What is the recommended maximum drive level for crystals used with the AM1808?
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
What is the recommended maximum drive level for crystals used with the AM1808?
Hi Jeff
FYI, I think the crystal on the LogicPD SOM for the 24 MHz is the following
http://www.citizencrystal.com/images/pdf/m-cs10.pdf
This specifies a DL value of 50 uW, and this is working fine without any known issues.
Also
1) In general the best way to confirm a crystal would be to actually test this out on the board (swapping out the crystal etc).
2) Most crystal application notes mention measuring the Irms value across the crystal to ensure that its power dissipation does not exceed the maximum specified Drive Level value
Do you have crystal in mind that you are considering for your design and what is the drive level specified in the crystal datasheet?
Regards
Mukul
Citizen has discontinued the CS10 and suggests either the TXC-7A or the the ABRACON ABM7 as replacements. At 24 MHz the TXC-7A would have a max ESR of 50 Ohms and a max drive rating of 200 uW. The ABRACON ABM7 would have a max ESR of 50 Ohms and a 100 uW max drive rating.
There are two problems in regards to simply measuring the drive level. (1) The evaluation module doesn't provide a convenient means to do so, and (2) simply measuring it on a board won't give me worst case. Finally, the fact that the 50 uW crystal on the LogicPD SOM is working fine without any known problems doesn't give me much reassurance in terms of long term reliability. Exceeding a crystal's max drive level won't necessarily cause an immediate failure, it could just cause excessive drift or shortened lifespan.
It seems to me, that someone on the design team should be able to provide a worst case drive level. OSCIN/OSCOUT pin capacitance values would also be nice.
LogicPD used two 33 pF Caps in parallel with the 24 MHz, 18 pF load, CS10 crystal in their design, so it appears they were assuming approx. 1.5 pF worth of stray PCB and processor pin capacitance. That seems a bit low. On the other hand, the 12.5 pF load, 32.768 kHz, FSX327 crystal used for the RTC has two 18 pF Caps in parallel with it, which would indicate about a 3.5 pF allowance for stray PCB and pin capacitance. The FSX327 has a drive level rating of only 1 uW. Incidentally, the FSX327 has also been discontinued.
It would make my life much simpler if TI could provide a more complete specification.
Hi Jeff
Jeff Oberski said:It seems to me, that someone on the design team should be able to provide a worst case drive level. OSCIN/OSCOUT pin capacitance values would also be nice.
Thanks for providing more details on the issue at hand. I wasn't aware of Citizen discontinuing CS-10. I will try to check with design team on this and get back to you if we have some analysis to share. Traditionally I don't see this specification provided in most of our device datasheets --> All the same your questions/concerns are valid IMO.
Let me see what I can find.
Regards
Mukul
Thanks Mukul.
By the way, I was mistaken about the FOX 32.768 kHz Crystal being discontinued. Fox has discontinued the FSX family, however the FSXLF family, which is used on the Eval module, is still in production. Sorry for any confusion.
--Jeff
Hi Jeff
I don't have complete answers for you on this. To put together a complete specification (if that is the route we take) will take some time. So my plan to see if we can provide you some tactical inputs.
I am still researching on the 24 MHz crystal query.
For the 32KHz RTC crystal the, spec for the crystal max drive level is 1 uW.
On your query on capacitance , I checked with design team, and following is the response;
For the question of exact cap (based on recommended cap for crystal):1.5 pF worth of stray PCB and processor pin capacitance or 3.5 pF.This is not that critical (unless the you want very exact crystal freq). The package + on chip cap is around 2pF (add expected board cap to this).
Hope this helps.
Regards
Mukul