Spacedoggys WB Build Diary

[spacedoggy]

Hi Guys

I too am making a WB Build Diary also after reading cornz' thread
I think it's good for all of us to share our successes and failures so we can all learn...

here are my battery boards. I thought I'd try both incase I messed 1 up...


http://i.imgur.com/Mtel2.jpg

I have a question about the polarity of the LED (see red arrow)
the green line indicates the negative terminal on the led
but on the board which is the negative pad?

also if you can see any blatant errors, please let me know.
I am awaiting the IC for charging the battery, I ordered a few
of them because there was a minimum order of 100 dollars ;(

VCOs in the post also.

One final question, I intend on using the nokia battery in the pic,
seems right voltage. is protection circuity a must here? is this battery ok?
I see the WB 2010 had overheat detection, it this for a reason, has anyones
WB exploded or anything while charging?

please give comments and feedback and I'll post updates of my progress.

SD

[TheFallen]

According to my schematics you want the negative side of the LED to be closest to the ~3mm plated through hole.

This means that in your pictures you need to have the green line on "top" as it were. Opposite to how you have it placed in the picture.

[spacedoggy]

ok, I'm still awaiting my final IC for the li on charger board, but can proceed using the AAA battery board.

so the power supply...
http://i.imgur.com/ITGnW.jpg

I made the mistake to put everything on before testing...

not good, I have a problem I heard about on here before. shortly after i connect power i hear a high pitch sound, from the board, and the voltages measures go nuts, varying from 0V to 2x the rails correct value. i measured 50V on the 28V rail. I cant guess where I have soldered incorrectly.

also for R1 and R2 I could not assertain their values and so have not added them yet. can you please explain the value of these resisters?
e.g. what does this mean? "Chosen value resistors for low battery detect: R1/(R1+R2)*Vlow = 1.1"

once again, please take a moment to look at the pics, and consider my battery questions above.

many thanks.

[TheFallen]

Looking at the picture, are you sure IC2 is on straight, it's awfully wonky.

R7 should be a 1M (105) resistor, it looks like a 100M resistor (107) however R6 looks okay. Check this out, it's a semi-completed picture of my WB.

Could you do us a favour, start it up, get the initial voltages of ALL the regulators, 3v3, 5/6V, 12V and 28V. Then, and this is the scary part, leave it for a little while watching for smoke then measure again. If it starts smoking TURN IT OFF!! The idea is that the initial turn on the regulators have to fill some large capacitors (C14 and C18) and because they're not full of charge yet the voltage across them is not what it should be (12V and 28V respectively). This will cause them to generate higher than expected voltages for short periods of time at start up.

R1 & R2 are for the low battery indicator (thus lbi), i.e. the divide the input voltage down by R2/R1+R2, in this case 220k/(510k+220k) ~ 0.3. So if Voltage in is 3V3 (a lithium battery) then the output is about 1.1Volt which then gets fed into the ATMega168/88's ADC. It would not cause the problems you're describing.

[spacedoggy]

thanks fallen for your feedback and your pic...

IC2 is indeed a bit off, I removed the inductor to fix it and it melted, lol, gonna have to order another one grrr
I'll order it with my WB2010 parts

R7 is indeed a 105

so R1 & R2 are 510k 220k

if I'm using the AAA battery board is this the same?

this is a great project for learning how to fix terrible mistakes. I'm learning loads by being a complete fuck up, LOL

[TheFallen]

Nope. This is a TERRIBLE project. It's not only SMD, but tightly packed fine-pitch SMD. In a lot of cases not only are mistakes expensive to fix, but impossible, if you screw up the LMX2433, your screwed. Oh and the VCOs are not easy to repair if you mix the high and low bands up.

The AAA board takes 3 1V2-1V5 batteries. That should give you between 3V to 5V which then, through R1 and R2 gets reduced to 1.1V and 1.5V. The voltages are still okay for the ATMega***'s ADCs but as far as I know there's no code in the firmware that takes advantage of the LBI. You could easily ignore them if you wanted.

[spacedoggy]

thanks fallen, you're a machine! will let you know how i get on when replacement bits get here.

[spacedoggy]

hi all,

i got my replacement inductor at last (melted it). and resumed testing...

I noticed that C10 was extermely hot. and it appeared to be shorted, so i removed it and measured the 3.3V as holding at 3.6V (!)
then probed the pad of the removed capacitor for 5 V and found it at 18V

have you guys any thoughts on how I can further test this?

can the C10 cap be temporarily replaces with one of the many normal 100uf caps i have lying about for breadboard projects, or are these yellow ones special?

I gonna start on the WB2010 as soon as the trace issues are sorted, I'll be a little more careful taking measurements this time... might make a seporate thread for that though as I'd like to see both working.

[spacedoggy]

also I have 1 question about the lion battery board. when you are charging it does the green light turn off to indicate charging is complete?
I had mine plugged in for a day and I would like to check that it is not being overcharged etc.

as always comments and knowledge scraps appreciated

[TheFallen]

+3V6 out of an unloaded +3V3 regulator is perfect. Even loaded, that's well within most tolerable values. Less than +10% out.

The LED, well *shrug* I don't know, check the datasheet

# clicked submit instead of preview...

Yes, that LED should only light up when the battery is being charged. D1, 1N5818 stops current from flowing back from a fully charged battery. The LM3622 also stops charging when it senses the battery is at maximum capacity. So you don't have to worry about over charging.

[spacedoggy]

OK...

replaced cap.

applied power, no noise...(yet) seems to be holding...

taking voltage measurements...

rail measurement
3.3V 3.7V
5V 12V
12V 12V
28V 28V

High Pitch Tone and room filling FLASH!

I recheck all rails after allowing time to discharge...

I measure the 3.3V rail...
-it 3.6V consistently for 20 Seconds
-I hear tone and remove power, reading remains at 3.6V during this time

I measure the 5V rail...
-it reads 12V coming down to 5V over 10-15 seconds...
-on 5V I hear tone again and remove power.

I measure 12V rail...
-it reads 12V consistently for 60 Seconds
-I hear tone and remove power, reading remains at 12V during this time

I measure 28V rail...
-it reads 26V climbing to 28 over 40 seconds...
-on 28V I hear tone again, voltage jumps to 51V

[TheFallen]

May have a wrong resistor divider (R3 = 39k/R4 = 12k) on the 5V/6V supply, I hate that chip. Only problem I've had on my WB that wasn't my fault (put the wrong VCO in the wrong spot otherwise).

I'm getting:
28V -> +26.4V (taken across C18)
VCC -> +3V3 (taken across C5)
Vin -> +3.96V (taken across C1)
12V -> +11.99V (taken across C14)
6V - > +5.23V (taken across C10)

Don't worry about tones, mine buzzes too.

[spacedoggy]

ok, put in R1 and R2 after looking at the diagram and comments, applied power, IC1 up in smoke.

banned hell.

WB on hold. I'll try WB2010 in another thread

*sigh*

[TheFallen]

Hang on a sec, R1 & R2 are the LBI (Low Battery Indicator) something which is pointedly ignored in the code IIRC. It should have no affect on anything by soldering them in.

IC1 is the TPS79333, the Vin -> +3V3 regulator, a linear drop out regulator, which does no switching. I'd say you have a short on your +3v3 rail somewhere. In fact I'd start by checking the top connection of R2, pretty much where the silkscreen label is. Also check between IC1 and C2, the ground/ +3V3 is pretty close there too.

If that doesn't solve it, send me a private message, I may be able to help.

[spacedoggy]

thanks fallen, I'll check this.