• 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.

Possible bad real time clock?

Status
Not open for further replies.

1-3-2-4

Member
I have a ipod dock with a clock some may know it.. it's the Xtreme Mac Luna.. anyways I've had it about a year now.. but one of the things I noticed is the time keeping on this thing sucks... I was late to work one day because of it.. In about a months time the time will be about 5 min off.. I set the clock about a week ago and it's already about 1 minute off now.

So what I saw inside when I opened it is a 10khz crystal and the 32.768 KHz crystal.

I don't have a oscope to test it but the simple thing to do is replace the crystal?

I was doing some reading online and I found this site about RTC selection

http://www.maxim-ic.com/appnotes.cfm/an_pk/617

"Crystal Selection
There are two types of crystals used for RTCs: 6pF load and 12.5pF load watch crystals. Generally, an RTC that uses a 12.5pF crystal has a timekeeping current of 1.7X more than an RTC that uses a 6pF crystal (that is, 6pF RTC current = 300nA @2V whereas 12.5pF RTC current = [email protected]). Timekeeping current is that measured when there is no serial bus activity and the RTC is only using current to run its 32.768kHz oscillator and count real time. A standard, low-cost, 50ma-hr, lithium, backup battery powers an RTC with timekeeping current of less than 570nA for up to 10 years.

A 12.5pF load crystal oscillator is somewhat more stable and less susceptible to noise and PCB layout stray capacitance than a 6pF load crystal oscillator. This is partially due to the capacitance from each crystal pin to ground, internal to the RTC, which is 25pF per pin for the 12.5pF crystal RTC and 12pF for the 6pF crystal RTC.

There are also inventory issues for 32.768kHz watch crystals. Load crystals at 12.5pF are readily available through many distributors. In contrast, load crystals at 6pF are not as readily available and can require a minimum-quantity order to be purchased. Check with your local supplier for availability and price.

It's important to use the correct specified crystal with an RTC, because the wrong crystal can cause as much as a 100ppm error in the 32.768kHz oscillator frequency, which translates to a 4.3-minute error over a one-month period. In addition, the wrong crystal can cause excessive timekeeping current or failure of the oscillator to start properly."

hmm 4.3 minute error over a one month period.. sounds almost like my issue..
 

Attachments

Last edited:

gabeNC

Member
Sounds like typical clock drift to me.

Clock drift - Wikipedia, the free encyclopedia

Even some of my very expensive large mainframe servers at work drift, some worse than others. That's why we use NTP to synchronize everything. I wouldn't imagine that a somewhat cheap consumer device would have that much added circuitry for extremely accurate clocking.

You can purchase clocks that mount on your wall and have remote atomic clock sync.

Amazon.com: Low Vision Atomic Solar Wall Clock - Self Setting: Health & Personal Care
 

1-3-2-4

Member
I don't care about it being 100% right 100% of the time.. I don't feel I should be loosing 1-2 min of time each week.. It will never gain time but will always get slower and slower.. at one point it was 5 min behind in just 3 weeks!
 

unclejed613

Well-Known Member
Most Helpful Member
if you use a trimmer cap, a cap between one pin and ground will slow it down, a trimmer cap in series with the crystal will speed the clock up, just keep that in mind when you go to make any changes in the circuit. if it's running slow, the cap should be in series, which increases the frequency.
 

1-3-2-4

Member
well I changed the crystal first.. I brought a cheap clock which also gave me 2 free tact switches... but replaced the crystals I will report back in a few weeks on how it's going.

first picture is from the new clock taken apart and the 2nd is where it went into.
 

Attachments

Last edited:
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top