Icetube clock instruction manual

For RTC breakouts, etc., use the Other Products from Adafruit forum

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
User avatar
adafruit_support_bill
 
Posts: 88092
Joined: Sat Feb 07, 2009 10:11 am

Re: Icetube clock instruction manual

Post by adafruit_support_bill »

I now tried another brand new battery and it goes down to the same value under load. Under no load is about 3.3V.

I assume this shows a higher than expected drain?
Yes. With a fresh battery you should get 3v or more in-circuit.

User avatar
phild13
 
Posts: 247
Joined: Mon Sep 10, 2012 1:05 pm

Re: Icetube clock instruction manual

Post by phild13 »

Looking at the schematic, I believe that if the Q3/Q4 (the ZVP3306) FET is installed incorrectly, then it will basically be "on" all the time because of the internal diode, draining the battery as the battery will try to supply the MAX6921 (square) chip and bias the tube.

I would verify that the FET is installed correctly according to the written instructions http://learn.adafruit.com/ice-tube-cloc ... d-assembly. If the writing is on the rounded side of the FET then don't go by the silkscreen outline on the board as the FET will be installed incorrectly. Install it so the rounded side is facing the large capacitor and the 22 ohm resistor. If the writing on the FET is on the flat side then install it like the silkscreen outline shows with the flat side facing the capacitor and the 22 ohm resistor.

niznai
 
Posts: 28
Joined: Fri Aug 09, 2013 10:51 pm

Re: Icetube clock instruction manual

Post by niznai »

Yep, the FET is installed correctly. If it wasn't, there would be other problems as well. BUt I can assure you mine is installed correctly because I have read the part in the instructions. Mine is with the writing on the round side, so it's facing the other way around to the outline on the PCB.

Any suggestions with the voltage divider anyone?

I haven't found a battery that will keep above 3V in the clock, power disconnected, so I think there's some problem with the componentry.

User avatar
adafruit_support_bill
 
Posts: 88092
Joined: Sat Feb 07, 2009 10:11 am

Re: Icetube clock instruction manual

Post by adafruit_support_bill »

niznai - can you post a photo of the component side of the main board?

Also, with the 3v cell installed, what voltage do you measure on pin 13 of the Atmega processor?

User avatar
adafruit_support_rick
 
Posts: 35092
Joined: Tue Mar 15, 2011 11:42 am

Re: Icetube clock instruction manual

Post by adafruit_support_rick »

Check diode D4. It's right next to the ATmega processor. If that is in backwards, then you won't get any current flow from the battery.
The white stripe should be towards the battery end of the diode.

niznai
 
Posts: 28
Joined: Fri Aug 09, 2013 10:51 pm

Re: Icetube clock instruction manual

Post by niznai »

Like I said before, I get about 0.25V on pin 13 on the battery.

The diode (D4) is installed as printed on the board (figure on this page http://learn.adafruit.com/ice-tube-cloc ... d-assembly ), which would suggest the non-stripe end towards the battery. That is consistent to the circuit diagram as well (see this page http://learn.adafruit.com/ice-tube-clock-kit/design ).
Or I don't understand what you mean when you say "towards the battery".

PS. I posted all the voltages in the battery circuitry all the way to pin 13 on the previous page. All values are read with the mains supply removed, of course.
Last edited by niznai on Fri Aug 16, 2013 2:23 pm, edited 1 time in total.

User avatar
adafruit_support_rick
 
Posts: 35092
Joined: Tue Mar 15, 2011 11:42 am

Re: Icetube clock instruction manual

Post by adafruit_support_rick »

niznai wrote:Or I don't understand what you mean when you say "towards the battery".
And it's right that you shouldn't understand. That was a brain-cramp on my part. I was thinking that the battery was someplace else. :oops: :oops:

OK, so the stripe is in the same place as the stripe on the silkscreen. What voltage are you seeing on either pin of that diode?

niznai
 
Posts: 28
Joined: Fri Aug 09, 2013 10:51 pm

Re: Icetube clock instruction manual

Post by niznai »

There's about .6 V drop across the diode, so about .7 on the cathode and battery voltage on the anode (1.2V). 0.6V drop on the diode looks fine to me.

User avatar
jarchie
 
Posts: 615
Joined: Sun Jun 24, 2012 2:16 pm

Re: Icetube clock instruction manual

Post by jarchie »

niznai wrote:There's about .6 V drop across the diode, so about .7 on the cathode and battery voltage on the anode (1.2V). 0.6V drop on the diode looks fine to me.
A 0.6v drop across the diode seems really high to me. Unfortunately, I don't have my clock in front of me, but I remember the voltage drop being much lower--less than 0.2v. Such a high voltage drop might indicate excessive current consumption during sleep.

Would you you check to see if the VFD heater is being turned off during sleep? To do that, measure the voltage on R3 (either side) without external power. Voltage should be essentially zero.

User avatar
phild13
 
Posts: 247
Joined: Mon Sep 10, 2012 1:05 pm

Re: Icetube clock instruction manual

Post by phild13 »

The FET is probably installed correctly if the directions were followed.

"If it wasn't, there would be other problems as well"

Not necessarily since the FET is used as a switch.
Not sure if I can explain the thought properly though

The P channel FET is connected with a positive source (VCC) and a more negative drain in the circuit schematic. In the FETs off state, the gate would be held at the source voltage (VFDPWR logic high). To turn the fet on, the gate is lowered with respect to the source (vfdpwr logic low). If the clocks power is removed the vfdpwr pin is neither high not low and the FET will also turn off.

Now if the FET is installed backwards, then the internal diode can conduct causing the FET to essentially be on all the time. This probably wouldn't be noticed as we normally want the FET to be on. If the clock is unplugged, the battery runs down trying to power the VFD chip and tube bias.

Another possibility may be a polarized capacitor is backwards, but I think there would be other issues if that were the case, but is worth verifying.

User avatar
jarchie
 
Posts: 615
Joined: Sun Jun 24, 2012 2:16 pm

Re: Icetube clock instruction manual

Post by jarchie »

PhilD13 wrote:The FET is probably installed correctly if the directions were followed.
I was wondering if the microcontroller was turning off the FET while the Vcc is below the brown out-detection threshold of 1.8v. But I suggested testing at R3 to also rule out a misinstalled/malfunctioning FET.

It's also possible that the voltage drop across the battery protection diode is greater in cases of low voltage--something I can't test right now. Rigging a test setup to measure the battery current during sleep is tedious, but might be worth doing.

I would also be curious if voltages on the battery detection diode are different with a brand-new 3v coin cell.
PhilD13 wrote:Another possibility may be a polarized capacitor is backwards, but I think there would be other issues if that were the case, but is worth verifying.
Good call; definitely worth checking.

niznai
 
Posts: 28
Joined: Fri Aug 09, 2013 10:51 pm

Re: Icetube clock instruction manual

Post by niznai »

The FET mounted backwards was addressed in another thread where the person seeking help described the symptoms and they were quite clear. From memory, the screen would light up at plug-in and then turn off shortly. That is how it was discovered some FETs have the text on the curved side and the instruction manual was rectified to include this possibility. If it wasn't such a drag to take pictures and upload here, I would do it, but you have to trust me, I installed mine correctly.

As for electrolytics, I would be indeed very surprised if one was the wrong way around. I'll have another check, but don't think so.

Will check the resistor as suggested by John and the drop on the battery Schottky again. Keep in mind, such a low drop (<.2V) over a diode is not exactly typical from what I have read on the web. My diodes seem to be MBR160 as listed in the parts list, if that says something.

User avatar
jarchie
 
Posts: 615
Joined: Sun Jun 24, 2012 2:16 pm

Re: Icetube clock instruction manual

Post by jarchie »

niznai wrote:The FET mounted backwards was addressed in another thread where the person seeking help described the symptoms and they were quite clear. From memory, the screen would light up at plug-in and then turn off shortly. That is how it was discovered some FETs have the text on the curved side and the instruction manual was rectified to include this possibility.
You may be right, although I know of one user who claims his clock works perfectly with the FET in backwards! (He is however using my firmware... I don't know why that matters.)
niznai wrote:If it wasn't such a drag to take pictures and upload here, I would do it, but you have to trust me, I installed mine correctly.
I still believe you. I suggested the R3 test for other reasons.
niznai wrote:As for electrolytics, I would be indeed very surprised if one was the wrong way around. I'll have another check, but don't think so.
I also doubt you made a mistake here, but still think it's worth checking.
niznai wrote:Will check the resistor as suggested by John and the drop on the battery Schottky again. Keep in mind, such a low drop (<.2V) over a diode is not exactly typical from what I have read on the web. My diodes seem to be MBR160 as listed in the parts list, if that says something.
It depends on the diode. Schottky diodes have much lower voltage drop than power blocking diodes, and all of the diodes included in the kit are Schottky diodes. From Wikipedia, "[a] normal silicon diode has a voltage drop between 0.6–1.7 volts, while a Schottky diode voltage drop is between approximately 0.15–0.45 volts." Voltage drop across a Schottky diode is lower at low current and the Ice Tube Clock uses only ~50 uA during sleep--not much at all.

I remember an observing a voltage drop of 0.15v on one Ice Tube Clock, but do not currently have access to my clock to confirm that measurement. :-( Also, that clock had more battery voltage (>3v on both sides of the diode).

niznai
 
Posts: 28
Joined: Fri Aug 09, 2013 10:51 pm

Re: Icetube clock instruction manual

Post by niznai »

This is one thing that is really worrying me. The fact the batteries drop so much in voltage to me says there is a high drain somewhere. Keep in mind I tried three brand new batteries in this clock and they all display the same massive drop in voltage when in the clock (wall wart unplugged). I can believe one battery was bad, maybe even two at a stretch, but three?! What is the chance of that? The last one I used is a Swiss made high capacity Renata that was twice as expensive as the normal ones (from the kit) I tried before. All batteries are well within their best by date.

I take your point re. Schottky diodes. I will check again more carefully the Schottky on the battery.

User avatar
jarchie
 
Posts: 615
Joined: Sun Jun 24, 2012 2:16 pm

Re: Icetube clock instruction manual

Post by jarchie »

niznai wrote:The fact the batteries drop so much in voltage to me says there is a high drain somewhere. Keep in mind I tried three brand new batteries in this clock and they all display the same massive drop in voltage when in the clock (wall wart unplugged).
Okay, not a bad battery... And I agree that the most likely explanation is a high drain somewhere. In addition to R3, it might also be worth checking the ATmega's BOOST output on battery power with the wall wort unplugged. Even if the ATmega is in a reset condition due to BOD, R5 should pull the BOOST output (Q2 input) low.

If measurements of the BOOST output and at R3 are both zero volts, then I would suspect a high impedance short somewhere.
niznai wrote:I take your point re. Schottky diodes. I will check again more carefully the Schottky on the battery.
I didn't mean to imply that your readings were wrong; they are probably an artifact of highish current and lowish voltage. I did mean to imply that a high voltage drop suggests a high drain somewhere.


Also... I might not be able to access the internet regularly over the next week, but since you have the (mis?)fortune of having four people trying to help you simultaneously, I hope it won't matter.

Locked
Please be positive and constructive with your questions and comments.

Return to “Clock Kits (discontinued)”