Adafruit - THE MUSICAL - is here! Get a musical tour of Becky's favorite tools around the Adafruit factory (video) and subscribe!

Sow how are they making one go 300ft plus?
Moderators: adafruit_support_bill, adafruit

by shadow on Fri Aug 22, 2008 3:45 pm

OMG, my ehp boards get here today, and god help me if UPS is late.

Probledude- srry about the LED shortage

if your in Phoenix, Arizona, i could spare some extra
shadow
 
Posts: 22
Joined: Sun Mar 02, 2008 5:59 pm

by Probedude on Fri Aug 22, 2008 4:54 pm

shadow wrote:Probledude- srry about the LED shortage

if your in Phoenix, Arizona, i could spare some extra


Thanks for the offer! I'm sure I'll find them in the next day or so. I did place an order for some more yesterday too so in the worst case I'll have them mid next week.

Attached are pictures of the board I did - used the photopositive method. The silver rectangle next to the board is the LiPo battery that I am going to use. I've only been testing with a single cell so far (3.6 - 4.2V) but will try doubling it this weekend once I find the TSAL6100's. (I only installed 4 LED's of the Everlight variety so far)

As you can see the battery is small and thin. Even 2 stacked on top each other will be thinner than the PCB + components.
Image
Image
Probedude
 
Posts: 87
Joined: Sun Dec 23, 2007 3:16 am

by Probedude on Fri Aug 22, 2008 11:53 pm

Tonight I installed some more Everlight IR LEDs, put in the 5V regulator, and went and got one of my RC helicopter lipo packs (2s pack, so 7.2 - 8.4V). Lipos have very low internal resistance, hence what happened next.

Ended up frying 6 of my LED's - the symptom was no IR output. They were working for a bit, then not.

The official SHP TV-B-Gone is using 10 ohm base resistors - so I did too, but I was using PN2222A vs 2N4401 transistors. I changed the 10 ohm resistors out for 47 ohm, replaced all the IR LED's (again with Everlight - my TSAL order hasn't arrived yet). Activated it 3 times and it's still working.

Transistors still get pretty hot though! Would not want to use sot-23 sized smt parts with this much power.
Probedude
 
Posts: 87
Joined: Sun Dec 23, 2007 3:16 am

by shadow on Sat Aug 23, 2008 12:50 am

ok, so im starting to make my Holy sh*t thats more power than need tv b gone and am wondering- should i make all internal of the project box (6 batterys and ten ir leds, or have two 3X AA battery packs outside of the box and have 22 LEDS?
shadow
 
Posts: 22
Joined: Sun Mar 02, 2008 5:59 pm

by oPossum on Sat Aug 23, 2008 5:55 am

woody1189 wrote:just experimenting with the chip design some more...i noticed some problems with 4 leds off of 3 fresh alkalines in that the codes wouldn't cycle. i dont think the alkalines can supply enough current for 4 emitters but when you use just 2 leds it blasts em. lost 3 tsal6100s in the process :(


Change the 47 ohm resistors to 220 ohm or higher. That will reduce the LED current and should stop the AVR brownout/reset problem.
User avatar
oPossum
 
Posts: 635
Joined: Fri Oct 26, 2007 12:42 am
Location: Michigan, USA

by oPossum on Sat Aug 23, 2008 6:01 am

woody1189 wrote:i was using a multimeter but then i realized that those readings were most likely wrong. i was hoping to at least see some spikes above 100ma but all i got were low readings. i'm gonna test it on a scope once i go back to school next week.


If you have a dual trace scope, measure both current and voltage.

The current should be measured between the battery and the whole circuit. Measuring only the current of one LED with a shunt will starve that branch and give inaccurate readings.

The easy way to do this is to connect a 0.05 ohm (two 0.1 ohm in parallel) between the battery negative and the circuit. Connect the ground clip of both scope probes to the battery side of the resistor. Connect one scope probe to the other side of the resistor - this will measure current. Set the scale to 20 or 50 mV/div - multiply by 20 for and actual reading. Connect the other scope probe to the battery positive. Set the scale to 1 or 2 V/div.

The battery voltage should not drop below 2.5 Volts to ensure reliable operation. If it does, increase the value of the base resistors.
User avatar
oPossum
 
Posts: 635
Joined: Fri Oct 26, 2007 12:42 am
Location: Michigan, USA

by oPossum on Sat Aug 23, 2008 6:16 am

Probedude wrote:Ended up frying 6 of my LED's - the symptom was no IR output. They were working for a bit, then not.

The official SHP TV-B-Gone is using 10 ohm base resistors - so I did too, but I was using PN2222A vs 2N4401 transistors. I changed the 10 ohm resistors out for 47 ohm, replaced all the IR LED's (again with Everlight - my TSAL order hasn't arrived yet). Activated it 3 times and it's still working.

Transistors still get pretty hot though! Would not want to use sot-23 sized smt parts with this much power.


Go higher on the base resistors, not lower. Mitch's SHP has a 9V battery with a very high internal resistance, so there is no need to limit LED current by any other means. The EHP design assumed AA alkaline batteryies, so peak current was somewhat of a concern, but not a big one.

The battery you are using can easily fry the LEDs if the AVR voltage drops too low. The base resistors should be increased to limit the LED current and reduce the battery voltage drop.

I think you are very optimistic about the internal resistance of Li- batteries. They certainly can deliver high current under steady load, but 20 uS pulses make very different demands of the cell. I suspect the effective internal resistance will be much higher with the TV-B-Gone than with a steady load like a motor.

I don't see a diode on your PCB. It would be a good idea the have a Schottky diode between the battery and the AVR, and a 100 uF cap in parallel with the AVR. That will provide some protection against voltage drop during the high current pulses.

You can easily test the general health of circuit by shorting the base and emitter of the emitter folower transistor. That will greatly reduce the LED current. If that works, then start tweaking the value of the base resistors for a safe LED current and tolerable battery voltage. A 100 ohm pot between the emitter folower and the junction of the base resistors would provide easy adjustment. When a suitable resistance is found, multiply by 8 and add to the value of each base resistor.


I agree that sot-23 would not tolerate the current, but sot-223 would probably be OK.
Last edited by oPossum on Sat Aug 23, 2008 6:19 am, edited 1 time in total.
User avatar
oPossum
 
Posts: 635
Joined: Fri Oct 26, 2007 12:42 am
Location: Michigan, USA

by woody1189 on Sat Aug 23, 2008 9:55 am

sounds good opossum ill try that out sometime next week.

i hooked it up to both 2 packs of 3 aa's in parallel and a cell phone battery and it worked for both so i'm assuming the 3 aa's just cant provide enough current? i took it outside using the cell phone battery and couldn't hit my tv from over 150 feet away so i was kinda disappointed.

opossum, in your chip schematic where it says cut for high power, does that mean cut the line out if we are using 3 batteries or cut the line in if we want high power? i've been using it without that line.
woody1189
 
Posts: 85
Joined: Sat Jun 21, 2008 11:19 pm

by Probedude on Sat Aug 23, 2008 4:16 pm

oPossum wrote:Go higher on the base resistors, not lower. Mitch's SHP has a 9V battery with a very high internal resistance, so there is no need to limit LED current by any other means. The EHP design assumed AA alkaline batteryies, so peak current was somewhat of a concern, but not a big one. The battery you are using can easily fry the LEDs if the AVR voltage drops too low. The base resistors should be increased to limit the LED current and reduce the battery voltage drop.


Which is why I went from 10 ohms from Mitch's design (which burned out the LEDs) to 47 ohms, not the other way around. I fried the LEDs using the 2 lipo cell setup with a LDO regulator feeding the MCU and a cap. No worries about dipping voltage and it didn't fry the LEDs because the MCU didn't have enough voltage.

I think you are very optimistic about the internal resistance of Li- batteries.

Where's the optimism? All the other TV-B-Gone designs have been powered by alkaline AA, AAA or 9V batts. This particular lipo pack is rated for 20C continuous discharge - it's not a cell phone lipo and the IR is much better than a series string of the above batts.

I don't see a diode on your PCB. It would be a good idea the have a Schottky diode between the battery and the AVR, and a 100 uF cap in parallel with the AVR. That will provide some protection against voltage drop during the high current pulses.


Agreed for other setups - but not an issue with my 2 lipo setup (see above). My 1 lipo cell setup doesn't have an issue either. The bigger concern is to make sure I don't overdischarge the lipos.

I agree that sot-23 would not tolerate the current, but sot-223 would probably be OK.
Likely better - power dissipation values for the sot-223 package on a properly designed PCB if I remember correctly is better than the to-92 case. Downside is the pcb footprint is larger on the sot-223 vs the to-92. Going back to Mitch's SHP schematic he shows sot-23 transistors being used. It works because of the 9V battery. It won't work with a full charged 2 cell lipo (8.4V) though if anyone else follows my footsteps and uses the SHP schematic verbatim. That's what I get for not taking measurements :)

Testing
Last night I did some range testing. Very good distance inside the house, ~ 70' - down the hall and around a corner I can nail my TV no problem. Could only do half the hall distance before.

From outside my house standing on the porch, pointed through the closed front door security window, around a corner to the TV in another room - success!

Today's adventures.
Went into Best Buy and Sam's club - turned off a few TV's for as little time as I wanted to spend standing in one spot facing that direction. For some reason my success was zero at Walmart. Had this problem before too. Don't know why - wonder if they tape over the IR receivers on their wall of TV's.

Tonight I think I'll take the dogs for a walk and see if I can turn off neighbor TV's from the sidewalk pointed toward their living room/ bedroom windows.

edit: Just bought a dummy Blackberry phone on Ebay. TVBG EHP with 2 cell lipo in a Blackberry Pearl! Woohoo!
Probedude
 
Posts: 87
Joined: Sun Dec 23, 2007 3:16 am

by gotwake424 on Sun Aug 24, 2008 9:34 am

wow, around corners, NIce! I think the problem your haveing at walmart, im having the same problem. if your trying to get the ones hanging from the celiing they have cardboard over the whole front panel, so i dont think the IR signal can reach them. Thats just my theory. btw i have V1.0 it worth it to get v2.0 or is it v1.5 anyway.

Gotwake424
Soldering irons get hot?
gotwake424
 
Posts: 13
Joined: Sat Aug 23, 2008 8:40 pm

by woody1189 on Sun Aug 24, 2008 5:43 pm

probedude, wire one of these up to your lipo pack and you wont have discharge or overcharge problems.

http://www.batteryspace.com/index.asp?P ... rodID=2774
woody1189
 
Posts: 85
Joined: Sat Jun 21, 2008 11:19 pm

by Probedude on Sun Aug 24, 2008 7:03 pm

woody1189 wrote:probedude, wire one of these up to your lipo pack and you wont have discharge or overcharge problems.

http://www.batteryspace.com/index.asp?P ... rodID=2774


Thanks Woody! Nice looking protection board.
I shouldn't have problems with overdischarge as long as I remember to unplug the battery when I'm not out and about with my stealth Blackberry :) Will look into those IC's and may use them if their spec sheet is ok.

In regards to charging it, I've got a plethora of chargers for all my RC planes and helicopters. I've been using mostly this for my lipos
http://www.fmadirect.com/detail.htm?ite ... section=45
It's a balance charger so each cell is brought up to 4.2V.
Probedude
 
Posts: 87
Joined: Sun Dec 23, 2007 3:16 am

by magician13134 on Mon Aug 25, 2008 6:21 pm

Yay, I just got the email that my boards shipped today from China. It should only be three days tops unless there are unforeseen customs problems!! 100 EHP and 150 CHiP full spec boards! I'll post pictures/try to get everything in stock in the MSI store as soon as they come in
User avatar
magician13134
 
Posts: 1119
Joined: Wed Jun 13, 2007 9:17 am
Location: Ann Arbor

by woody1189 on Tue Aug 26, 2008 11:25 am

will the avr output the same current from a 3.3v source as it would from ~4.5v?
woody1189
 
Posts: 85
Joined: Sat Jun 21, 2008 11:19 pm

by woody1189 on Wed Aug 27, 2008 12:15 pm

just making sure my numbers are right...

tsal6100s can dissipate 210mW so:

at 2.1V*.5A the power is 1.05W, assuming 25% duty cycle, 1.05W/4=262.5mW.

I'm assuming this is why the chip design has been killing my tsal6100s by putting 510ma into them with 47 ohm resistors. I figure I can get by with 210mW and have them not die on me since those ratings are often lower than in reality. So:

210mW*4=.84W
.84W/2.1V=400mA

Beta=40
.400A/40=10mA input

from 4.6V given by three AA's, 2V lost be each transistor;

.6V/.01A=60 Ohms

So for maximum(and Safe!) power off of three AA I'll need to use 60ohm +/- a few.

Does that make sense?
If so, i was hitting my tv from 140 feet with ~150ma goin through each led. I'm gonna try 400ma today.
woody1189
 
Posts: 85
Joined: Sat Jun 21, 2008 11:19 pm



cron