Date:2010-09-10 05:06:33 (9 years 2 months ago)
Author:Werner Almesberger
Commit:df071c3398fed1aa6a3a64eea67b2d4b3a1f993d
Message:New ECN about clock circuit of atusd boards. Measured (absence of) effect of adding an inductor to the clock line.

- ecn/INDEX, ecn/ecn0005.txt: new ECN discussing the switch to a
capacitative clock voltage divider. With frequency measurements.
- atusd/tests/send-20100910: measurement of transceiver performance when
placing an inductor in the 16 MHz clock line.
Files: atusd/tests/send-20100910 (1 diff)
ecn/INDEX (1 diff)
ecn/ecn0005.txt (1 diff)

Change Details

atusd/tests/send-20100910
1Same test setup as for send-20100909. Board 100908-A was modified to have
2a 220 nH inductor in the C7 position, instead of a 0R resistor.
3
4Distant Local
5--------------- --------------- ---------------
6100813-red 100813-orange 100908-A (with inductor)
7
8TX 255, -91 255, -85
9TX 255, -90 255, -85
10TX 255, -91 -
11TX 255, -90 255, -85
12TX 255, -91 garbled
13
14- TX 255, -47
15- TX 255, -47
16- TX 255, -47
17- TX 255, -47
18garbled (-90) TX 255, -47
19
20255, -82 255, -43 TX
21255, -82 255, -43 TX
22255, -82 255, -43 TX
23255, -82 255, -43 TX
24255, -82 255, -43 TX
25
26Control experiment:
27
28100813-red 100813-orange 100908-B (unmodified)
29
30TX - 255, -83
31TX garbled (-91) 255, -86
32TX - 255, -86
33TX garbled (-91) 255, -86
34TX - 255, -86
35
36255, -83 255, -43 TX
37255, -84 255, -43 TX
38255, -85 255, -43 TX
39255, -84 255, -43 TX
40255, -84 255, -43 TX
41
42
43Summary
44-------
45
46Direction No inductor (1) With ind. (2) Change
47------------------------------- --------------- ------------ ------
48100813-red -> 100813-orange 255, -87 255, -91 -4 dB
49100813-red -> 100908-A 255, -88 255, -85 +3 dB
50100908-A -> 100813-red 255, -85 255, -82 +3 dB
51
52100813-red -> 100813-orange 255, -90 - ?
53100813-red -> 100908-B 255, -90 255, -86 +4 dB
54100908-B -> 100813-red 255, -87 255, -84 +3 dB
55
56(1) Data from send-20100909
57(2) Only 100908-A was modified. Changes in the measurements with
58    100908-B show inaccuracies of the test setup.
59
60
61Conclusion
62----------
63
64The differences observed are the same for the modified 100908-A board
65and the unmodified 100908-B board. The inductor therefore does not seem
66to have a significant effect on transceiver performance.
ecn/INDEX
440002 Done Add load capacitors to 16 MHz crystal
550003 Edit Replace balun and filter with integrated balun
660004 Edit Take into account layout considerations for RF
70005 Edit Correct atusd clock voltage divider
ecn/ecn0005.txt
1Correct atusd clock voltage divider
2
3
4According to section 9.6.3 of the data sheet, an external clock supplied
5to the AT86RF230 has a minimum peak-to-peak voltage of 400 mV and a
6maximum of only 500 mV. Furthermore, the signal must be DC-free.
7
8The voltage divider in the 20100903 and 20100908 designs is too sensitive
9to interference and the PLL constantly unlocks, rendering the device
10dysfunctional.
11
12A more robust divider circuit can be obtained with the following
13replacements: C7 -> 0R, R3 -> 33 pF, R4 -> 220 pF. After reworking the
1420100908 boards, they no longer suffer PLL unlocks.
15
16A Qucs simulation of the circuit can be found in ../atusd/sim/cdiv.sim
17
18Frequency measurements yield the following results:
19
20Host Board Error Meas. accuracy
21            (f, ppm) (ppm, nom.)
22------- --------------- --------------- ---------------
23Ben #2 20100908-A +2 99.7
24            +1 99.6
25Ben #1 20100908-A +23 99.7
26            +24 99.9
27Ben #1 20100908-B +24 99.4
28            +24 99.7
29Ben #2 20100908-B +2 99.8
30            +2 99.9
31
32The Ben's 12 MHz crystal has a tolerance of +/- 30 ppm, which is better
33than the +/- 40 ppm required by IEEE 802.15.4.
34
35What remains to be verified is whether this change causes interferences
36that may affect transceiver performance and may also violate emission
37regulations.

Archive Download the corresponding diff file



interactive