Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

What can be wrong with this board?

Status
Not open for further replies.

chris99

New Member
Hi,

I'm an amateur toy-fixer by necessity, since my two little kids don't seem to be able to hold things still in their hands for more than a few seconds. Their last victim is a LeapReader, an electronic pen used to add audio interactivity to some special books. It was working fine when suddenly it started behaving strangely, in what seems to be like a slow-motion operation. Basically everything works, but it takes about 3 times more time, with lights flashing slowly and audio slowed down to devilish sounds. After putting it in re-charge for a couple of hours, resetting and repairing it via its companion software without success, I decided to open it and I couldn't find any signs of trouble.

What I found out though is that by holding the pen in a particular position (flat, face down), it works perfectly fine. As soon as I move it by about 20 degrees in any direction it goes back to slow mode. I disassembled everything and there aren't any moving parts left that could cause this. I tried to push and pull pretty much every component and contact point without being able to replicate the effect. It seems like gravity is the only force in action, but I can't identify where exactly lies the problem.

What can I do?
 
Interesting and challenging. I can only guess, but perhaps there is a crystal or resonator in there that is cracked, causing the entire clocking to run slow. The crystal is actually a mechanically vibrating device smaller than your fingernail in a small metal can usually.

Or it could be a hairline crack in a circuit trace?
If gravity is definitely a factor then look for any heavy components that might slightly flex the circuit board and cause an open trace or opening a weak solder joint.
It must be in there somewhere, you might have even pushed on the mystery fault without noticing it!

This might be one in a million, but I have actually discovered an IC that had internal failures based on orientation to gravity. Must have been a bit of metal in a void of the chip package.

I hope you find it, and please let us know what you found! This is a weird one!
 
Interesting and challenging. I can only guess, but perhaps there is a crystal or resonator in there that is cracked, causing the entire clocking to run slow. The crystal is actually a mechanically vibrating device smaller than your fingernail in a small metal can usually.

Or it could be a hairline crack in a circuit trace?
If gravity is definitely a factor then look for any heavy components that might slightly flex the circuit board and cause an open trace or opening a weak solder joint.
It must be in there somewhere, you might have even pushed on the mystery fault without noticing it!

This might be one in a million, but I have actually discovered an IC that had internal failures based on orientation to gravity. Must have been a bit of metal in a void of the chip package.

I hope you find it, and please let us know what you found! This is a weird one!
This could be also one of those rare cases that Murphy was a realist rather than optimist. :)
 
Thanks for the reply. The board is small and very light, so I can't really see any heavy components bending it. I also tried bending in all directions by hand but nothing changes. Here are the pics in case they might help:

Like this it works: **broken link removed** (is the crystal the metal rectangle on the left of the big black round? It's labelled "FT 6.000")

Like this it doesn't: **broken link removed**

What I find amazing is that if I lift it to a vertical position with the long side as base (mid way between the 2 pics), the threshold is very clear at 90 degrees. Tilting it back and forth by 5mm (a few degrees?) make it change speed and you can even hear the sound acceleration, similar to a record going quickly from 33 to 45 rpm, in 10ms.

Is there a way to understand if it's the crystal? Can I replace it myself?

Thanks a lot!
 
There's plenty of surface mounted devices on both sides of the board, a dodgy solder connection on any of them could be your issue. Also, one of them may have a stress crack, particularly capacitors. Apart from a really good visual under a strong magnifyer, I would gently prod each component and chip leg when it's in its non working orientation and see if you can bring it back to life that way. At the least it should narrow your focus to a particular area of the board. Incidentally, the Crystal/ Oscillator is the little metal can right in the middle of your board in photo 1, marked FT 6.00, as you mentioned, it is replaceable, with care and the right tools :)
 
After prodding everything without results, I tried something more aggressive and found out that by shorting two of the oscillator's legs (see pic **broken link removed**) I can repeat the problem, but unfortunately not fix it... I'll try replacing the oscillator then as the last possibility. What are the right tools to do this? I just have an amateur soldering iron and a lot of patience... are they enough?

Thanks!
 
Last edited:
I had something similar with a tv remote, it pulsed the led but the tv didnt recognize it, turned out the ceramic resonator had changed its freq after the owner chucked it at his dog.

If the old osc is dead and you have a new one try heating it with a solder iron while gently pushing it with a pair of tweezers, it should come away when the solder and adhesive has melted.
 
Might be worth trying tap tap tapping on the oscillator can with the pen held at various angles - you might find an angle where it will jump between working and not working. Doesn't fix anything, but it might help pin down the fault a bit more. Also, inspect with a very strong magnifier around the legs you tried shorting. You might find a tiny whisker of something conductive lying around.
 
This could be also one of those rare cases that Murphy was a realist rather than optimist. :)
I really think he always was that.
 
I tried tapping with no success and I'm unable to remove the osc. I tried heating it as much as my iron could do, but I can't find a spot where the knife can slide under and the glue doesn't give in. the osc seems to be fused to the board! Anything else I can try?
 
I just noticed there is another crystal on the other side of the board - little silver cylinder. Might be worth having a poke around there. Look very closely at the end caps of all the components to see if there is corrosion, since it is a hand held device for children this is entirely possible! Also I'm wondering if it's worth giving the board a good wash under the tap, let it dry a day or two before testing again - this cleared up a bunch of faults on a game controller for me.
 
It certainly sounds a crystal is toggling between fundamental and 3rd overtone.

6MHz is possible to make as a 3rd overtone
The cylindrical crystals tend to be watch type 32kHz

It suggest that the 6MHz Xtal Oscillator is at fault. Since it would have the CMOS gate and caps internal, it should not be sensitive to load capacitance or orientation. 3rd overtones are usually thickness regulated to control and sometimes helped with an LC bypass.

So I suggest testing my assumptions;

- apply Finger tip capacitance on either crystal pads
- apply pressure with a pencil erasure tip. on either side of Xtal
- use your hand as an EMI shield touching board common and near xtal.
- gravity might apply to a microscopic disconnect inside but would be affected by minor stress.

My hunch is an electrical contact loss inside the unrepairable 6MHz oscillator from inductor to pad not vibration or EMI or capacitance will fix.

4291950600_1425703178.jpg
 
It certainly sounds a crystal is toggling between fundamental and 3rd overtone.

6MHz is possible to make as a 3rd overtone
The cylindrical crystals tend to be watch type 32kHz

It suggest that the 6MHz Xtal Oscillator is at fault. Since it would have the CMOS gate and caps internal, it should not be sensitive to load capacitance or orientation. 3rd overtones are usually thickness regulated to control and sometimes helped with an LC bypass.

So I suggest testing my assumptions;

- apply Finger tip capacitance on either crystal pads
- apply pressure with a pencil erasure tip. on either side of Xtal
- use your hand as an EMI shield touching board common and near xtal.
- gravity might apply to a microscopic disconnect inside but would be affected by minor stress.

My hunch is an electrical contact loss inside the unrepairable 6MHz oscillator from inductor to pad not vibration or EMI or capacitance will fix.

4291950600_1425703178.jpg
I wonder if the User interface is an IR LED and PD which may change sensitivity with end tilted away from lens. The Xtal @6MHz is typically fundamental mode at this f, but could be modified to make it operate at 18Mhz with sufficient gain in CMOS a inverter. Changing the inverter speed from stray capacitance might change the filter characteristics such that 6MHz dominates and they need 18MHz for CPU Flash wave file player(guessing) or a PLL is used that locks on the wrong harmonic after a large counter divider
 
Last edited:
I wonder if the User interface is an IR LED and PD which may change sensitivity with end tilted away from lens. The Xtal @6MHz is typically fundamental mode at this f, but could be modified to make it operate at 18Mhz with sufficient gain in CMOS a inverter. Changing the inverter speed from stray capacitance might change the filter characteristics such that 6MHz dominates and they need 18MHz for CPU Flash wave file player(guessing) or a PLL is used that locks on the wrong harmonic after a large counter divider

You do realize this thread is well over a year old....?
 
i see.
 
Maybe Tony was precisely tilted thus reacting three times slower as the IC in question...:) :)

A good weekend to everyone.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top