Puma board for FreeEMS

[Fred]

AND, for anyone to make progress, they need to order parts, not all the parts, just the parts they need, so the BOM must be the priority, BOM with "if this then this" sections all through it.... how can someone have trouble with assembly/mods if they can't order the parts they need accurately in the first place... again, no replies till I've dealt with the above stuff. I was furious when I read it this morning. Failure to take me seriously in this post and the last one will not be pleasant.

[Fred]

I expected that spin1 would need some patches and changes as it's the first run and there are still many unknowns etc. What I didn’t expect was the absence of documentation explaining those patches, let alone no BOM.

Especially when most of the stuff is documented, in this thread, by me, and in other ways. It needs pulling together.

I know the HW guys have put in a TON of effort too. But if were not on the same page were not really headed towards the same finished product. Look at what happened to all the guys who left diyefi.org to peruse things on their own.

Yes, at least three failed spin off projects, so far.

All the key people as of now have SPIN 1. We need to get those built so we can get the SW DONE(at least for the most part).

A fair assumption, but possibly a wrong one. Marcos hasn't had a lot of time, and had to spend time on things, that didn't progress the project, for his SASE entry. He'll be back from his holiday soon and hopefully will get the docs in tip top shape and the remaining boards built and sent.

The is what the FreeEMS project needs.
1. Good docs for Spin1.
2. The software guys to get busy getting things sorted and tested.
3. MORE TESTING
4. The software guys to give the hardware guys a set of std cpu pin-outs etc.
5. Finalize Spin2

Exactly, this project is utterly worthless without the software being good, and it can't get any real world testing without people installing it or trying to. The fact that it worked for me, so far, twice more or less straight off the bat is just a testimony to the care put into the code. It still needs real world testing to be proved robust, and to finalise some of the design parameters that AFFECT THE BOARD DESIGN. (see point 4)

You can work on 1 and 5 and the same time to a certain extent, so it's not like your are bored.

Exactly, but we need the spin 1 stuff in at least an initial close to good state for people to make ANY progress.

TBC...

[Fred]

What I didn’t expect was the absence of documentation explaining those patches, let alone no BOM.

I understand there is a BOM, on puma.freeems.org. That has been there right all along. Also after spin was released, a reference designator BOM was created.

That BOM is useless due to the number of optional components, different options, necessary omissions, value specialisations, etc. That is why a BOM is required with a "everyone needs THIS" section and appropriate other optional and configurable and customisable sections.

The money isn’t really a concern(I spent $25 big deal). What is a big deal is time. Go on ohloh.net, look up our associated projects and have a gander and the estimated project cost(s).

Nice, does that mean I should put in for a raise? It claims the cost so far is $116k. Which I believe to be low if one were to try to hire a team to do it. I don't see how that's relevant to spin 1 or the next spin.

Time is money. By not documenting things and not distributing boards to those that have ordered and/or paid for them, we are wasting their time, and, more importantly my time. My time is limited. I'll be in full time employment again in a few months, and we need to make the most of my 24 hour development NOW while we can. Understand that I can't write code for people if they can't use it. I've promised to get people up and running with code that they require as they request it for the first people to try. As yet, no one has tried. No one can try. There are no good docs to allow them to do so.

All the key people as of now have SPIN 1.

Thanks for verifying I'm not a key person on this project. Or perhaps you know something I don't? I don't have a spin 1 in my hands at this point, so it's bit hard for me to help directly, but I'm willing to help with what I can.

This comment was completely out of line. I was || this close to tearing you a new arsehole this morning. Sean has been invaluable over the last 3 years, treating him this way when his comment was CLEARLY innocent is unacceptable. Pull your fcuking head in.

We need to get those built so we can get the SW DONE(at least for the most part).

If you can't physically assemble it from the docs available, I'm at a bit of a loss. We have told you that resistor XX is digikey number XXX-ND and goes in this location.

You have, yes, for the original design, which is so broken in so many ways, that most of the original docs are wrong. Most importantly, he can't assemble it without parts. And he can't cost or time efficiently order parts without an accurate and modular BOM. This doesn't exist and the one in the project tree needs to be forked. Marcos is aware of this and will no doubt do it soon.

Code: Select all

Perhaps it's simply a communication issue, I'm not sure what the problem is right now.

No suitable BOM.

Also the boards that have been built, have had several pictures to help iron out potential problems. What more is it you need? Can we get your help in such a task?

No suitable BOM. Sean can help with feedback once he orders and starts building and installing and using his board... he can't do that yet, because there is:
No suitable BOM.

Perhaps you can build yours and take photos along the way to help document the process.

I'm sure he would love to help, once he has a suitable BOM.

What is your general layout, low Z high Z, ect.

Not relevant, we need to do this right ONCE to save you and Marcos time. Repeating the same info thirty times is inefficient and error prone at best. One set of carefully written instructions and one modular BOM will cover everyone in the same amount of work, or quite probably MUCH LESS.

I know Fred's mods have included high Z, but I haven't seen those mods documented yet.

Yep, you're right, I've not bothered because I've had bigger fish to fry and because there was no point, Marcos was hell bent on the SASE thing, so off he went and did it, great. When he comes back online properly, I'll talk with him about it and get my contributions added to the docs along with his and yours and huffs and jeans etc. Not until then. Do it once, do it right. Let's show some professionalism to the world.

I'd like to see a vote option, such that we can get feedback about how many are out there with similar problems to what you expressed.

Are you fcuking joking? Many of the people with boards that need info are not regulars on this forum and should not be expected to understand the entire hw design inside-out to get theirs built... I'll distribute links to them once they exist. RIght now they do not.

TBC...

[Fred]

Can you elaborate more about why you haven't built your board? Is it because you don't know where a particular resistor should be located?

I was told to wait on building it "as is" because there were problems with certain circuits etc.

Yes, I told him that, and I'm right, ordering the BOM as is will result in significant wasted money, extra unused parts, and missing parts that are required, therefore resulting in a secondary order, delay, shipping costs, and overall wasted money and time. NOT GOOD.

My point was that we're all putting in huge efforts and we need to keep all team members happy and working(within reason).

Exactly, keeping the hardware ball rolling is not useful if there is no software to use on it... and the software requires some hardware to test on, and this, by default, has become it. People can't order their parts without the detailed option specced BOM. They can't build their stuff without the parts. They can't install and test the board (required for spin2) without it being built. People can't test the software without it being installed. Hence the first thing in this list is a BLOCKER for the rest. Not for me, I'm sorted, but for EVERYONE else many of whom are important, especially Sean. If you can't make sense of this sentence and agree with it and run with it, then we're wasting our time.

LOL sorry man, I figured you would have had a board in your hands long before me :) I didn’t mean it that way, poor choice of words on my part sorry. Hopefully all contributors have HW in hand.

I don't think it was, Jared read something between the lines that was not there. You shouldn't be apologising for that!

Enough of the "we wont document it, lets make another board that no one can use first" bullshit, carry on with whatever you want, but don't suggest docs aren't required or won't be produced. That is totally unacceptable. Especially when Marcos is already working on them... This is the final word. Further posts should be tech only.

My wrist is now in pain after writing all of this. Please, reduce my admin work load and keep the eye on the ball which is working spin 1 boards out in the wild getting used tested reported upon modified improved criticised measured, etc, and code on them getting the same treatment. Nothing further on this, please. Except possibly an apology to Sean. I'll accept that.

Fred.

[jharvey]

Fred, if you put your rant energy into pictures and such doc's about your puma mods, I could work on a wiki or the BOM you used to make a HighZ board. Right now, I'm still left with a lack of information, that prevents me from being able to do the docs you requested.

[Fred]

https://github.com/nitrousnrg/puma/tree/Spin1_branch

Marcos has made the branch and is underway with the docs! Excellent. He even did it correctly from the tag! Happy days. I'll help him with this as soon as he is within communication range :-)

I don't like his name for it, but it'll do.

Fred.

[nitrousnrg]

Sorry for the dissapearance, I've been far, faaar away.

As Fred noted, there are some docs to build a board. Its not complete, but at least is something. I'm still improving that.

Many things must go to the site, I've been working on it too. There aren't big changes, just a couople of needed ones (link updates, pics, "installed systems" section, etc)

About the USB connector, I'm not biased by mini or micro, but if micro is more reliable, I'd go for it. I don't have the cable, but I realize there are lots of devices that uses those connectors.

Fred, the branch was going to be named as just "Spin1", but it conflicted with the Spin1 tag. I don't like it neither.

Jared made lots of changes in the design, and they aren't yet in my repo! I've been looking at his work, but didn't merged yet. So, Jared, keep it up if you want, I'm into docs right now, when I finish/get tired I'm going to ask you and do a merge to start from you latest mods. I came from buenos aires with many ideas.

For example, I found a place to test and certify the boards against vibrations, temperature, humidity, salty atmosphere, etc. Its an university, they usually test automotive and aerospacial things. There is also a place to do conformal coating here.

There is a nice A0 poster of Puma-FreeEMS too, made for the event (ridiculous competition, btw). I'm posting it here as soon as I find the picture.

[Spudmn]

Glade to see the documentation starting to come through. I'm going to need this very soon.

I just picked up my Puma board.

Thanks Fred. Nice to meet you in person.

I am pretty sure I already have a box for it I just need to find where I put it.

[Fred]

Marcos, see your email, I have a LOT of feedback for the PDF, and a little for the latest site. Thank you very much for your doc efforts so far, a good start. I should probably start a thread and link your pdf and paste most of my email contents in there so people have some idea what is missing from the document.

As for the branch name, I was thinking "Spin1-Fixes" or "Spin1-Updates" or something like that. It doesn't matter, though, JHarvey would surely call it shed colour, and he'd be right this time. Another strategy could be to tag as "spin1-release" branch as spin1 from that, and do further tags as "spin1-update1" etc. I quite like that. I don't mind you retagging on the same hash with a new name so you can rename the branch, either. Thoughts? We should come up with a plan for next time too, as it will be much the same deal with additions and docs to be added after the design is sent to be made...

Looking forward to the poster, I hope you got it in high res, and some of you with it, and some of you there presenting, etc!!!

Spudmn, nice to meet you too! It was good talking shop, sorry if you didn't get a word in edge ways :-)

Fred.

[Fred]

Marcos, the diode we were discussing, D1, is 263mV and not an issue. I tested operation of the unit down to 4.5V input this evening, more than good enough for any application, my issue was different...

That poster:

http://stuff.fredcooke.com/100_5483.JPG

Fred.