Whew! It's the end of the day, and I got in a good 5 hours (possibly more!) in on this project today, and I now have it functioning great! I have a built in clock that timestamps when the package comes and prints the time to the display. I have also added a Parallax 5 posistion switch to the prototype board, and it's not that much extra to put on. It is used to set the clock and clear the register.
I've posted some pics of the current screen and updated board. Hope you like it!
I need your opinion on Ken's project construction. I am torn between creating a project that is cheap and simple (what I currently have) or one that is expensive and feature-packed. Does a mail-alert system need lots of features? Although a retrieving robot would be AWESOME, that tacks on a huge amount of money. And although this current system is nice, would it be worth it looking more impressive to have a big fancy system?
Tell me your opinion, I would be glad to hear it.
I would do what the customer asked and nothing more. If the customer wants more features, then you should discuss additional payment for your time and additional parts.
With that said, you can give the customer a very well tested and well working system. Something which works better than the customer's expectations.
Find some volunteers in your neighborhood and have them use the system. Have them set the time, etc. Can they easily figure out how to do this? Do they easily understand how to operate it?
*You* may easily understand its operation, but when you turn the controls over to the general public, interesting things happen! They can't figure out the most basic things!
Then how about weekends, holidays, and daylight savings time change? (Power saving mode when business is closed...)
Why not throw a little flair in the display? Make pending messages easily noticeable. Perhaps some silly comments based on how long the package has been waiting?
@Micro, Bill190 has the right advice. You can add some some minimal extra features, but don't go too far overboard - meet the "spec" even though it was written quite loosely. For example, you'll need a clock and it appears you already built one into the project (but I didn't specifically ask for one - not sure how you'd be able to turn the display on/off without one anyway). Looks like you're well underway. I'll check my e-mail for your message now.
(I just figured out who you are, you've been programming some good stuff and are like in middle or high school or something).
I'll tell you what, the Army won't let me hire you at your age but when my project gets going in Jan/Feb, it might be worth a little out of pocket to have you participate on some discrete aspect; assuming it's ok with your folks. We also do high school summer internships (in Maryland and maybe Calif). When the project concept is approved, I'll share it on the forum - three hints: 1. it is educational, 2. it will use more than 30 networked propellers total (not in one device) and 3. it might save lives.
Oh, Microcontrolled. It won't cost anything in parts to make the text display attractive and colorful. Go for minimum parts and an elegant, reliable solution in a quality case.
FYI - If you want to add a "wiz bang" power saving feature which keeps the current date/time, changes the time automatically for daylight savings time, and also calculates holidays, following is some helpful information...
First as to "business hours", there would be Parallax business hours and FedEx business hours. Parallax business hours might be difficult to figure out as I see posts here from Parallax people very early in the morning as close to midnight sometimes! Technical types don't always work 8 to 5.
FedEx can work late sometimes, but I don't know if they would drop something at a business after normal closing hours?
Then in large cities, different businesses will sometimes have "flex hours". This is to help with traffic congestion problems. So worth investigating. In some cases the "official business hours" might be 8-5, but the "shipping/receiving person" might work from 10-7 everyday. And the FedEx guy might know this and sometimes deliver packages late.
So it is possible to design a system, use the correct "business hours", then later find out that will not work for the person using the system! Good to get it straight from the horse (shipping/receiving person) if possible, then if something is different from what the "boss" said, discuss it with him so everybody will be in "sync".
And the people who ultimately use your system will be the people who give the boss feedback as to if they are happy with the system or not. So it is a good idea to involve them in system design if possible.
(We have all seen the cashier who can't figure out how to enter something or delete something and is cursing the person who designed the system. Nobody ever talked to the people who use the system 8 hours a day!)
As to keeping time, see page 62, "Timekeeping Applications" of the "Propeller Education Kit Labs" document. This discusses accurate time keeping with TimekeepingBad.spin and TimekeepingGood.spin.
Then for daylight saving time calculations, search google.com for the words...
algorithm daylight savings time
Note that businesses are not always closed for every holiday. And FedEx may or may not deliver on some holidays?
And businesses as well as the federal government will change their minds! They may decide to add a new holiday at some later point. (Just like they recently changed the daylight savings time dates.)
@micro: Another way to make a few bucks might be to get one of your designs sold through Gadget Gangster - you get all the fun of development and none of the hassle of selling :-)
Thanks to everyone for thier comments/suggestions/offers!
@Invent-O-Doc: Who told you? :-) Anyway though I would be happy to help if I can.
@simonl: I have a project on Gadget Gangster, and in well over 1 year I have made $5 off a $20 parts investment. Now a days you have to have professionally manufactured boards just to sell your product on the site anyway, so that's over $100 in down costs for something that will probably make $10 or less a year. I support Gadget Gangster, and I like and buy the development products, but I don't think it's for me.
@All: I am making out the order list today, and my parts should ship out soon. I will be upgrading the display to look more visual appealing, as I could change to a better text driver or maybe even have room for the graphics driver. Before that, though, I'm on the search for a nice pushbutton. Preferably one with a built-in LED, for the delivery side.
Before that, though, I'm on the search for a nice pushbutton. Preferably one with a built-in LED, for the delivery side.
I don't have any specific recommendations, but you want to make sure it's a high-quality switch that will be durable. It should probably also be rated for outdoor use.
One option - and I'm serious about this - is to pick up a doorbell switch from Home Depot.
From a quick look at the data sheet, it doesn't look like there is any way to mount that in a panel - it's a PCB mount. You can work with them, but for a one-shot application it will probably be much easier to work with something that can be mounted in a panel.
From a quick look at the data sheet, it doesn't look like there is any way to mount that in a panel - it's a PCB mount. You can work with them, but for a one-shot application it will probably be much easier to work with something that can be mounted in a panel.
My wisdom tells me you shouldn't mount a switch on a circuit board. You should have it on a panel. Pressing on the circuit board could weaken it and you should keep in mind or have an understanding that the sun is going to bleach parts that are outside and that incliment weather will play its role. You want something rugged to last but also not too expensive because it may have to be replaced. You would do good to look at some commercial intercom systems because they are built to last. No offense to Radio Shack but we kept having weather related issues and having drivers back into our intercom so our company had a professional intercom installed.
I haven't read through the specs again so I may have missed this before...
Is the switch going to be mounted in a box with all the electronics (at the transmitter end)?
Or, is the switch going to be mounted outside with wires running inside to the box (which is obviously more secure and less exposed to weather)?
In the latter case, you probably want the switch to be mounted in a panel in the wall (a hole in the wall, essentially) or you want to find a very short switch so you don't need a deep box.
From my understanding the switch will be mounted in the same case as the PCB, which is a Mountain King acrylic enclosure.
They are nice cases, but they are not weatherproof (or resistant). If the electronics at the door end are in one of those cases (which will probably look cool), I'd expect that the case would be inside and the switch would be separate.
I guess it's something to clarify with the customer.
I thought of that too, and if the customer wants it, I have a nice weatherproof case here that is used (has a few holes in the bottom for wires) but is still is nice condition. However due to size I may have to cut up to a centimeter off the bottom of the protoboard to get it to fit. Then again I could always just order a weatherproof case....
If you wantto mount a pushbutton in the weather there are standard industrial solutions for that. They're pricey but they work very, very well. You want to be looking at a Hoffman or similar enclosure which is reated NEMA IV for water exposure, a similarly rated switch (Hoffman and other companies sell boxes pre-drilled for the more common industrial control switches), and CGB glands on round wires for getting the signals out of the box without letting air and moisture in. This will survive rain and repeated sun/night cycles. (If the site is shielded from rain and direct sun, you can get away with a lot less hardening.)
I'm at home right now but I can scare up some links tomorrow if you're interested. I've been doing this sort of thing since 1985 or so.
Industiral boxes like this are made to receive a metal back plate onto which you would mount your electronics (you don't drill holes in the box because that would mess up the weatherproofing, and nobody expects the box to match what's going in it).
The punches and drills necessary to make the holes for the industrial switches and CGB's are a bit expensive; we have those at our shop and I'd be willing to do that part for you for free. Parallax may also have those tools.
I have to say after my original wet-blanket reply to you Microcontrolled after reading this thread I have never been so delighted to be proven wrong in my life. Good luck!
Also, don't make the insanely common mistake of installing a "blind pushbutton." Provide a light or something to let the driver know the button push was accepted, so he can alert you if it isn't working.
Good call there Roger. That will eliminate a lot of test cases.
User pushes button, has no clue:
pushes it again
holds the thing down at random
pushes it a lot of times quickly, slowly, etc...
fails to engage it on the first push
slow push, with long hold in "pushed" position
a tap, not quite a push, etc...
Best case is a latched, delay release where the "pushed well enough, thank you" signal is triggered, and that status displayed for a good, solid amount of time. Keep that light on for long enough that they get it, and move on, but not long enough to cause early wear of the bulb.
It's simply amazing the number of things people will consider doing when asked for input without some clear, immediate feedback.
My wisdom tells me you shouldn't mount a switch on a circuit board. You should have it on a panel. Pressing on the circuit board could weaken it and you should keep in mind or have an understanding that the sun is going to bleach parts that are outside and that incliment weather will play its role. You want something rugged to last but also not too expensive because it may have to be replaced. You would do good to look at some commercial intercom systems because they are built to last. No offense to Radio Shack but we kept having weather related issues and having drivers back into our intercom so our company had a professional intercom installed.
Good advice. Micro, the Prop Proto Board Enclosure could be used to mount the switch. Although this enclosure is not weather resistant, it will be placed in a dry location so it should be okay. Like I said in the PM, these guys know their stuff!
"pushes it a lot of times quickly, slowly, etc..."
That will be the pedestrian crossing request light then. Even when the "Wait" light is lit, people (bless them) still think that another 20 pushes will "just make sure", and if those pushes are done very fast and extremely hard, then those lights are bound to change earlier.
Comments
I've posted some pics of the current screen and updated board. Hope you like it!
Tell me your opinion, I would be glad to hear it.
With that said, you can give the customer a very well tested and well working system. Something which works better than the customer's expectations.
Find some volunteers in your neighborhood and have them use the system. Have them set the time, etc. Can they easily figure out how to do this? Do they easily understand how to operate it?
*You* may easily understand its operation, but when you turn the controls over to the general public, interesting things happen!
Then how about weekends, holidays, and daylight savings time change? (Power saving mode when business is closed...)
What happens if that occurs?
What happens if the FedEx guy presses the button two or three times? Or presses the button quickly 6 times?
(Testing...)
Why not throw a little flair in the display? Make pending messages easily noticeable. Perhaps some silly comments based on how long the package has been waiting?
(I just figured out who you are, you've been programming some good stuff and are like in middle or high school or something).
I'll tell you what, the Army won't let me hire you at your age but when my project gets going in Jan/Feb, it might be worth a little out of pocket to have you participate on some discrete aspect; assuming it's ok with your folks. We also do high school summer internships (in Maryland and maybe Calif). When the project concept is approved, I'll share it on the forum - three hints: 1. it is educational, 2. it will use more than 30 networked propellers total (not in one device) and 3. it might save lives.
Good luck on the FedEx arrival indicator thingy.
First as to "business hours", there would be Parallax business hours and FedEx business hours. Parallax business hours might be difficult to figure out as I see posts here from Parallax people very early in the morning as close to midnight sometimes! Technical types don't always work 8 to 5.
FedEx can work late sometimes, but I don't know if they would drop something at a business after normal closing hours?
Then in large cities, different businesses will sometimes have "flex hours". This is to help with traffic congestion problems. So worth investigating. In some cases the "official business hours" might be 8-5, but the "shipping/receiving person" might work from 10-7 everyday. And the FedEx guy might know this and sometimes deliver packages late.
So it is possible to design a system, use the correct "business hours", then later find out that will not work for the person using the system! Good to get it straight from the horse (shipping/receiving person) if possible, then if something is different from what the "boss" said, discuss it with him so everybody will be in "sync".
And the people who ultimately use your system will be the people who give the boss feedback as to if they are happy with the system or not. So it is a good idea to involve them in system design if possible.
(We have all seen the cashier who can't figure out how to enter something or delete something and is cursing the person who designed the system. Nobody ever talked to the people who use the system 8 hours a day!)
As to keeping time, see page 62, "Timekeeping Applications" of the "Propeller Education Kit Labs" document. This discusses accurate time keeping with TimekeepingBad.spin and TimekeepingGood.spin.
Then for daylight saving time calculations, search google.com for the words...
algorithm daylight savings time
For holidays...
holiday algorithm
Calendar...
calendar algorithm
And Federal Holidays in the U.S...
http://en.wikipedia.org/wiki/Federal_holidays_in_the_United_States
Note that businesses are not always closed for every holiday. And FedEx may or may not deliver on some holidays?
And businesses as well as the federal government will change their minds! They may decide to add a new holiday at some later point. (Just like they recently changed the daylight savings time dates.)
@Invent-O-Doc: Who told you? :-) Anyway though I would be happy to help if I can.
@simonl: I have a project on Gadget Gangster, and in well over 1 year I have made $5 off a $20 parts investment. Now a days you have to have professionally manufactured boards just to sell your product on the site anyway, so that's over $100 in down costs for something that will probably make $10 or less a year. I support Gadget Gangster, and I like and buy the development products, but I don't think it's for me.
@All: I am making out the order list today, and my parts should ship out soon. I will be upgrading the display to look more visual appealing, as I could change to a better text driver or maybe even have room for the graphics driver. Before that, though, I'm on the search for a nice pushbutton. Preferably one with a built-in LED, for the delivery side.
One option - and I'm serious about this - is to pick up a doorbell switch from Home Depot.
but the doorbell button is likely to do better in the weather.
OBC
http://www.magneticlocks.net/products.asp?cat=19
See the button in that picture? That's an arcade button. To be more specific, YOUR arcade button. :-)
I've been searching the web for something more durable and weather proof, though. I'll look at that site you linked to.
http://www.e-switch.com/entity/tabid/95/entityname/category/categoryid/14/sename/anti-vandal/default.aspx
http://www.mouser.com/ProductDetail/E-Switch/LP4OA1PBATR/?qs=sGAEpiMZZMvxtGF7dlGNpqqmTFthXBmfKA6hYk5DC6Q%3d
http://www.mouser.com/ProductDetail/Bulgin/MP0042-2/?qs=TJSU9qitBVQzNOeT2EcPhQ%3d%3d
This is not vandel-proof, but the location this will be in is not likely to vandels.
My wisdom tells me you shouldn't mount a switch on a circuit board. You should have it on a panel. Pressing on the circuit board could weaken it and you should keep in mind or have an understanding that the sun is going to bleach parts that are outside and that incliment weather will play its role. You want something rugged to last but also not too expensive because it may have to be replaced. You would do good to look at some commercial intercom systems because they are built to last. No offense to Radio Shack but we kept having weather related issues and having drivers back into our intercom so our company had a professional intercom installed.
http://www.mouser.com/ProductDetail/Schurter/124166211120000/?qs=VO2/VXZcKR0q6Y2q9v1PLg==&tab=Notes
I'm not going to use any that are PCB mounted, for the reasons that you said.
Is the switch going to be mounted in a box with all the electronics (at the transmitter end)?
Or, is the switch going to be mounted outside with wires running inside to the box (which is obviously more secure and less exposed to weather)?
In the latter case, you probably want the switch to be mounted in a panel in the wall (a hole in the wall, essentially) or you want to find a very short switch so you don't need a deep box.
They are nice cases, but they are not weatherproof (or resistant). If the electronics at the door end are in one of those cases (which will probably look cool), I'd expect that the case would be inside and the switch would be separate.
I guess it's something to clarify with the customer.
I'm at home right now but I can scare up some links tomorrow if you're interested. I've been doing this sort of thing since 1985 or so.
Industiral boxes like this are made to receive a metal back plate onto which you would mount your electronics (you don't drill holes in the box because that would mess up the weatherproofing, and nobody expects the box to match what's going in it).
The punches and drills necessary to make the holes for the industrial switches and CGB's are a bit expensive; we have those at our shop and I'd be willing to do that part for you for free. Parallax may also have those tools.
I have to say after my original wet-blanket reply to you Microcontrolled after reading this thread I have never been so delighted to be proven wrong in my life. Good luck!
User pushes button, has no clue:
pushes it again
holds the thing down at random
pushes it a lot of times quickly, slowly, etc...
fails to engage it on the first push
slow push, with long hold in "pushed" position
a tap, not quite a push, etc...
Best case is a latched, delay release where the "pushed well enough, thank you" signal is triggered, and that status displayed for a good, solid amount of time. Keep that light on for long enough that they get it, and move on, but not long enough to cause early wear of the bulb.
It's simply amazing the number of things people will consider doing when asked for input without some clear, immediate feedback.
Good advice. Micro, the Prop Proto Board Enclosure could be used to mount the switch. Although this enclosure is not weather resistant, it will be placed in a dry location so it should be okay. Like I said in the PM, these guys know their stuff!
That will be the pedestrian crossing request light then. Even when the "Wait" light is lit, people (bless them) still think that another 20 pushes will "just make sure", and if those pushes are done very fast and extremely hard, then those lights are bound to change earlier.