of a problem. So on the temp sensor where the input responce time is several seconds if not minutes, I suggest we use a large cap and a cut off of say around 10 hz. With the TPS we should have a roll off of around a 200 to 300 hz.

Good suggestion, but I think such things should come with a discussion about reasoning and some suggested values and perhaps min values too? Good idea though.

Under Extra Stuff, I suggest we specify a default pwm pin, but do it via small table so it can change if required for different layouts. (Jared ducks to avoid pie plate tossed by Fred)

I'm actually all for configurable PWMing etc, BUT, it may not be as easy as you might think to do that sort of thing. No pie for you! :-)