Need help with electronics please
#1
Posted 01 November 2011 - 12:00 AM
Hi,
I am in search of help with the Tron Disc Project. First of all, I don't know jack about programing or electronics, but I'm a quick learner.
I am currently following the two guides here:
Original Guide Hackerspace/Coding4fun
Thundergod Guide using netduino mini
This is the basic circuit. But I would like to alter the following:
- use a premade USB charger unit like this: sparkfun micro usb charger - Data sheed with schematic here
- add more LEDs: 30 outer ring, 16 inner ring (8 on each side) and 8 for the solo lights with bezel mounts on the inner surface of the disc (4 on each side) = total of 54
The basic idea is to create the same animation as in the movie (youtube link @ scene 34 sec).
Now the questions are:
- Can I simply smolder the usb charger directly to the netduino and and attach >>both<<??? battaries onto that one charger? If yes can you please clarify how the batteries should be wired/smoldered to the usb charger?
- If I go with 54 LEDs:
- Do I need other batteries? I bought 54 x nichia 6500mcd 45° - data sheet here
- If the batteries stay the same, will my LEDs simply run out of power faster or are they going to shine less bright?
- If they shine less bright, what battery do i need?
Thanks in advance
#2
Posted 01 November 2011 - 07:04 AM
#3
Posted 01 November 2011 - 05:16 PM
I don't have much experience about LiPo-charging, but to be honest, I don't like the charging/detecting circuit above.
I am not sure, but it seems that you have to balance the charge when two or more LiPO batteries are connected together. Another clue is that there's no LiPo batteries chargers for series pattern. You find balanced chargers, single-cell chargers, but no double-cell chargers...at least I wasn't able to find them.
Mario, you are very much correct. LiPo batteries *must* be balance charged.
The other thing I would say is if you are charging LiPo batteries, you really should get a "LIPOSACK." Should something go wrong with the batteries, it can mean the difference between some smoke damage in your house and a full on fire.
-dan
#4
Posted 02 November 2011 - 05:03 AM
I didn't want to talk about the danger about the Lithium batteries, because -as stated- I have no experience enough. However, the Polymer-Lithium (LiPo) should be aware from being explosive, I guess.The other thing I would say is if you are charging LiPo batteries, you really should get a "LIPOSACK." Should something go wrong with the batteries, it can mean the difference between some smoke damage in your house and a full on fire.
Anyway, you are confirming me that it's better to charge them separately.
Cheers
#5
Posted 02 November 2011 - 11:35 PM
#6
Posted 03 November 2011 - 05:13 AM
A single cell gives you 3.7V, thus is not enough to power the Netduino. You must use at least two cells, for a total of about 7.4V. To tell the truth, even 7.5V is just the minimum threshold of usage for the Netduino regulator, thus that's not the best solution.
Well, the most decent solution I may suggest is the following:
- use just a LiPo cell (3.7V), with the related chanrger;
- add a step-up module (input 3..5V, output 9..12V) to power the Netduino *ONLY*;
- power the led set (MAX included) directly using the battery power (no conversion);
Hope it helps.
Cheers
#7
Posted 04 November 2011 - 02:16 AM
Update:
I like the idea of only 1 lipo battery.
Lipo Battery I wanted to use: 850mah 2c 3.7V Will this be enough for the 50 LEDs?
I prepared and new Schematic ... was a pain in the ass to learn how to use a circuit diagram program :E but it was worth it. The Schematic is far from being completed though.
- Don't know how to connect the battery directly to the max and the LEDs O_o? Do I need another Step Up Module for that too? At the moment it looks like it is powered over the netduino with 5V ... or do you mean by "no conversion" that it will run just fine with 3,7 too directly from the battery too?
- Don't know how the step up module will look like. I found a premade version with a build in potetiometer, supposed to be regulating from output 2V-35V and input ... this will take some time to research for me, already had a quick peek at it though. If i understood correctly i can switch the potention meter from 10k ohm to 1k ohm, since i only need to set it up from 3,7 to around 9 volt for easier adjusting
- GND is Ground ... where does this go in the actual setup, do i have to wire it somewhere, if yes where to? Guess this one is total noob question
- Don't know if i copied the wires correnctly going from arduino to the mx7219
. Don't know if I set up the LEDs correntcly, since i added more to the original design, i simply expanded it though, so it should be fine
And in case anyone wants to adjust the schematic and contribute, here is the .sch for Eagle:
Schematic Download on Megaupload for Eagle
#8
Posted 04 November 2011 - 06:34 AM
== Battery ==
Is it enough a 3.7V/850mAh?...Well, it depends on the usage.
Assuming that:
- most of the energy need for the leds: Netduino and MAX(*) itself, suck very little;
- *ALL* the battery energy is available for the circuit (high efficiency, no heat waste);
Now, ask yourself: how is the current used for one led? how is the average number of leds enlightened?
It follows the simple calculation...but the experience tells even more.
(*)excluding the led
== Step-up ==
The circuit in the lower-right corner is *NOT* a step-up, but a normal regulator having low-efficiency.
A possible step-up module could be this one.
However, the current available is pretty low (about 100mA), thus *of sure* it cannot used to power the leds.
The actual problem is that the MAX cannot be powered at 3.7V, thus you must power it at 5V along with the Netduino. This would not be a problem, unless the MAX has to power the led set.
== Led power ==
The simplest way to solve the problem above, is to insert some kind of buffer/driver on the MAX's SEGx lines. It has to act as voltage level shifter, from the 5V MAX level to the 3.7V needed for the led.
Let me some time to find out a decent solution, but I think there are no other (simple) way to solve it out...
== GND ==
What's problem about GND?
It's the "ground" wire/plate: the electrical point common for any circuit having a conventional potential of zero Volt.
There are some minor optimization can be made on the whole circuit, but we'll talk about them later.
Cheers
#9
Posted 04 November 2011 - 04:20 PM
Nichia 5mm LED white 6150mcd 45° NSPW510DS
Current 10 mA 20 mA 30 mA
Voltage 3,00 V 3,20 V 3,30 V
light power 3075 mcd 6150 mcd 9225 mcd
So 50x20mA = 1000mA ... meaning i could power the whole disk nonstop for about 45min before recharging. Thats plenty of time. And at home it will be connected to usb anyway the whole time as a lamp near my pc.
== Max7219 ==
I though the whole purpose of the chip was to give the right voltage to the targeted bits/leds? It says on Maxims homepage Only one external resistor is required to set the segment current for all LEDs. So that shouldnt be problem right, since the resistor is alrady set (20k ohm in the original schematic).
But then again, this is only my assumption, with my limited electronics knowledge >_>
Within the original design it looks to me like it is going from two battaries 7.4V to the netduino, then the netduino has its own 5v out going to the max, and the max is set to 3,5V output (thats the volatage of the LED used in the original design) through the resistor at the build in converter or something. My LEDs need only 3,2V, guess I'll need another resistor then?
#10
Posted 04 November 2011 - 07:01 PM
I'd get a step back to the pair of batteries, thus the overall voltage will be 7.4V.
This seems the most convenient option, because the led need a voltage that is too close to the single cell battery. Moreover, with a single cell battery (i.e. 3.7V) there are the problems mentioned in my previous post.
Let's forget the battery charger for a while.
You should use two batteries in series, and a DC-DC converter like this one. It allows as input a wide range of voltages, and is capable of an output of 1A at +5V.
At this point, you'll use the DC-DC regulated supply for either the Netduino (directly on the +5V pin), and for the MAX, leds included.
With this trick you'll double the total capacity, thus the duration becomes almost twice.
I'll send you a solution for the charger: let me some time to draw it.
Cheers
#11
Posted 04 November 2011 - 09:26 PM
From what i learned today about charging multi cell lipo batteries, it is possible but dangerous and both batteries need to be checked by the factory to be usable together ... and even then they require a rather complicated charger. Found it here. I have also found some multicell chargers but they are all huge.
Frankly i don't even understand how they made it work in the original design O_o?
== LEDs Voltage ==
I found out how to set the output voltage for the max7219 from the data sheet (table 11). The Resistor (R1 in the schematics) needs to be ~24kohm for 3.2V. Yay \o/ ... So the max takes in 5v from the netduino and passes on 3.2v to the Leds.
I'd get a step back to the pair of batteries, thus the overall voltage will be 7.4V.
This seems the most convenient option, because the led need a voltage that is too close to the single cell battery. Moreover, with a single cell battery (i.e. 3.7V) there are the problems mentioned in my previous post.
Another alternative would be to find a rechargeable battery that has an output of 5V or 9-12V right? Since the netduino can be powered by either. I'll check if I can find something like that.
EDIT: I am wondering how other ppl use their netduino??? can't find anything on power source related problems ... am I the only one who wants a circuit to run through usb and a rechargeable battery like a cellphone ^^??? I thought this technology is quite common now days? Even my PC Mouse is working that way, has a 1.2V rechargable battery, but is chargable via usb which is 5V. Or am i missing something here?
Edit 2: Uh actually found someone: netduino forums link
#12
Posted 05 November 2011 - 06:15 AM
A knife would kill a people, but you can't avoid to eat or shaving, because has a blade...
As LadyAda pointed out, the only allowed pattern is to place in series two batteries. Note that this is *NOT* a totally safe pattern, but it would be not even with any other kind of battery.
That's because the two (or more) batteries involved may not be perfectly equal, and their discharge may be different. Even the current flowing thru is exactly the same.
For instance, one of them discharges faster, and the voltage drops are different.
BTW, your car battery (12V) has 6 2V-cells in series, and it's safe...how is it? That's because the whole battery has been made with the same material, same manufacturing, same operator, etc. What would be the difference?
It's much when you take 2 children, and two twins.
Instead, when you take any battery (typically purchased), you have any guarantee that they come from the same lot, same operator, same material, etc.
Under these circumstances, a *VERY GOOD* pattern for a series batteries, is to monitor the total voltage, as long the intermediate voltage. Thus, you may see when a battery is below a certain limit.
...ANYWAY...
I wouldn't be scared...at least, I'd try some experience...
Below is depicted my suggestion.
Consider the bottom section of the scheme. You have two batteries (don't care about the alkaline pair: keep it as a single LiPo-cell), which have:
- to be placed in series, when operating;
- to be charged separately (independently), when not operating.
- to be excluded when you want your circuit switched off;
The breadboard would be the disc circuit.
Your disc will have a 5-holes *female* connector, connected to the batteries and to the circuit (breadboard) as depicted.
Afterward, you will have another TWO *male* 5-pins connectors: one for operating, the second for charging.
When no connector is plugged in, the disc is switched off.
The operating connector has just a couple of wires, as depicted in the upper section of the picture.
If you try to follow the current path, you'll see that the batteries are actually placed in series, thus your circuit is 7.4V powered.
The second "charging" connector is trivial, and must be wired to a couple of simple LiPo chargers. Any kind of charger would fit (accordingly to the battery type).
All the remaining considerations, about the DC-DC to efficiently step down the voltage from 7.4 to 5V, are still valid.
Hope it helps.
Cheers
Attached Files
#13
Posted 06 November 2011 - 07:42 AM
I was following up on your first post Mario, when you said:
The circuit above is powering the MAX+leds with +5V. This supply is given by the small regulator embedded in the Netduino Mini, and it's *NOT* designed for large powers. It would result in a overheating.
Amd I just found out 2 things:
1. netduino microcontroller max current: 200 mA total
2. USB max current: 500 mA
Since I need 1000 for the LEDs alone 80 something for the netduino and probably another 80 for the max7219, I won't be able to run this thing via USB. Guess my dreams just got shattered :L
So back to square one. No USB charger. Back to two batteries. Will try using the charging and charge detechtion circuit from the original design >_> but add a step down from batteries -> max7219 instead of batteries -> netduino -> max7219 as you already suggested. I will need a step down from 7.4V to 5V then.
I also checked out the charging circuit and it was designed by someone who knows what he is doing. It is indeed a way to charge 2 cells in series. I've been checking his circuit against mine and found out what i need to alter to use it for 7.5 batteries. Well basically it was already set correct in the original deisgn fron Harford Hackerspace guys. But I was actually able to understand it and recalculate the resistors and voltage, since it is explained in detail on shdesign.org ... Even the 7.5V enrty makes sense now to me \o/ yay.
In case you're interested:
Main article on the charging circuit
In depth explanation
Original schematic
Improved Schematic
Things to do now:
- find matching stepdown module or parts for atleast 1A (50x20mA)
- order battery
- order female dc connector
- order stepdown module / parts
- prototype and test that it actually works (also check voltage, set pot, etc.)
- then finally build it >_>
New Schematic
#14
Posted 06 November 2011 - 08:18 AM
#15
Posted 06 November 2011 - 08:40 AM
#16
Posted 10 November 2011 - 04:22 PM
I found a balance charger circuit in an rc forum!
Can anyone tell me if it will work like this? I was already told i need to "adjust/change the resistors divider of LM2577" in the charging circuit , no idea how to do that yet
eagle schematic here
#17
Posted 10 November 2011 - 05:47 PM
I have not tried to understand the circuit, but personally, I would use a specialized dual cell charger IC (from TI, Maxim, Microchip etc.) - it will be probably a little bit more expensive than discrete components above, but it will have correct charging management (switching among various charging modes/algorithms), over/under voltage and thermal protection, adjustable cut-off voltage etc. Those ICs often require minimum external components, can have LEDs connected directly to signal the charging status and there is usually typical application schematic included in the datasheet.Can anyone tell me if it will work like this? I was already told i need to "adjust/change the resistors divider of LM2577" in the charging circuit , no idea how to do that yet
Alternatively, I would consider using batteries with connectors and charge them externally
Edit: With the power requirements (50 LEDs) the cell phone battery is probably not a good idea, they have typical capacity about 1000 mAh.
#18
Posted 10 November 2011 - 06:51 PM
#19
Posted 10 November 2011 - 07:49 PM
#20
Posted 10 November 2011 - 09:02 PM
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users