KISS Eval board (general discussion)
ManAtWork
Posts: 2,178
The first post of this thread is reserved for the most important informations about the board. Features are:
* P2 with all 64 IOs available on pin headers
* pin compatible with P2 EVAL accessory boards
* 3.3V LDO, 1.8V SMPS, 5V in
* 512kB boot flash ROM
* NO video (onboard)
* NO hyper RAM
* prog plug connector for programming (NO USB onboard)
* fits on standard breadboard
* 25MHz crystal
* dimensions 30.5 x 76.5mm (3" x 1.2")
* optional SD card on newer version
Manual and schematics are available in the links below.
* P2 with all 64 IOs available on pin headers
* pin compatible with P2 EVAL accessory boards
* 3.3V LDO, 1.8V SMPS, 5V in
* 512kB boot flash ROM
* NO video (onboard)
* NO hyper RAM
* prog plug connector for programming (NO USB onboard)
* fits on standard breadboard
* 25MHz crystal
* dimensions 30.5 x 76.5mm (3" x 1.2")
* optional SD card on newer version
Manual and schematics are available in the links below.
Comments
You can solder single row pin headers into the outer row on the bottom size and into the inner row on the top side. The bottom pins go into column "I" and "B" of the breadboard and the top pins can be connected with jumper wires. JP2 (power supply) goes into columns + and -.
To use accessory boards double row pin headers on the top side are required. If you need both you can use longer pins in the outer rows so that they are available on top and bottom.
Can somebody confirm that the width of an accessory board is 20.3mm (0.8")? Do they fit if the gap between adjacent headers is 0.3" (2 pins missing at 0.1" pitch)?
Only some of the accessory boards are 0.8". A few are 1.2", and the VGA/AV board is 1.3".
I'd call it SimpleVersatileP2 (SVP2). An excellent starting point. I had better name for it but just didn't like the short of it (those connotations from the past long gone)
I'll only provide SMT assembled boards. Because of the different pin header options everybody has to solder the THT parts by himself. And there will be no individual options. So if anybody needs a bigger ROM or a differen crystal she/he needs to take the soldering iron... Sorry, to save time I only took components I've already got in stock.
I live in Germany, so shipping inside Europe will cost ~13€. Overseas shipping will be much more expensive so if anybody in America or Australia is interested it would be good if we could bundle shipping.
My only concern is the lack of a USB port, but these are not hard to layout on any user PCB, same with a VGA connector, etc. After market add-on boards are also available. Possibly add/include in documentation a sample PCB layout example for those who may want to copy & paste onto their own PCB, along with a BOM of required component parts. Question: You are using a 25MHz crystal, the EVAL board uses 20MHz crystal (any reason?)
As for the width of the EVAL accessory boards. I have a EVAL Protoboard and it is 1.2" wide. The regular width is o.8" wide. This is not a problem because anyone using the KISS board probably has an EVAL board and has moved on to designing their own (production) product using the P2 chip. The KISS board has everything I need - and has the parts that I can not readily design & manufacture ie. I do not have a pick & place machine and reflow soldering ovens to do the fine pitch work ( I assume you have production equipment ? )
* It often did cause problems with the P1, resetting the propeller when it shouldn't
* it costs money and board space
* Windows installs a new driver instance for each board, so if I test 100 boards I have 100 VCP ports installed. After the 255th windows has a problem... you have to de-install them all manually
* everyone I know already has a prog plug
* I don't like micro USB connectors, they tend break or wear out
I have 25MHz crystals in stock because it is the standard frequency for ethernet PHYs.
Making the pitch between adjacent IO port groups 1.2" would make the whole board very large. I think it would be no problem to add a spacer if neccessary (a 12 pin female header with long pins formerly known as "wire wrap socket").
Ditto!
It's early days but I just received a bunch of those magnetic adapters (aliexpress) and they seem to be OK. I decided to order a bunch more of a slightly different design and that actually swivel.
I am surrounded by metal chips and so I have had to clean one, once. It's nice that if you are clumsy, like me, you can catch the cable with your foot and the phone no-longer goes flying across the room
Since I will be stocking all of Peter's boards for North America distribution, I can add all your products for the same.
PM me for additional info.
I think windows needs to see change in the port ID info, before it increments.
There looks to be room to allow for both USB and bare-wires power on your layout.
I like the USB connectors with full stakes, as those are very sturdy, the one shown (below) is 4c/100, so has minimal cost impact.
If allowing for 'bare-wires power', maybe choose a wider-supply regulators would be a safer choice ?
Watch JonnyMacs Edge Breakoutboard on Zoom.
I quit.
Sad. It's your decision but, seems to me, a bit premature. You've already put some work to it and clearly described the board and the goals in the first place. I'm sure there is room for the board like this. Just give it some more time and people will come back asking for it.
I like the approach you were taking, and it would be good if you could continue to pursue it. I think the Edge is good, but I'm concerned about the connectors required to use it. A few days before you made your initial post I was thinking about something like what you were proposing. I think using 0.1" spaced pins would be better than using the Edge connector. I was thinking along the same lines as Peter's P2D2, but with a minimal amount of components like the KISS Eval board. You might consider two versions of the KISS board -- one with the Eval board pinouts, and another with just the 64 P2 I/O pins, and minimal ground and power pins. That latter version would be slightly shorter, and have fewer pins than the Eval board.
If there's a product B that can do everything product A can do and B is better in every perspective then there's no longer any reason to continue to make A. And even if the Edge board takes a little longer and cost a bit more people will wait and pay for it because your signature is on it, Jon. I can't compete with that. If I stop now I loose a day of work. If I continue I'll loose more and nobody gains anything, IMHO. Making eval boards is not my main business. I have lots of other stuff to do.
If anybody disagrees feel free to use the files attached.
Two key points
- your board is much more compact, about 4 vs 28 sq inches
- EU made is important for EU, i believe
cheers Bob, retired electronics engr.
Exactly. My servo project is still waiting for completion.
Information leak... There will be a small 0.1" breakout adapter announced soon (if not already), part number #64019, which accepts the Edge and gets you to a breadboard or veroboard. I didn't see yesterday's Zoom yet, but if that wasn't included you will find it shown at the next Zoom. Item on target to be in stock November 1st, along with the Edge module.
Both valid points.
Man, that Edge thing will be huge at 28 sq inches. Easy to find on a cluttered desk but no fun when one tries to incorporate it into some actual commercial product if it ever comes to that.
Commercial product could just include an edge socket, then plug the edge in direct- no breakout needed? That's one of the principals of the first edge module version- to be as low cost and simplified as possible, suitable for integration.
Sorry @ManAtWork for going OT with your thread. I personally think your solution has it's own great merits- it's something I wanted to do as well; to extend from the FLiP design to make the FLiP2. Maybe one day that will happen, but if the community makes their own boards then that's brilliant too. There's always going to be different needs, and I think the more choices the better. Unfortunately it seems we all have far more ideas than time these days, but together we can move mountains silicon
You have to compare the Edge board against the KISS board without the breadboard. I think only the big breakout board is 28 sq inches. The breadboard in my picture is even bigger.
The idea of the Edge, if I understood correctly, is to use both together for experiments and only the Edge board in the final product.
When it's available at Mouser you can order it without paying customs fees and shipping will be cheaper than now.
Veroboard? Get outta here. While that was ok back in the 70's for discretes and the odd dip chip, I use matrix board with uncommitted through hole and plated pads on proper FR4 board thank you. That old veroboard would bend and tracks would lift or they would short because you didn't cut them clean etc. But 100mil matrix board is limiting anyway, which is why I am getting my 50mil microMAT boards made up. I know they don't come with octal sockets, but they suit smd components as well as through-hole and 50 and 100mil header pins. Even 10 years ago DIP CPUs were very hard to come by, so I think this mindset needs to change. The P2D2 has 100mil headers but they were taking a lot of extra room and besides, it was too hard to find low-profile headers for it. Now the P2D2 has two 44-pin 50mil connectors on a 1" row spacing, and it is feeling so much better for it. Mind you though, I can slap that slim P2D2 smd fashion onto a fat 100mil header board exactly like my last version, so you can have the best of three worlds, 100mil, 50mil, and smd.
My bad then. I took the 4 vs 28 comparison for granted hence the comment. It pays off to pay attention to detail and verify things. Thanks for pointing this out.
What really matters is the total cost (tax + shipping). With the Parallax brand on it I do not expect it to be even close to your offering. Time will tell.
Here's the thing: Ken showed me their version and gave me one shot for feedback -- a lot of back-and-forth would make it impossible to deliver when they want to deliver it. Herein lies the problem with design-by-public-committee; it shreds schedules and you end up with people that aren't happy, anyway.