Just got my Irtoy and was a bit disappointing, wonder what's up with seeed, the quality isn't as good as usual. Replacing the IR led I can live with but my resistor values are all wack, R4 for example is 390 Ohm so I'm getting a whopping 7mA through the transmitter. I'm swapping everything out now to get it working.
[EDIT]
Replaced 3 resistors, R2, R4 and R5 and everything works now. This is a pretty big mistake if you ask me, not everyone has a SMD resistor book and hot air station handy.
I guess it has to do with moving into their new office. I also had a couple of issues with my latest order with them (two weeks ago).
But they say that they will solve it all and send replacements (haven't received the replacement parts yet, but I'm feeling confident)
Getting R4 wrong was a pretty big mistake, the range isn't great as it is and they effectively cut it in half. I wonder how many people have this same problem and don't even know it, as I have it R4 should be about 150 which is pretty close to the original design.
My IR toy also has the same problem as s3c .What is the point of ordering a "professionally assembled" board if it is defective anyway and needs to be fixed by hand.The components are bent out of position and are crooked - Soldered that way not bent during shipping.Looks like someone was in a hurry and also hand soldered the components ???.
I have a 390Ω for R4 as well... I thought the range seemed a little short. I don't have any smd resistors and I don't think I have the soldering skills to replace that one without seriously risking the board anyways... Any suggestions?
[quote author="DTR2"]The components are bent out of position and are crooked - Soldered that way not bent during shipping.Looks like someone was in a hurry and also hand soldered the components ???.[/quote]
My impression is that this sort of thing depends upon the quality of the pick&place machine. I design electronics, and one of my clients decided to purchase a pick&place with no experience - not unexpectedly, the quality of the boards they produce is less than others who have better equipment (and more experience).
But you raise a good point. I would like to know whether this is a mistake, par for the quality and cost, or even expected.
They must have switched manufacturers or personnel or something then because the bus pirate builds were excellent. If this happens again they're losing my support.
Please don't blame this on Seeed. I limited the LED to 10mA in the prototype, that partlist went to Seeed for the initial quote. I updated the value on the Eagle files I sent for production, but they use the quote partlist. I should have sent an updated quote partlist with production values. I'm really sorry. This is the same reason they used that funky shielded LED instead of a cheap 'normal' IR LED, I sent an Eagle-exported bill of materials for the quote and it happened to list that footprint (which was incorrect), so that's what they used.
I will verify the production partslist and each part number for every project in the future to make sure it meets my requirements exactly. I will also make sure that only the values from the article and/or final PCB files are used.
There have been lots of little production problems in the first six months of the Dangerous Prototypes experiment, but this is the first time I personally put a mistake on a board into production. I can't tell you have devastated I am. First with the backwards LED (show stopper bug!), and now with the 10mA IR LED. I am going to do what I can to make it right. I will do repairs and send replacement parts, let PM or email.
I totally understand if anyone doesn't want to buy my designs in the future, and I would totally support that decision. I do my best to design an interesting open source hardware project with hardware availability every month. I try to keep the prices low to where the parts "already stuck to the board" are less than buying it yourself in 1sies. After all, what fun is open source hardware if you're the only one using and developing it? I strive for perfection, but my designs are not commercial consumer products, and that often shows. Every design is a 'work in progress', and I understand how frustrating that can be to a user.
As the deadlines get tighter, and the orders bigger, everyone's messing up. I didn't do a project last month because both the OLS and the IR Toy were pending. I may also skip this month because we're going to try to have projects production ready before publishing them in the future.
Again, my sincerest apologies for the defects. I'm happy to make repairs and/or send replacement parts. I'll reimburse your postage costs by paypal (up to $3, first-class mail only please). I will do my best to manage the projects better in the future, and I will take more control over production part selection.
Wow, now that's an apology, after hearing the reasons for this mishap I can't be to annoyed, this sound like the kind of thing that would happen to me as well. No worries Ian, sorry for my rant. The reverse LED is still interesting though, before replacing it I looked at it through my camera and it did light up dimly, could it be that some of the LEDs are in fact correct but the R4 mishap made it appear otherwise?
As for having the production done when the project is released, how do you feel about some community collaboration before starting production as was done with the Logic Sniffer? The IrToy for example could have had a couple of awesome features that were suggested after production started which wouldn't really have increased the cost.
Well I for one appreciate everything what you and the rest of the people involved have done and I fully understand mistakes happen so if you need help with covering the expenses that you might incur because of the situation or need help with anything then I will be more then happy to help out with what I can. Hope everything works out for everyone.
By the way are all IR Toys defective or just some?
[quote author="Ytsirk"]
I have a 390Ω for R4 as well... I thought the range seemed a little short. I don't have any smd resistors and I don't think I have the soldering skills to replace that one without seriously risking the board anyways... Any suggestions?
[/quote]
You can either pop off R4 with two soldering Irons, holding both ends and lifting it off the board and then replacing it or solder a 330 or 360 on top of if to get a total resistance of about 150 Ohm. Most decent suppliers stock SMD resistors but use some extra flux and don't solder directly onto the board but onto the Iron tip instead. The other two resistors (R2 and R5) aren't crucial and you can leave those as they are.
All the first batch of IR toys had reversed LEDs, you can read more here. It really was a bad situation. Bottom line is Eagle silk was wrong, the assemblers relied on that instead of the schematic. I didn't verify and I should have, I didn't realize how much they used the partlist and silk instead of the final schematic and board files. I should have at least removed that SFH reference from the eagle BOM and replaced it with '5mm LED':
http://dangerousprototypes.com/2010/04/ ... -1-defect/ (http://dangerousprototypes.com/2010/04/05/usb-ir-toy-preorder-1-defect/)
The self-test actually did catch the LED problem, but the wrong solution was implemented. The IR LED is visible when viewed through a camera, it's probably leading current or a small difference in ground values that causes it.
The R value is slightly different because if that were the only problem I think overall the situation it would be less terrible. Reduced range is unfortunate, but DOA hardware is embarrassing.
I am trying to do more development in the forums, and I will continue to do more. There will always be some projects I want to develop in private, but I want to open up development as much as possible. Also, I'm going to start making blog posts of the production preparation and quality control stuff so that is a more open process.
Hey Ian...
I'd like to say that I feel privileged to be on the ground floor of such a cool little project. Anyone who thinks there are not going to be problems with a first generation open source design is, at best, simply naive. You have demonstrated yourself to be a decent and generous guy to offer to work out and fix these issues and have no doubt that v2 will be even better. I have already found the IR Toy to be useful educationally, and will continue to enjoy playing around with it and hopefully contributing something useful back to the project.
No one likes to get something that doesn't work, but the bugs will get worked out...
"This open source hardware and software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. If you can't accept this risk, please do not buy this hardware. "
I got 3 USB IR Toys
preorder 1 - all had a 390Ω R4 and the IR LED was installed in reverse.
Ian's apology is very acceptable to me and his offer to repair or send replacement parts is generous, considering the warning in the USB IR Toy product description on Seeeds web shop (http://http://www.seeedstudio.com/depot/usb-infrared-toy-p-607.html?cPath=61_68) Ytsirk quoted in his post.
Having the right tools and parts handy it took me about 10 minutes or so to fix my 3 USB IR Toys and they work just fine.
Things like this can and and do happen with pre-orders (= pre-production runs), regardless how careful everything has been validated and checked before.
The USB IR Toy can't really be compared with the SUMP LA ([Pre-order] Open Workbench Logic Sniffer) project
1) Seeed has not shipped any [Pre-order] Open Workbench Logic Sniffer yet!
2) the project was a cooperation of 2 developers (Jack & Ian) and was supported by at least 3 persons who built prototypes from parts provided together with the PCB (I think robots got the PCB only for his contribution) by Ian and who tested them independently.
3) the SUMP LA project was far more complex
4) the name "USB IR Toy" implies that you can/must play around with it (to get it to do what you want it to do) ... :D
... can't wait for the Open Workbench Logic Sniffer pre-orders to arrive ...
This "mishap" with the USB IR Toy will not stop me from ordering more of Ian's projects as pre-orders at Seeed in the future ... quite the oposite. :)
In one point I am a bit confused. First you say:
... because we're going to try to have projects production ready before publishing them in the future.
then:
... I am trying to do more development in the forums, and I will continue to do more. There will always be some projects I want to develop in private, but I want to open up development as much as possible. Also, I'm going to start making blog posts of the production preparation and quality control stuff so that is a more open process.
Sounds a bit like a contradiction to me. I can see reasons not to publish projects before they are production ready ... but it might help a lot if you had a number of (experienced) beta testers who would get a chance to take a look at the design, make suggestions, validate and test them (i.e. breadboard or prototype kits like with the SUMP LA) or even contribute before the design will be sent off for production ...
Your open source projects give everyone full insight into the process of developing, designing, manufacturing and using advanced electronics! No one likes errors like the ones we have experienced with the USB IR Toy but they are part of almost every project - regardless if it's a commercial or experimental project. By publishing your projects and making the whole process including errors and mishaps that happen along the line available to the public many get a chance to learn from you ... including what can go wrong and how to deal properly with such a situation!
Thanks to everyone for understanding and the kind words.
Sorry about the contradiction. What I mean is that, even for projects developed 100% in public forums, I'm going to start waiting until Seeed is 'ready to go' before publishing the actual article and opening the preorder. For the OLS that would be after the production prototype is constructed and a test-plan is in place, then the lead times should be minimal, and major production issues will be worked out before the clock is ticking.
Should the problem with the value for R4 overly limiting the IR LED current be mentioned in the blog posts for others?
- http://dangerousprototypes.com/2010/04/05/usb-ir-toy-preorder-1-defect/ (http://http://dangerousprototypes.com/2010/04/05/usb-ir-toy-preorder-1-defect/) - USR IR Toy pre-order 1 defect
- http://dangerousprototypes.com/usb-ir-toy-manual/ (http://http://dangerousprototypes.com/usb-ir-toy-manual/) - USB IR Toy Manual
I'm not very confident in my SMD skills to try to replace R4 without damaging something else. Ian, what country are you in (if I decide to ship it to you for your generous offer)?
Thanks,
--Rob
[quote author="ian"]
All the first batch of IR toys had reversed LEDs ...
[/quote]
Well my IR Toy is from the first batch and I therefore need a new one, if I make a purchase from Seeed now will the IR Toy I received both have the IR LED correctly installed and the correct value for the R4?
Hey rhyvu - If you send me your batch 1 IR Toy I'll fix everything on it for you, no need to buy another.
Ian
[quote author="ian"]
Hey rhyvu - If you send me your batch 1 IR Toy I'll fix everything on it for you, no need to buy another.
Ian
[/quote]
I thought that if we purchase from Seeed we support you and the open source development? If I send it to you then instead of supporting you I end up costing you time and resources that could be used on better things. I would rather keep my current IR Toy as a sort of Inverted Jenny (http://http://en.wikipedia.org/wiki/Inverted_Jenny) (although not as rare) since I doubt we will be seeing many errors made by Dangerous Prototypes in the future. So to sum it up I would gladly purchase another working IR Toy, I just need to know if I make a second purchase it will have the issues resolved?
That's true, but I feel terrible about this mistake and I'll do everything I can to make it right. It's not a problem. Everyone has been so understanding and kind about it, it would be my pleasure to fix it for you to say thank you for the support and patience.
The next batch of IR toys will have the correct LED (regular not sht-41) and LED orientation. I'm not sure the request for the resistor change was in soon enough to beat it to manufacturing.