Teensy 3.1 based BotBoarduino type carrier board

I did play a little more with the idea of having the ability to separate the USB and EXT going into the Teensy. Here is an updated 3d image showing a 2x2 at the top left. Not sure yet, if I need another thing like this that requires a jumper to run, especially if you are not cutting the trace on the bottom of the Teensy. Optionally could add in a solder jumper as well, so you only need to put in this 2x2 if you are going to use this option. Not sure if it should be a normally connected and you have to remove the etch (like the teensy), or open and you need to solder over it to make the connection.

Thoughts?

Also put in the ability for having 4 pin serial connectors for AX like my other boards which I believe the OSH Park ones will arrive tomorrow. Not sure when the Seeedstudio ones will arrive. The mail system says it can take 1-30 days from China. Shipped on Friday.

Depending on how long Seeed takes, may have to order some of these to see if it works.

Forgot to mention, that my Teensy 3.1 breakout board (v0.2) arrived a few days ago from OSHPark, which I have assembled one of them:


The one on the right is the new one, with the Teensy 3.1 also from OSHPark, so have matching purple. This one has the larger terminals to make it easier to hook up the larger wires, as well as the adapter soldered on the bottom of the Teensy, which gives access to several more IO pins. Which there are Male IO pins that make them useable. Not enough room to add 3 pin headers for all of them. The XBee area was made more friendly for the XBees as to not have any electrical stuff running in the main keep out area (area in front of the last IO pins, where the Ant is. Did leave ground plane on bottom of board in this area, as it sounds like wire Ant likes…
I have started testing and so far nothing bad with it… Note: I also ordered a set of them from Seeedstudio to see how well they do. They arrived yesterday at my Mailbox and are waiting for me to go into town to pick them up. Hopefully they will be good!

Kurt

The new ones from Seeedstudio look fine :slight_smile: As you can see with it next to the ones from OSH-Park. Note: They sent 6 instead of 5

Very very nice… :wink:

This is very nice, and very flexible. I like! :slight_smile:

Very nice! :smiley: I think this gives maximum flexibility for using both kinds of servos at the same time. I can see possibilities where we might want to use mostly standard Hitec type servos, but maybe an AX or more to do heavy lifting or other tasks.

Oh, I will be doing a full build when I get boards. This will give me a good excuse to get one or two AX servos to play and test with too. :slight_smile:

I think the way the BotBoarduino has things is plenty nice enough. I love the ability to jumper between Vl and USB for powering the electronics. I jumper to USB for powering the electronics when I am testing. I’ll rejumper to Vl when I want to power off a battery, and run the robot on the floor.

Again, I like the way things are setup on the BotBoarduino for power. Unless you can see a real need to do more, I don’t think it’s necessary.

I think this is a desirable feature. I’ve heard (but can’t remember where) some boards have auto sensing for the IOREF, but am not sure this is necessary. It’s probably also a pain to do. I think the jumper setup is plenty nice. I can’t imagine wanting (or needing) to mix 3.3V and 5V shields.

8-Dale

If my thinking is correct, and I understand what you are talking about, I believe the USB power should normally be separated from the EXT power. Things powered via USB should never be connected to things requiring EXTernal power. I’m thinking in terms of things like the AX servos, which require 11V.

I am getting excited to see how this works also! I am really liking the way this is turning out and am looking forward to getting some boards for myself.

8-Dale

Very nice!! :smiley:

8-Dale

Yep on Power I was trying to make it as completely general purpose as possible. With two sets of input connectors it is setup to allow them to be used multiple ways. Example;
If you have AX servos, you can use the one set for those, and the Other one to feed voltage Regulators which puts out the +5v/+3.3v on the board and these can be fed into the Teensy or not… And you can choose which of the two main power inputs to feed the VS pins. If you have all of the jumpers in place and one main battery you can also use it to daisy chain to another board, the way we do from SSC-32 to …

As for making it do what the Botboarduino does, I am not sure I have that complete flexibility or not. That is I am not 100% sure what the Teensy actually does when you cut the jumper trace on the bottom of it… Don’t really want to experiment that much on mine. Works fine so far with out that on my Carrier board. Currently the board is looking like:

Notice I added 3 pin jumper for IOref. Personally I am thinking of removing the double jumper stuff at the top, that seperates VL from VUSB (personally I would never use it). Currently I have a simple soldered in jumper on the bottom, which the verification code of Diptrace hates.
Solder-Jumper.jpg
I may remove it. WIth it, I don’t need to populate the 4 pins and it works normally. Only if you need to play around would you cut it and then use the external jumpers. The other option which the program would like better is to add a 0 ohm resistor connecting the two circuits.

But I think I am pretty close to being done with what I will be doing here.

Kurt

Quick update: After receiving some more feedback up on Trossen forum (forums.trossenrobotics.com/showt … 3.1/page12) I decided to leave the 4 pin jumper at top. I did not like the trace jumper on the bottom as the verification process did not like it and showed up as short… So I converted it to a 0402 resistor on the top, that can be populated by a 0.0Ohm resistor. I decided it was not much of a cost hit. $.10 for 1 and $.14 for 10…

I also decided that while I am not sure what I will do with them, I spent enough time on it, so I put in an order yesterday to Seeedstudio, for a batch of 5. In case anyone else is interested, I have included a zip file with the Gerbers I sent them. The zip file has the files that Seeed wants as well as the names that OSH-Park like.

In addition the zip file has the design documents. It has the Schematic and Board layout files. Note: I am using the Beta version of Diptrace right now and these file formats are not compatible with their released version. But can export to their Ascii version and import back into released. I also have PDF files in this directory where I did a print to adobe for both the schematic as well as the layout.

In addition this directory has an excel document with the parts list. I put in a little color coding into this document where the parts needed for AX-12 servos is called out. Likewise parts for possible update for full duplex servos (which I will not populate) as I have non of these.
TeensyDuinoV001-wide.zip (250 KB)

Look what i got !!!

http://farm4.staticflickr.com/3756/13158567325_f6b315c26d_c.jpg

Looks familiar :slight_smile:

I just finished assembling a 2nd of the V0.2 breakout boards. This time one of the green ones fabricated at Seeedstudio. I used a Green Teensy 3.1 ordered direct from PJRC… Still doing some testing.

Actually I already noticed one thing I should have caught earlier. Logically I think the TX/RX led indicators should be reversed. Always interesting on how to label, should TX be TX from the processor or from the XBee. In this case it is from the XBee, but logically makes more sense that TX should be what is going out from the board (DIN on xbee…) No big deal…

This time I cut the etch on the bottom of the Teensy that connects USB to VIN. So in that state only powered when external power is applied. I soldered on a jumper to the VUSB pin, that I can connect up to any VIN pin and power up that way to experiment). Would do slightly different now, and make it a little easier, but soon will be on robot so won’t matter.

Note: I did setup actual jumpers on my logical Teensyduino (with arduino headers) board, that can easily do with shunts. I have been tracking the shipment of the fabrication of this board and as of this morning cleared customs and is waiting to be transferred over to USPS. So should have probably within the week. So I checked my supply of parts and ordered what I needed from Digikey to assemble one or two of them.

Not sure how much testing of the shield parts I will do as I don’t have many shields. I have a prototype shield and I an V1 Servo Shield I can try out.
Thinking of trying out an XBee on one, but would like a shield that is not 100% tied to pins 0,1. Also cheap as this is only a test. Thinking that something like: robotshop.com/en/xbee-shield … =GoogleUSA
might do the trick $6. May have to get my hands on one, but the cheapest shipping is $13 and will take a week+, maybe should order something from Amazon…

The batch of Teensy breakout boards with Arduino Headers arrived today from Seeed studio. Here is a picture of one of them where I test fit the arduino headers, as a first test to make sure I did not … it up. In the picture is the pile of the other 4 of them, plus the most recently assembled of the other design. I also show the few other shields I have for Arduinos (non-mega), that I may try out on it.

May be a few days/week before I will get a chance to assemble one of them. I also have a couple of other shields I purchased and were shipped today from Robotshop which should arrive in about a week.

Quick update: I assembled one.

Not sure if any of my current shields are compatible enough to play with. Ones I ordered at Robotshop should arrive Monday.

That’s a beauty, with round corner… :wink:
I still have to order all my things to populate mine.

Yep - Looks pretty nice with the rounded corners. I have done a little playing around with this new board, with a couple of different shields. A couple of them arrived from Robotshop on Monday. I will soon try out the Seed Studio XBee shield (this one allows you to choose which IO pins to use). The unfortionate part about some of these shields is they extend aways in the direction of the Battery connector instead of the other direction as I would expect. So they extend over many of the 3 pin headers I put on the board. I could help it out some by using taller headers as to give more room or if I were to make another version, I could either lengthen the board or try to rotate where I put the headers to be normal… But that would be a lot of work…

Personally I am playing more with my other design, that is the same size as the Botboarduino/SSC-32… that has an XBee on it.

Currently playing with the DIY remote control and converting it over to be Teensy based. Have most of it in place now:

The underside is still a ratsnest but mostly all there. Need to wire in battery and probably some debug cable that I will leave in place.

I am also thinking of simplifying the protocol I use in the remote control. Not sure of having the Robot ask for data when it want’s it as I currently do is better than simply sending the data N times per second and the robot grabs it when it needs to. Will see. Still want the bi-directional communications that allows the robot to have the display show messages, play sounds, display battery status…

Kurt

P.S - When I get the Quad parts, it may be hard for me to resist from using this and a Teensy on the Quad :wink:

As I mentioned else where, I now do have the Teensy 3.1 on the SQ3 where it is also driving the 12 servos itself, using my ServoEx library…

I also wanted to mention that I put my design files for the two different Teensy 3.1 break out board up on github: github.com/KurtE/Teensy3.1-Breakout-Boards

Your remotes are looking very nice guys, I think I’ll use that thread in the futur :unamused:

Just thought I would give an update here. I ran into an issue, where I found my 3.3v circuit was not working properly. At first I could not find any obvious problem. But the 3.3V was getting pretty much 5v, which could be bad if what you are connecting too does not support 5v. I checked the output of the 3.3v regulator and found there was pretty close to 0 resistance between the 5v input and the 3.3v output pins. So suspect short, Checked on unpopulated board the two circuits are separate. First I thought maybe bad solder job. Was maybe true on first board, but not obvious on 2nd. On the 2nd one, I cut the VR off of the board and cleaned up solder. Checked an no voltage on 3.3v circuit. I then ran a jumper from the Teensy chip 3.3v output to one of the optional PU resistor pins for I2C and now have 3.3vs on that circuit.

What I am thinking, is that maybe the capacitor is too close to the VR, which makes it easy to short and/or where I ran etch off of the pattern could make it easy to short and/or maybe the via that is under the VR could short out. So I edited my layout to change all three of these cases and will probably order a V0.2 from Seeeduino with these changes.

I made a few additional changes as well, including:
a) Did not have a 1uf capacitor on the output of the 3.3v VR as datasheet recommended so added one.
b) Not sure in all cases I need 3.3v regulator as Teensy has it’s own. so added optional pin, you can connect to the teensy, which uses the 3.3v output, in which case you should not populate the 3.3v VR
c) I don’t like those little switches I used, so switched back to switches like on Botboarduino.
d) Move the logical shield a little logically closer to the USB side of the card, as to give more clearances to the 3 pin headers on the other side.

I think that is about it. Again may order new batch from Seeedstudio. But may wait a few hours or day in case anyone quickly thinks of other changes I should make.

Kurt

P.S - Diptrace released the new version of their software, which I am now using instead of the beta version.

Decided to go ahead and order the new batch from seeedstudio. In addition to what I mentioned in the previous post, I also played a little with the Analog circuits. In particular, I no longer simply connect up the AGND pin directly to ground. Instead, I created a new circuit, and used the AGND as the GND for the Analog input only pins. From some postings up on PJRC, found out the AGND pin is setup to hopefully be a cleaner ground for more sensitive circuits, which is why I brought it out to the Analog only pins…

I uploaded the Schematic/Layout files to my github project. Soon will reorganize that project to more easily see the different versions. Also will update the parts list. Main changes is, new capacitor, different switches, but will also try to document some more which parts are optional…

Kurt