|
Where could I source another LED like this? I don't know the correct terminology to search for it. I know is it's a 1/2 W red LED (that is now broken after dropping my bicycle light many times). If I find another one, would I need to do anything besides (attempt) to solder it in?  Thanks
|
#
?
Jun 2, 2009 16:27
|
|
|
|
| # ? May 21, 2024 14:27 |
|
|
clredwolf posted:It's pretty sound, keep in mind your motors aren't going to be 100% efficient. Usually it's rated, so look at it, factor that in. If it's 80% efficient, it'll take 12 kilojoules over a second to make 10 kilowatts of output. Also batteries deteriorate over time, so you'll get less and less runtime. Tractor's still going to be light compared to stock (unless I add bank #2) so weight isn't really an issue. The transmission is geared so low I should be ok for the power of the motor (I hope). According to the brochure at the store, these batteries are good for 300 charge/discharge cycles with a 100% depth of discharge, which I never intend to do, so hopefully they'll last for 1000 uses or so. Running 2 banks in parallel, do you think that would be better than having 2 separate banks that I can flip between? How about charging, I have a 20 amp 48v charger, should I charge the two banks as a single unit, or individually? Thanks for checking my work, I appreciate it. 
|
|
#
?
Jun 2, 2009 17:20
|
|
|
thylacine posted:Where could I source another LED like this? I don't know the correct terminology to search for it. I know is it's a 1/2 W red LED (that is now broken after dropping my bicycle light many times). If I find another one, would I need to do anything besides (attempt) to solder it in? That's a weird LED, I'm guessing it's one of these: http://www.unique-leds.com/index.php?target=categories&category_id=159 http://www.futurlec.com/LED/LEDFLUXRED.shtml The light has a low power mode and high power mode right?
|
|
#
?
Jun 2, 2009 22:53
|
|
|
thylacine posted:Where could I source another LED like this? I don't know the correct terminology to search for it. I know is it's a 1/2 W red LED (that is now broken after dropping my bicycle light many times). If I find another one, would I need to do anything besides (attempt) to solder it in? Looks like a PLCC package to me: http://www.mouser.com/Search/ProductDetail.aspx?qs=sGAEpiMZZMuus4pmcqQnzirNWwQCfkT5DbY3pR2W7OU%3d Something like that. Hey, even gives you the chance to go for a nifty blue light if you wanted to 
|
|
#
?
Jun 3, 2009 01:04
|
|
|
yeah, definitely a plcc or some derivative of such. They're reasonably common.
|
|
#
?
Jun 3, 2009 02:12
|
|
|
ANIME AKBAR posted:yeah, definitely a plcc or some derivative of such. They're reasonably common. If you have a multimeter, make sure the anode/cathodes are in the same place, and if you have a micrometer, make sure they are the same size. If that is the case, you should be good - LEDs are pretty flexible as far as driving currents and stuff. If the packaging is the same, they should be interchangeable.
|
|
#
?
Jun 3, 2009 04:09
|
|
|
Delta-Wye posted:If you have a multimeter, make sure the anode/cathodes are in the same place, and if you have a micrometer, make sure they are the same size. If that is the case, you should be good - LEDs are pretty flexible as far as driving currents and stuff. If the packaging is the same, they should be interchangeable. Rad, I have access to a micrometer and I vaguely know how to use a multimeter - I'm sure the internet will tell me how to figure the rest out. Thanks electronics thread bros!
|
|
#
?
Jun 3, 2009 06:39
|
|
|
The other thing is that your batteries you are looking at will not have the capacity your after.. if it is a 12V 100Ah battery (1200 watt-hours) you will only get 1200 watt hours out of it IF and ONLY if you discharge it slowly and evenly over a priod of say 24 -48 hours. If your calculating on draining that battery in 2 hours what you will actually get out of it is an hour and maybe 15 mintues ( at most). Its like a car, if you drive slowly your fuel economy is higher... Ah is the size of your fuel tank ... Which is also why you connect all packs in parallel and use them at once, if you have 2 '2 hour' rated packs you run them one at a time you would get maybe 2 and a half hours out of it, but if you run them in parallel you might get 3 hours out of the 4. Also NEVER discharge batteries fully, even if rated to '100%' depth of discharge, I wouldnt run a lead acid pack like that below say 60% Also if your connecting multiple cells in series some of them can have higher voltages than others, after a few cycles you can get to the point where some cells are completely empty and others still nearly full, after that the rest of your pack drives one or more cells in reverse, essentially destroying the pack, make sure your charger has a 'balancing' function, it charges the pack as a whole at high current, then when nearly full it charges each individual battery slowly to top it off and match voltages of all packs. Also i hope you wont be connecting the motor directly to the battery pack wihtout some sort of driver or regulator otherwise its going to be full speed or nothing. And maybe add some of those large 2 farad capacitors as well to your pack As well as temp probes/ current etc.. Find a way to limit inrush current when you turn your motor on, you could damage your pack otherwise/ other stuff conencted to it.
|
|
#
?
Jun 4, 2009 07:05
|
|
|
Bush Ant posted:The other thing is that your batteries you are looking at will not have the capacity your after.. I've got a 400 amp motor controller:  The motor and controller are 48v, so I have to have 4 12v batteries in series to form a single power pack. Are you saying I would be better off with 2x 48v packs connected to the controller in parallel than 2 separate packs that I can switch on and off with a toggle? Also I have a 48v 20amp charger:  Should the bank not self balance? since the 4 batteries are all in series the cells are all connected and should be mostly in sync with each other in terms of balance. Right? I don't intend to ever run these batteries dead. What would you suggest as a limiter for the initial starting current? I've never heard of a component that would do that. Thanks for the advice!
|
|
#
?
Jun 4, 2009 07:27
|
|
|
That controller should limit inrush current (unless its a POS). http://energy.ece.illinois.edu/Balog/images/intelec02.pdf This will give oyu an overview on cell balancing and the benefits of greater capacity over time and longer life of the pack, but basically you want to charge all cells and then make sure the voltage of each cell is within 0.01V of the other cells. Also a formula for calculating the time your pack will last. t = h(c/ih)^k Where t is time the pack will last, h is the hour rating of the pack (usually 20 hours or so, basically how long a period of time the battery was discharged to get its rating) c is capacity (at h rate) i is current and K is a constant which is about 1.2 (up to 1.3 for older packs) So say you have two 150Ah packs For one pack (roughly 80 amps for oyur motor) t = 1.68 hours or 1 hour 9 minutes x 2 packs (switching when first goes dead) is 2 hours 18 minutes at full load. For two packs in parallel t = 2.68 hours. Is about 2 hours and 40 minutes at full load See how you gained 20 minutes by connecting the packs in parallel instead of running one then the other? also connect each battery pair in parallel, not a string of 48 and another string of 48 (or just get batteries with double the rating or whatever, jsut conect them up as one big pack the cheapest way you can. Also if you run the motor at a constant low speed versus full power then idle then full again you will see a bigger increase in running time, e.g. Half the time its at 100% throttle and other half its just sitting there doing nothing you would get 5 hours 20 but if you ran it continuously at half power you would get 6 hours 10 minutes usage. The banks will not self balance, I woudl get an accurate voltmeter and a specific gravity tester (if these are wet cells) that will allow you to test the batteries, ideally you would charge all at once and then top off any cells that are a bit behind ( if you understand what I mean) with a 6 or 12 volt charger (depending on your battery) NOTE ideally you would do this with each cell in the battery, so 48v pack has 24 series cells but as I said you dont want to be discharging the thing 100% each time you use it becuase the pack wont last a month. 60% is about what you want, a car battery doesnt get discharged below about 95% during normal use and lasts years, as after the car starts it is immediately charged back up. So multiply those times by about 0.6 if you want the battery pack to last.... The reason you are realising that this thing wont run for as long as you thought is because the energy density of hydrocarbons is an order of magnitude better than batteries EDIT: buying the charger before deciding what pack you are going to build wasnt the smartest idea, too fast a charge (greater than say 8 hours or so) reduces battery life
|
|
#
?
Jun 4, 2009 11:29
|
|
|
The point of what the guy above me is saying, is that by putting both banks in parallel (the exact series-parallel combination used to be determined by charging requirements, which I do not know a lot about), you're reducing the equivalent series resistance of the bank, wasting less energy. So if you're going to have two banks, put them both in.
|
|
#
?
Jun 4, 2009 12:03
|
|
|
Bush Ant posted:The reason you are realising that this thing wont run for as long as you thought is because the energy density of hydrocarbons is an order of magnitude better than batteries I'm well aware that the storage capacity of batteries are far inferior to gas or diesel. There are many disadvantages to electric vehicles, and especially to DIY ones like this where I have no access to a full engineering lab. Also, the charger is configurable, I can change it for use with wet cell, agm, gel cell, and other settings too. I knew I was going to use a 48v battery pack, since my motor takes 48, and this one fit the bill. And is a 20 amp charge rate really too fast? I mean the banks are going to have 416 AH each, at a 20 amp charge rate, at 100% efficiency it would take 21 hours to completely charge a single bank. I know it will never be 100% efficient, so it'll probably take 2 full days for a complete charge for a single bank of 4 batteries, right? Thanks catbread, as soon as I get another set of batteries, I will connect them in parallel. It'll save me from buying another contactor too.
|
|
#
?
Jun 4, 2009 17:49
|
|
|
I thought you said 150 amp hours for the pack previously? If its 416 amp hours then a 20 amp charger should be fine. Get a decent multimeter and if you see that the voltages between batteries differes by more than about 10mV get a small capacity 6 or 12v charger (possibly for motorbike batteries or something) as a top up charger and test the pack maybe once a month or so (and top up any batteries that are a bit behind)
|
|
#
?
Jun 5, 2009 03:02
|
|
|
My first PIC project is up and running (well, programmable/debuggable). DSPIC33F running at 40 MIPS. It's for my inverter control project. Using the ICD 2. I got the board milled, it came out really nice. Still waiting on an FTDI USB chip which will sit in the bottom left. 
|
|
#
?
Jun 6, 2009 23:45
|
|
|
catbread.jpg posted:My first PIC project is up and running (well, programmable/debuggable). DSPIC33F running at 40 MIPS. It's for my inverter control project. Using the ICD 2. So what does it do? I looked through your history in the thread but I don't think I saw you explain it. It looks pretty hardcore for a first PIC project, mine was just a digital thermometer with an LM35 and 4 7-segment displays. Right now I'm working on USB firmware for a PIC18F4550. I know that Microchip provides their own, but I thought I'd learn a whole lot more if I wrote it myself. It's nearly complete, it successfully enumerates as an HID keyboard and it has LEDs to indicate caps lock and such. Next step is to hook up a 4x4 keypad and have it send keycodes for button presses on that.
|
|
#
?
Jun 7, 2009 00:01
|
|
|
Oh, it's going to implement a modified space vector modulation control scheme for a three phase inverter. 3 analogue inputs for feedback from a signal processing system to sense harmonic current levels and thus half-wave core saturation, 3 PWM outputs, 3 switches and 3 pots for control and real-time control variable tweaking. USB bridge for debugging/anything cool I want to do with it. The reconfigurable digital peripheral pins have really come in handy for this board, that was a lot to try and connect to a 28-pin package. Writing USB firmware is a cool project. I see that PIC has USB hardware, maybe when you're done with that you can try and emulate it with bit-banging GPIO!
|
|
#
?
Jun 7, 2009 00:54
|
|
|
catbread.jpg posted:I see that PIC has USB hardware, maybe when you're done with that you can try and emulate it with bit-banging GPIO! It's hard enough when the PIC is doing the signalling stuff for you 
|
|
#
?
Jun 7, 2009 02:29
|
|
|
Bush Ant posted:I thought you said 150 amp hours for the pack previously? 150 amp hours per battery if I went with 6 volts, the 12v were 104 each. And I will, thank you.
|
|
#
?
Jun 8, 2009 04:12
|
|
|
catbread.jpg posted:My first PIC project is up and running (well, programmable/debuggable). DSPIC33F running at 40 MIPS. It's for my inverter control project. Using the ICD 2. Nice! Where did you get your board milled and what did it cost? I just used bactchPCB for the first time and while I like their prices having to wait a month to get the board was terrible.
|
|
#
?
Jun 8, 2009 15:43
|
|
|
Slung Blade posted:150 amp hours per battery if I went with 6 volts, the 12v were 104 each. um I think your missing something here If you have four 12v 104 amp hour batteries, and connect them in series, you will have a 48v 104 amp hour pack, not a 48 v 416 amp hour pack.
|
|
#
?
Jun 9, 2009 07:21
|
|
|
Keebler posted:Nice! my university, for free 
|
|
#
?
Jun 9, 2009 08:16
|
|
|
Does anybody know an electronics hobbyist supply store in Boston? I can't find one anywhere. I used to go to Radar Electric in downtown Seattle, that was one hell of a store. My dad took me there all the time as a kid, I had such a blast looking at big bins of vacuum tubes, motors, photovoltaics, etc. Too bad they closed.
|
|
#
?
Jun 9, 2009 20:43
|
|
|
Just an update, I got that FPGA software for the crazy ADC working. Half of it was replacing the clock, the other half was asking another engineer very nicely for some help. Thanks everyone who helped me out!
|
|
#
?
Jun 11, 2009 01:50
|
|
|
I just got some PCBs printed for the first time from Goldphoenix. They quoted me 18 pieces for a 3.66"x1.35" PCB and sent me 30 pieces on two panels. They were uncut (but are scored) so now I have to figure out how to separate them properly. Ordered the parts I need from Mouser yesterday and it got here via UPS 15 minutes after the FedEx guy with my boards. I'm impressed by Goldphoenix's turnaround time and Mouser's shipping time. Now to separate and assemble one of these boards and hope that my design wasn't a total failure.
|
|
#
?
Jun 11, 2009 17:23
|
|
|
Kire posted:Does anybody know an electronics hobbyist supply store in Boston? I can't find one anywhere. http://youdoitelectronics.com/index.htm This place looks pretty awesome.
|
|
#
?
Jun 11, 2009 17:43
|
|
|
Scarboy posted:I just got some PCBs printed for the first time from Goldphoenix. They quoted me 18 pieces for a 3.66"x1.35" PCB and sent me 30 pieces on two panels. They were uncut (but are scored) so now I have to figure out how to separate them properly. Ordered the parts I need from Mouser yesterday and it got here via UPS 15 minutes after the FedEx guy with my boards. I'm impressed by Goldphoenix's turnaround time and Mouser's shipping time. I've never ordered from Mouser because their shipping costs to Canada is kind of high, but I am extremely impressed with Digikey. For only 8 dollars they ship it from the US to my door in 2 days with no customs bullshit.
|
|
#
?
Jun 11, 2009 18:22
|
|
|
BattleMaster posted:I've never ordered from Mouser because their shipping costs to Canada is kind of high, but I am extremely impressed with Digikey. For only 8 dollars they ship it from the US to my door in 2 days with no customs bullshit. I usually order from Digikey, but I needed these 9 pin SIP bussed resistor networks that Digikey doesn't keep in stock. Shipping was $20 for 1-2 day UPS to Canada. I would have gone for their $8 USPS shipping but I wanted these parts before the weekend. They claimed to not have any customs/fees with UPS/FedEx and it turns out there were none.
|
|
#
?
Jun 11, 2009 18:41
|
|
|
Does anyone have any source for cheap IR pass filter material?
|
|
|
#
?
Jun 11, 2009 19:09
|
|
|
Tmavomodry posted:Does anyone have any source for cheap IR pass filter material? Exposed developed film negatives. Edit: apparently two layers give good results: http://www.davesastro.co.uk/equipment/irpass.html Mill Town fucked around with this message at 20:04 on Jun 11, 2009 |
|
#
?
Jun 11, 2009 20:00
|
|
|
BattleMaster posted:I've never ordered from Mouser because their shipping costs to Canada is kind of high, but I am extremely impressed with Digikey. For only 8 dollars they ship it from the US to my door in 2 days with no customs bullshit. mouser is top tier if you are in texas. I order from them all the time, I usually pay the lowest FedEx rate and get the product the next day.
|
|
#
?
Jun 11, 2009 21:00
|
|
|
Okay, question: I just started a part time internship at a company, and apparently they want their electronics to utilize PIC microcontrollers. I can work with them, but it's been a while, so I'm not sure what models to use. To start, I want something very simple, small, and very low power. I only need a few I/Os, and it doesn't have to be fast (32KHz would probably be fine, and the clock tolerance doesn't really matter so it can be internal). It would also be cool if it had good sleep options and a variety of wakeup sources. So far I've been looking at the 10f series, but I'm not liking the fact that they can only run off their 4MHz internal RC oscillator, meaning it's going to waste a low of power on speed I don't need. Also, the only wake-up sources are pin changes and comparator output changes; I'd really like something that can wake up on a counter overflow or something so I can have it wake up at fixed intervals.
|
|
#
?
Jun 12, 2009 17:11
|
|
|
The PIC10 series as well as the baseline PIC12 and PIC16 devices would be pretty worthless for you since they don't have hardware interrupts or timers that can run while sleeping. The PIC12F635 and PIC12F683 have internal 32KHz oscillators plus an 8MHz one that can be scaled from 8MHz to 125KHz. They have 3 timers that can trigger interrupts on overflow which make them great for precise timing. One of the timers (TIMER1) can wake it from sleep, but it requires an external 32KHz crystal because sleep mode shuts down the internal oscillators. Hooking up a crystal consumes 2 of the 6 I/O pins, but if you were looking at PIC10s with only 4 I/O pins anyways then it should be okay. BattleMaster fucked around with this message at 20:00 on Jun 12, 2009 |
|
#
?
Jun 12, 2009 17:36
|
|
|
Aluminum Record posted:http://youdoitelectronics.com/index.htm I want to check that out, but holy crap I'm going to have to take my bike on the commuter rail, then bike from the train stop to the store. Wow. That's going to be an all-day adventure.
|
|
#
?
Jun 15, 2009 15:39
|
|
|
I'm doing some prototyping right now and making all my connections with female headers and no connectors (wire in hole). This is the most painful thing I've ever done in my life. What kind of connectors can I get (need 2pin, 4pin, 15pin, 16pin) to mate with female headers or male headers (I'd rather use male headers on my boards because then I can just buy 100 pin breakaways). I've looked all over digikey/mouser but their connector sections are full of confusing products that don't look like what I need. I guess what I'd like to buy is something like this, but I don't know what parts to get (are the assembly/pins separate?) or what tool I need to crimp the connections.
|
|
#
?
Jun 15, 2009 19:07
|
|
|
I did some searching and found these sockets that would be perfect for me: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=A3024-ND http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=A3025-ND http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=A26965-ND http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=A26963-ND I found the female crimp connectors and they're ridiculously expensive (.30/ea) unless you buy like 10000 at once. I'd be spending nearly 20 bucks per board just on connectors. There is also the fact that digikey sells Tyco's crimping tool for 580 dollars! Is there any product line that does the same thing but is more affordable for someone not buying in bulk?
|
|
#
?
Jun 15, 2009 21:27
|
|
|
Scarboy posted:crimping tool for 580 dollars! Are crimp tools patented or something? Why doesn't someone start a company mass-producing these things and selling them for cheap?
|
|
#
?
Jun 16, 2009 04:36
|
|
|
BattleMaster posted:The PIC10 series as well as the baseline PIC12 and PIC16 devices would be pretty worthless for you since they don't have hardware interrupts or timers that can run while sleeping. Those look like pretty good compromises. The scalable clock is a feature I was looking for too. Thanks for the recommendations. It's kind of frustrating that I'm working for a company that deals with cutting edge liquid crystal technology, but when it comes to electronics they want the absolute dirt cheapest solutions at a cost of functionality.
|
|
#
?
Jun 16, 2009 13:00
|
|
|
Scarboy posted:I'm doing some prototyping right now and making all my connections with female headers and no connectors (wire in hole). This is the most painful thing I've ever done in my life. What kind of connectors can I get (need 2pin, 4pin, 15pin, 16pin) to mate with female headers or male headers (I'd rather use male headers on my boards because then I can just buy 100 pin breakaways). I've looked all over digikey/mouser but their connector sections are full of confusing products that don't look like what I need. I haven't found a good solution to that yet either. What I usually end up doing is getting a roll of flat ribbon cable and buying those rectangular press together ends for them. They come in different sizes and for the board side (usually the male end) you can get solder or wire wrap termination.
|
|
#
?
Jun 16, 2009 14:01
|
|
|
ANIME AKBAR posted:Those look like pretty good compromises. The scalable clock is a feature I was looking for too. Thanks for the recommendations. Oh, one thing I forgot to add is that all PICs have a Watchdog timer that can be used to wake it from sleep, but I never use it so it slipped my mind. The one on the baseline devices can only be set to overflow in a maximum of 2 seconds, but the one on midrange devices like the two I listed can be set to overflow in several minutes. TIMER1 with a 32KHz crystal uses less power and is more precise, though.
|
|
#
?
Jun 16, 2009 17:36
|
|
|
|
| # ? May 21, 2024 14:27 |
|
|
Cyril Sneer posted:Are crimp tools patented or something? Why doesn't someone start a company mass-producing these things and selling them for cheap? Because a cheap crimping tool is called a pair of needle-nose pliers. It is kind of ridiculous but if you ever have to do more than a hundred crimp connections in a day it makes sense. You have to be able to rely on the connections you make.
|
|
#
?
Jun 17, 2009 05:59
|
|
