Announcing the ORDuino !

Topics Related to the ORD Bot Printer

Re: Announcing the ORDuino !

Postby orcinus » Sat Mar 03, 2012 9:35 pm

Same here. I'd love to help, but can't commit because of a chaotic work schedule.
And i'd hate to let other people's time or resources go to waste.

Edit: BTW, do you plan on offering bare PCBs or kits?
orcinus
 
Posts: 720
Joined: Tue Feb 14, 2012 4:03 am

Re: Announcing the ORDuino !

Postby frob » Sun Mar 04, 2012 3:57 am

orcinus wrote:Same here. I'd love to help, but can't commit because of a chaotic work schedule.
And i'd hate to let other people's time or resources go to waste.
Edit: BTW, do you plan on offering bare PCBs or kits?


Thats too bad but i understand. Perhaps i should have more clearly explained what i am hoping for in terms of testing and help with firmware tweasks and adjustments, which is not very much actually - i'll do that a little later once the design is completely finished.

No plans to offer a bare PCB or kit mainly because i have a lot of experience assembling SMT boards by hand and i know how hard it is for some packages like the QFN the motor drivers come in.
when the parts don't dissipate any significant power you can get away with it ok most of the time, if you have a lot of skill and decent tools.
but like in this case, the motor drivers do burn quite a lot of power so it is critical to have a perfect, void-free solder joint for the center thermal pad to have a chance of reliable operation, and because that pad is purposely connected with very low thermal resistance ( a lot of vias) to a large heavy copper ground plane, it just is impossible to accomplish that without a proper reflow oven and carefully adjusted soldering temperature profile. plus even if you do have a reflow oven, its practically guaranteed you wont get it right on the first try.
Luckily i do own a real reflow oven (not a kluged toaster), as well as IR and hot air rework stations, and have good relationships with local CM's with very expensive x-ray equipment so i will be sure my prototypes are correctly soldered.
By the way these will use very high quality materials- production boards are silver plated, and will be lead free and 100% ROHS compliant so acceptable to export to Europe.
If demand is high i may even consider bringing these along on my next excursion to the the compliance lab and sneak them through FCC/UL/CE testing assuming there's time left on the clock when i'm done testing my work projects.

Anyways because the board is almost 100% smt , the automated assembly is the least expensive part of the process, once the setup charges as paid for.
The price reflects that and all the handling & logistics required to get these made and distribute them etc. the latter which is not insignificant.
packaging kits with dozens of little baggies of smt parts would cost a lot more than the automated assembly actually will, so it would mean you pay more, not less, for a kit.

The good news is that its 100% smt.
which means if this board gets popular and i can order production batches in larger quantities, the cost of ongoing production should come down fairly steeply,
and i can unleash my natural generosity and reward the early adopters with some very nice discounts on future production models. :)
frob
 
Posts: 260
Joined: Tue Jan 31, 2012 6:24 pm
Location: Montreal, Quebec, Canada

Optional thermocouple interface: yes/no? how many channels?

Postby frob » Sun Mar 04, 2012 3:51 pm

please vote today if you have an opinion on this!
# channels: extra cost (conservatively):
0 : $0.50
1 : $20
2 : $25
4 : $30

uses 18-bit I2C sigma-delta A/D converter and 1% accurate temp sensor for cold-junction compensation.

SMT screw terminal blocks push-in spring-clip type terminals ?

Note: adding the circuit but not installing the parts might add 1 sq inch of PCB area so about 50 cents.

Edit:
this also has a slight influence on another part of the design:
I'm including digital potentiometers to set motor current instead of manual trim pots ( uses less space in a critical zone - plus more reliable, less trouble / risk to adjust, and controllable through software - price is a wash with trim pots, so no extra cost really)
If i include the thermocouple interface, the digipot will be I2C types - otherwise i'd use SPI types.
frob
 
Posts: 260
Joined: Tue Jan 31, 2012 6:24 pm
Location: Montreal, Quebec, Canada

Re: Announcing the ORDuino !

Postby bdring » Sun Mar 04, 2012 5:07 pm

Are the digital pots non-volatile? If not what level do they start at? I am concerned they often might not see the setting code for while if people are playing with or reprogramming the unit.
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

Re: Announcing the ORDuino !

Postby orcinus » Sun Mar 04, 2012 5:22 pm

bdring wrote:Are the digital pots non-volatile?


Hope so, or nasty stuff might happen. Most Analog Devices I2C digipots are non-volatile, i think.

frob wrote:Thats too bad but i understand. Perhaps i should have more clearly explained what i am hoping for in terms of testing and help with firmware tweasks and adjustments, which is not very much actually - i'll do that a little later once the design is completely finished.


If there will be no coding from scratch, but debugging and tweaking (and if there isn't a strict schedule), i'm in, if you'll need me.
What's the ballpark price-point estimate for the beta boards?
orcinus
 
Posts: 720
Joined: Tue Feb 14, 2012 4:03 am

Re: Announcing the ORDuino !

Postby frob » Sun Mar 04, 2012 6:07 pm

bdring wrote:Are the digital pots non-volatile? If not what level do they start at? I am concerned they often might not see the setting code for while if people are playing with or reprogramming the unit.


The pots default to 50% when they power up or reset. They are available in non-volatile or volatile.
I'm planning to use non-volatile types, but it should be safe to use volatile (they're pin compatible)-

The motor drivers all have their reset, enables, and sleep inputs connected to separate microcontroller i/o pins, with appropriate pullups or pulldowns such that when the mcu is reset or unprogrammed, all those signals default to their disabled state, so motor drivers are shut down by 3 separate (redundant) means any time that happens.

Because there's not a lot of inventory of these at the moment, in case it gets difficult to obtain non-volatile types after the initial production run,
i plan to use volatile ones in the prototypes (well at least mine), to confirm that there is no issue- in case i have to substitute later due to lack of availability of non-volatile parts.
frob
 
Posts: 260
Joined: Tue Jan 31, 2012 6:24 pm
Location: Montreal, Quebec, Canada

Re: Announcing the ORDuino !

Postby frob » Sun Mar 04, 2012 6:29 pm

orcinus wrote:
bdring wrote:Are the digital pots non-volatile?
Hope so, or nasty stuff might happen.
Not too likely either way- i'm more of a "belt and suspenders" kind of guy- see my last post. :)
orcinus wrote:If there will be no coding from scratch, but debugging and tweaking (and if there isn't a strict schedule), i'm in, if you'll need me.
What's the ballpark price-point estimate for the beta boards?
Really only minor additions to the sprinter or marlin firmware- i have a couple people already signed up to help with that - of course the more the merrier, but at this point i'd only expect to just having it put through its paces as much as you're able to before i send out the production PCB order, and help make sure no bugs inadvertently slipped in with code improvements.
After everything is connected up i imagine that means a few hours to a day or so's worth of effort on the outside.
Of course having a real and already tested printer mechanism would help tremendously with that :-)

Protos/betas will be "all dressed" with LCD and complementary heated build plate and flex cable, all for $150.
I'm leaning towards throwing in the quad-channel thermocouple interface too.
frob
 
Posts: 260
Joined: Tue Jan 31, 2012 6:24 pm
Location: Montreal, Quebec, Canada

Re: Announcing the ORDuino !

Postby orcinus » Sun Mar 04, 2012 10:25 pm

Don't have a running printer yet, but do have a PrintrBot on the way (i hope) before the end of the month.
From what i've seen, it shouldn't be much of a problem getting it up and running fast. So i can test with that at first and with ORDbot when the kits arrive and i get it assembled.
orcinus
 
Posts: 720
Joined: Tue Feb 14, 2012 4:03 am

Re: Announcing the ORDuino !

Postby frob » Wed Mar 07, 2012 4:04 am

ORDuino.png

Here's a quick update :
Finally got the placement to a point to where i'm satisfied it will work well for routing the way i like and adequate isolation and grouping of sensitive bits away from noisy bits etc, and got the motor drivers routed on top, which you can see on the top row - XYZ on the left, extruder channels on the right.
In the end i was able to squeeze in both DAC''s and traditional trim pots, so i can use either.
Sorry there's no 3D models of the connectors and other large parts yet, those take time to track down and load in so i'll only get to that once i'm all done with routing.
As it stands the board itself is 170mm X 60mm but as you can see some of the connectors cantilever off the board a bit when they're installed, which might add another cm or so to the dimensions when you count that. i may increase the board just to better support those connectors but without pushing them further out, so effective size wont really get any larger.
There's quite an array of optional power connectors over on the left edge, some of them i don't expect to be popular overlap each other so you can put one or the other but not all of them - however you can get 2 and maybe 3 options in parallel, including the 1/4" faston tabs, ATX connector, and pluggable screw terminals.
Thermistors are on bottom edge on the right, and thermocouples on the right edge - they along with all the motor connectors all use the same 8x 3.5mm SMT screw terminal block - though i might possible redesign the pad to also support a through-hole version should the SMT pins prove to not be great in taking the usual abuse - though i have already used the SMT ones in a commercial design and they seem surprisingly strong.
There's a RS-485 transceiver that occupies the last 2 spots on the right end of the thermistor interface.
For the Thermocouple interface i actually manage to fit 3 different versions of the circuit so i can build it as a 1 channel, 2 channel, or 4 channel model - assuming there's any demand for that at all.
The heater outputs are along the bottom edge on the left, with the FFC header for my heated build plate as well.
As you can see I had placed the USB connector just above the CPU in between motor drivers, mainly because i wasn't sure if there would be room anywhere else, and that's the side of the CPU with the USB connection. Well it turns out i could fit everything with some room to spare so i will probably move that to bottom edge instead and put a motor voltage bus electrolytic cap up there instead.
Also i don't have the LCD connector on yet for same reason but that will go on the bottom edge below the CPU - also its hard to see the little SD card connector there just below the CPU a little to the right side, but i might be able to switch that back to a more traditional SD card connector along the bottom edge too.
There's still quite a lot of little tweaks, cleanup and DRC checks to run, but the hard part is finished, the rest of the routing will be a breeze -
Unfortunately i have never been so busy with work projects so the time i can put on it this week is going to be next to none, but i should have this finalized and ready to send to fab by the end of the coming weekend nevertheless.
Then i'll get all the 3D bodies on the board for some prettier renderings and output an .STL file for anyone who wants to play with fitting it on the ORDbots :)

EDIT: if you look carefully ya might spot a few other goodies i forgot to mention, like the dip switches for configuring stepper mode and a few other things, like locking the write protect on the DAC's eeprom, setting RS-485 bus address, etc.
there's 4 RC servo headers just above that.

Also it seems that of all the signals in the ATX connector, there's only one 12V pin/wire, but lots of 3.3V and 5V - so not so great by itself for all those pins. What we really want to use is the optional 4-pin xtension, which is often seperate, which is usually for CPU core voltage regulators - this one has 2x12V and 2x GND lines. so i'll probably eliminate the DIN connector and pit that in there instead, but scooch down the ATX connector and install it at the top left corner instead.
Last edited by frob on Wed Mar 07, 2012 5:02 am, edited 1 time in total.
frob
 
Posts: 260
Joined: Tue Jan 31, 2012 6:24 pm
Location: Montreal, Quebec, Canada

Re: Announcing the ORDuino !

Postby Enraged » Wed Mar 07, 2012 4:19 am

how many thermistors can it take? if it can use 3, that would let you control temperature on 2 hot ends and a heated bed.

for the LCD, are you planning to use I2C?
Enraged
 
Posts: 439
Joined: Fri May 06, 2011 3:13 pm

PreviousNext

Return to ORD Bot

Who is online

Users browsing this forum: No registered users and 4 guests