Society of Robots - Robot Forum
Electronics => Electronics => Topic started by: sonictj on July 10, 2009, 11:57:02 PM
-
I'll been logging the progress of my 640/1280/2560 dev board for a while now
http://www.societyofrobots.com/robotforum/index.php?topic=8060.0 (http://www.societyofrobots.com/robotforum/index.php?topic=8060.0)
, but I came across this
http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1246870444 (http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1246870444)
there is a seeduino mega coming out in a month for $48. I can't compete for that in price, but I have to say I'm not all that impressed by the layout. All of the i/o is thrown at the back so densely, I don't know how you could possibly make a shield and be able to pull it apart. I aslo think its incredibly annoying to have a board with that many i/o that is labeled by the arduino number scheme.
My question is there any viability left in my design as a product. In many respects I prefer my design, but I know that many consumers would not feel the same way. If you could tell me your opinion on the matter I'd greatly appreciate it.
pros to the seeduino over my design
*accepts arduino shields and additionally has .1" spaced i/o fro homemade shields.
*low price of $48
*can be run on usb power
*has 3.3v support
cons
*unorganized/clumped array of i/o. I'd be willing to bet pulling a shield off, connected to all that, will be nearly impossible.
*Its difficult to use the board with an environment other than arduino because the pin labels are specific to arduino
*I'm betting the 5v regulator can only source .5 to 1.5 amps of current, which may not be enough when a ton of sensors, and such are plugged in.(talking about power bus the mcu has a set 200ma limit for either board)
pros to my design over the seeduino
*all of the i/o is well organized and labeled according to the avr data sheet I have also modified the arduino environment to accept these labels at the arduino pin numbers.
*all i/o is .1" spaced for homemade shields
*my board can source 3 amps of +5v, and ~2 amps of input voltage which can come in handy on a board with this many i/o.
*because I make these boards by hand I can put a 640,1280,or a 2560 on the board (currently no bootloader support on 2560, but I bet I could knock it out in less than a week).
*The standoff holes in my board make a simple rectangle compared to the odd arduino mounting holes (4x.125")
cons to my design
*price I can go as low as $65 with an atmega640 on board, but I cannot beat $48 and profit
*no power via usb (was thrown out to allow for 3 amps of +5v current)
*does not support standard arduino shields.
Other than that I belive the feature set is identical (correct me if I'm wrong)
both boards have access to all i/o and support the arduino bootloader. Both boards are ~ the size of a standard arduino. They are also both rohs compliant.
My problem is I have to sell 14 boards just to break even. If I can't sell that many there is no reason for me to sell any at all.
I attached the pdf of my board I'll have real life pictures Tuesday night when my prototype arrives.
-
that seeeduino makes me want to vomit.
the layout looks like a three year old got hold of eagle... no offense to the creators but damn...
i shield would be pure madness for that board... for things where it isn't related to robotics, this may be fine but for robotics, i wouldn't buy it for $20.
sonic, for the extra ~$20 i would definitely go for your board! its like an axon jnr :P hell, id even pay $70 for it ;D
the only thing i slightly don't like about your board is the lack of power busness...
but im sure a shield wouldn't be much trouble to make up.
i like your design way better, not just because your sonictj. but because its truly better.good work man!
i hope to see these on sale soon... i know ill buy one!
-
I didn't include power bus to avoid competing with the axon. That is actually why admin allowed me permission to advertise my design. If Admin said no to me adverting, my design would be dead already. I wanted to build a board that appeals to the SOR community, and those who love the arduino environment, not to mention the power of the Axon.
Thanks a ton for the input, and the encouragement. If there are others out there who would like to voice their opinions for, or against my design I would appreciate it.
-
ahh ok, i can understand why admin wouldn't want a competing product advertised on his site.
with the 3A, is there a +5V pin somewhere for this, or what?
i really hope this becomes a product, it would be a great thing to have...
will it be open source? you releasing the files?
on a side note, is there any reason your file is called atmecha.pdf?
-
there are 2 +5v pins, 3 gnd pins, and 2 vin pins they are all down in the lower left. It was hard to make it obvious with the silkscreen labels. I used multiple pins because they are only rated 3A a piece, and I wanted a factor of safety.
-
do you have any plans for selling a shield for it?
and are the header pins male or female? in the eagle file they are female?
-
There are still so many advantages to yours.. go for it!! Very nice design!
-
There are still so many advantages to yours.. go for it!! Very nice design!
could you elaborate please, and would you be interested in purchasing one? By no means do you have to I just would like to get a feel for how many I could sell, since there are quite a few startup costs when starting a business.
Thanks for the input! ;D
To others don't feel afraid to bash my design. Its ok to not like it too. This is a great learning experience for me.
-
I'm currently not building.. but if I was, I would buy one.
-
I would ask Admin how many Axon's he sells.. that would give you somewhat of an idea of what the demand is like.
-
I would ask Admin how many Axon's he sells.. that would give you somewhat of an idea of what the demand is like.
not really... the axon would probably sell more because of the power bus and all the ads admin has everywhere...
-
I'm sure it would sell more.. but at least you have an idea of the demand. Does he sell 10 a year or a 100.. if only 10 then good chance not a good market.. if 100 then I'm sure it would be easy to sell more than 16. See what I mean?
-
true, and the price may bring in more people.
admin sells a fair amount of axons every year... i remember a while back he was looking for someone to package and post them for him... and maybe solder header pins in... i cant remember that well...
but this wouldn't be if there were only like 50 of them... and per year...
im thinking maybe near 1k axons...
-
sonictj good luck.. can't wait to see some pictures of the prototype!
-
sonictj good luck.. can't wait to see some pictures of the prototype!
yah I'll definately post them as soon as I have the board. I hope everything works electrically too! I spent 2+ months designing and redesigning/learning eagle. I tested my homemade prototype extensively so I'm almost certain everything will work, but I won't know until Tuesday. Wish me luck on that :P.
-
one big advantage i can see is the compatibility with axon code.
i can see this drawing in lots of newer, poor people who want the code...
how easy would it be to make a shield for this?
-
just buy any through hole plated proto board, that is big enough. Sparkfun sells the wombat. I think jameco has some cheap ones, and I know you can get some twin industries ones on mouser/digikey for ~16 or less I believe.
-
The board is definitely axon compatible. I think I could even load admins boot loader in place of the arduino one. I suggest going ispmkii though if you have experience. Its a lot less of a hassle in the long run.
-
but the axon uses the cp2102 and yours uses the ftdi chip right?
wouldn't that affect the bootloader software at all?
i gotta stop posting in this thread... my name is clogging it up :P
-
but the axon uses the cp2102 and yours uses the ftdi chip right?
wouldn't that affect the bootloader software at all?
It shouldn't the hardware operates in the same way. I've talked to admin a bit about compatibility, and it never came up as a possible issue.
-
To refocus the topic for new readers. Please look at my first post, on this thread, and tell me what you think of my design versus the seeduino mega. I need to know where I stand. I want to make my board available to the masses, but if there is not enough demand I will not be able to do so.
Your feedback is much appreciated ;D.
-
Firstly, congrats for such a well designed product. A lot of thought went into it and also a lot of struggle for making it Arduino compatible. I am an Arduino software user because of my lack of deep programming skills (I was a Bascom and NQC programmer before Arduino) so this board apeals to me for this reason. I build robots as a hobby so I have no use for Arduino style shields. Making my own shield is easy, so that would not hold me back for buying your board, but I would not get a seeeduino if I need all those extra pins that are so cluttered. For a factory made shield, that is fine, but for a home made one, it will be a pain and more than 2 hours of work, which I value at $20 per hour. I guess I could make a home made shield for your board in about an hour (just labour cost) so a $20-$30 price difference would set things straight.
So yeah, go for it, make it a product. You will sell more than 20 in a year, but you need to be patient. I made my Ro-Bot-X board 4 years ago an I just sold the last one from a batch of 10 boards (in a robot) a few weeks ago. When I made them, I wanted to make a robot kit to sell it in my country, but the market wasn't ready yet. Now it is, but for a product like yours, my board is kind of obsolete. Each board cost me $13 and some of them I sold, some of them I offered as presents for people interested in the hobby. I guess I sold a bit more than half of the boards and overall I didn't make a profit, other than helping people get interested in robotics.
Unfortunatelly, I have so many boards that I dont use that I don't need to purchase another right now. I have an Axon that I want to use for a biped robot (I'm held back by the servo costs) and thank you for making it Arduino compatible. However, if anytime I'll need another powerfull board for my projects, your board is the first on the list.
Thank you for all your past, present, and future efforts!
Edit: b.t.w. allways get the double sided through hole plated prototyping boards when making your shields! An if the pads are square instead of round that would make things even easier! Sparkfun has some...
-
If you bring it to the market before Christmas I think you will have a good chance at making it a success. If all you need to sell is 16 just take a shot at it! The key to this product will be advertising, nope not expensive adds all over the place, but just get the word out about the product. YouTube will be HUGE, start posting videos of projects that are possible with your board... make some sweet stuff if you can! what people will do is ask how to make that and then you could publish a couple tutorials on how to do some projects with your board.... Hopefully if all goes as planned, they will go off and buy your board... If not, well how many people use YouTube?
If you want to compete with that seeduino...
Take a look at the specs that they provide, is there anything that they have that will be possible to put into your board yet?
Get it to be sold on all of the big sites
-Trossen Robotics
-SparkFun
-Etc.
Personally go for it, If you don't sell 16, then well you learned from the experience what to do next time! right?
-
I uh might actually buy one if I sell my drumset as i plan to do... just gotta find a buyer
-
You guys are definitely starting to bring my confidence back. I'm still a bit hesitant due to this being around a $500 investment for me, and I don't have a lot of money. I know business is a gamble, but I'm trying to eliminate as much of that element as possible.
-
I'm not going to promise a purchase yet but if do purchase a board I'd take yours over the roboduino or anything arduino puts out.... as it is now my option iis basically the roboduino...though i will use arduinos software
-
I figured there arduino compatibility would be a plus I guess I was right about that. The arduino is how I started out just a year ago. It is a great platform.
-
the arduinos ok... i dont like the shield thing... let me know when you start selling...
-
let me know when you start selling...
will do if I go through with this.
-
*price I can go as low as $65 with an atmega640 on board, but I cannot beat $48 and profit
You aren't designing it for price properly, as I can definitely do your board cheaper than that. :P
Go back and redesign to reduce price.
You guys are definitely starting to bring my confidence back. I'm still a bit hesitant due to this being around a $500 investment for me, and I don't have a lot of money. I know business is a gamble, but I'm trying to eliminate as much of that element as possible.
Only $500?! ha!
Lets say it costs $35 per board to make. Then lets say you make a tiny batch of 100. Thats already $3500. :P
My problem is I have to sell 14 boards just to break even. If I can't sell that many there is no reason for me to sell any at all.
Experience. You'll learn a lot, even if it was a financial failure.
-
Lets say it costs $35 per board to make. Then lets say you make a tiny batch of 100. Thats already $3500. Tongue
not true if I make them to order. I plan on buying a ton of pcbs, but only a few parts at a time. This makes the total cost of a single board higher, but my potential financial loss is much smaller. I'm soldering stuff by hand not sending it out.
-
my strategy for building boards is to make as many that are needed. Meaning I estimate the demand and have enough on hand to sell. I don't want too much unsold capital sitting on the shelf because that costs money. I thinking small time here. I'm still a student, with no additional sources of income.
-
To Admin
I know it is possible to make the board cheaper even with the board design I have. It just requires a high risk to me financially which I cannot justify. I would have to pay huge upfront costs, that I cannot back. I aslo doubt the design would catch on enough for me to profit much. I see this as a huge learning experience, but I'd rather not risk breaking the bank.
I greatly appreciate your input. I was hoping I would get a bit of criticism too. Do you have any other suggestions?
-
Lets say it costs $35 per board to make. Then lets say you make a tiny batch of 100. Thats already $3500. :P
My problem is I have to sell 14 boards just to break even. If I can't sell that many there is no reason for me to sell any at all.
Experience. You'll learn a lot, even if it was a financial failure.
i would say a batch of 100 would be a bit much if you don't have the money, somewhere
around 20-50 boards would be better if you are tight on money.
Lets say it costs $35 per board to make. Then lets say you make a tiny batch of 100. Thats already $3500. Tongue
not true if I make them to order. I plan on buying a ton of pcbs, but only a few parts at a time. This makes the total cost of a single board higher, but my potential financial loss is much smaller. I'm soldering stuff by hand not sending it out.
that is a good plan because the pcbs will be cheaper the more you buy, so 100 will be dirt cheap, but try to go for price breaks on the parts for bulk buys, if you are only going to get 20 of a part that is ~$18 and there is a price break at 25 pieces that takes it down to ~$17 consider saving up a bit more to get the better deal.
-
but try to go for price breaks on the parts for bulk buys, if you are only going to get 20 of a part that is ~$18 and there is a price break at 25 pieces that takes it down to ~$17 consider saving up a bit more to get the better deal.
Thats the plan
-
I greatly appreciate your input. I was hoping I would get a bit of criticism too. Do you have any other suggestions?
Well, if you don't have confidence in your own product, thats a bad sign already :P
My suggestion is the same as before - to design an original product thats very different from those on the market, and that you have confidence in. When you have zero competition, you get 100% of the market. With just one competitor, you only get 50%, etc. There are quite a few established microcontroller products out there already. :P
Also, if you do a plot of cost vs # of units made, you'll find that anything below 100 units won't be very profitable. When I did the math for the Axon, I calculated I'd still break even given worst case scenario. So I took the risk . . .
(disclaimer - I'm financially motivated to scare away potential competition :P)
-
Well, if you don't have confidence in your own product, thats a bad sign already Tongue
I have a ton a confidence in my product, but I don't have a lot of confidence in the market ;D. I get your point.
Also, if you do a plot of cost vs # of units made, you'll find that anything below 100 units won't be very profitable. When I did the math for the Axon, I calculated I'd still break even given worst case scenario. So I took the risk . . .
I'm not looking for money really I just don't want to lose any. If I make a $100 whoo hoo, but I'm primarily doing this to say I have what it takes, and so I can put it on my resume.
I think of it as a design that I can sell cheaply to my friends, and the community, as a hobby not so much as a big money maker. To me 20 units is a huge success.
-
do you know what you are going to be calling it yet?
-
I thought ATmecha, to play on the ATmega series and because of the mechtronics applications.
-
cool!
well... I am going to add this in just for a little motivation,
As of October 2008 over 50,000 arduino boards have been sold. if you could capture (at least) .5% of these sales, you could possibly (in theory) sell over 250 boards, this is just based on statistics so actual results may vary...
(just thought it could boost your excitement a little) ;)
-
That is a good point
-
Level of difficulty:
1. getting a microcontroller to work
2. putting mcu on a PCB
3. getting people to buy your mcu
4. getting it cheap and easy to manufacture
5. mcu being a successful/profitable product
95% of all projects don't get past #1. I think you are aiming for #3? :P
-
yep I have #4 covered too I have access to a reflow oven and I can also get cheap stencils. And #5 is a result of 3.
-
yep I have #4 covered too I have access to a reflow oven and I can also get cheap stencils
Hand doing those boards won't be easy. If it took you say 35 minutes to solder, pack, program the bootloader/fuses, test for failures, accept orders, and mail each item, and did 100 units, thats 58 hours of work (an average worker does 40 hours/week). How much do you value your time? :P
When you get your first prototype, time how long it takes to do all the above. Its no small number! Basically, #4 means 'designed for manufacture'. One advantage of the Axon II over the original is that it'll take me about ~5+ minutes less to manufacture each one. A huge amount of time considering my sales :P
And #5 is a result of 3.
Not really. You can give up profits to improve sales. :P
If you sold 1000 units but made only $50 in profit, you failed #5. ;D
-
Hand doing those boards won't be easy. If it took you say 35 minutes to solder, pack, program the bootloader/fuses, test for failures, accept orders, and mail each item, and did 100 units, thats 58 hours of work (an average worker does 40 hours/week). How much do you value your time? Tongue
When you get your first prototype, time how long it takes to do all the above. Its no small number! Basically, #4 means 'designed for manufacture'. One advantage of the Axon II over the original is that it'll take me about ~5+ minutes less to manufacture each one. A huge amount of time considering my sales Tongue
I agree, but I just don't foresee selling enough for me to get too aggravated. I really love to solder I find it really relaxing. This whole thing is for fun. If the fun turns into hard resentful work I'd stop selling the product. Also if I had an really high demand I would look into outsourcing. You are definitely right though, its just something I'll have to feel out as I go.
And #5 is a result of 3.
Not really. You can give up profits to improve sales. Tongue
If you sold 1000 units but made only $50 in profit, you failed #5. Grin
agreed I don't plan on doing that lol.
-
i agree soldering is very relaxing, maybe when you start up just do a few boards a day, everyday and by the end of the week you will have ~15-20 boards done... although, if the demand is great enough maybe get an assembly company to do the work for you... although, if you are using a reflow oven, the soldering time may decrease a bit...
-
Try to market your product for the general public, not just robots/mech stuff. The Axon is pretty much only bought by robotics people because that's what it is advertised for. Axon may beat the Mega in some areas, but people who aren't very experienced may think "Oh thats for robotics, im not doing that so Ill get the other one"
-
Razor Concepts has a good point, it may drum up more sales that way ;D
-
I think you should still try to sell your board design because there are people out there who just can't solder. I find the soldering relaxing because I will start singing songs while I am soldering. I was singing the stupid armour hot dog theme the other day when I was soldering. There are a lot of newbies on this site and I know that they will be looking for something that is better than the $50 dollar board, and I think you have it. Some of us don't know how to design a board and may take a look at your board and think to themselves, "self, I made my $50 dollar board, but I don't know/want to make a better one from scratch, so I think I will spend the fitty and get this one." In a nutshell I think you should go for it. Like you said, your a student, so what else do you have to do besides drink beer and party???
-
I think you should still try to sell your board design because there are people out there who just can't solder. I find the soldering relaxing because I will start singing songs while I am soldering. I was singing the stupid armour hot dog theme the other day when I was soldering. There are a lot of newbies on this site and I know that they will be looking for something that is better than the $50 dollar board, and I think you have it. Some of us don't know how to design a board and may take a look at your board and think to themselves, "self, I made my $50 dollar board, but I don't know/want to make a better one from scratch, so I think I will spend the fitty and get this one." In a nutshell I think you should go for it. Like you said, your a student, so what else do you have to do besides drink beer and party???
Yes. I agree. I know basic electronics and can solder very good.. but I don't know much about mcus and I can't solder surface mount. At least sell your design.. if you don't want to build yourself. I think you did too much great work to just put on a shelf.
Tesla
-
Like you said, your a student, so what else do you have to do besides drink beer and party???
I wish lol.
-
Never soldered a smd mcu before, but I have an idea. Someone mentioned the stencils. I never saw them in real life to determine if they are good enough for my idea as material, but perhaps someone can find the proper material for it. So, here it is: would it work to have a sheet of thin material (think high temp resistant plastic) shaped like the board, with holes so you can properly align them, with a square cut to fit the mcu, with the sides like a comb to fit the space between the pins of the mcu. So you lay this plastic over the board, fill up the comb like sides with solder paste, place the mcu and solder the pins. Then, if there are still solder bridges, use solder wick to remove them. This should work fine, but there is one problem I can think: the comb like teeth will be so thin it will make them easily bendable and breakable...
-
you don't use a comb you us a squeegee.