View AN2052_8730883.PDF datasheet online --- IC-ON-LINE

Datasheet File OCR Text:

november 2014 docid10909 rev 2 1/13 13 AN2052 application note vipower?: implementing frequency modulation on the viper53-e to improve emi emissions introduction this report describes how the frequency modulation reduces t h e electromagnetic interference on an smps using the viper53-e as primary pwm-switch. the comparison of emi measurement data are presented in the following sections. www.st.com
contents AN2052 2/13 docid10909 rev 2 contents 1 test results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 circuit and description of test configurati ons . . . . . . . . . . . . . . . . . . . . 5 3 analysis and possible further development . . . . . . . . . . . . . . . . . . . . . . 8 4 optimization of external oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 5 reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 6 revision history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
docid10909 rev 2 3/13 AN2052 test results 13 1 test results figure 1. original, time average at 200 khz, 53.42 dbuv/m figure 2. fm configuration ?a?, time average at 200 khz, 48.07 dbuv/m
test results AN2052 4/13 docid10909 rev 2 figure 3. fm configuration ?b?, time average at 200 khz, 43.92 dbuv/m as shown in figure 1 , 2 and 3 , using the fm technique, an improvement is obtained. when frequencies are lower than 1 mhz, the device switching is the primary noise generator. using the configuration ?a? or ?b?, a 5 dbuv/m drop is obtained for frequencies lower than 200 khz. between 1 mhz and 10 mhz, both configurations ?a? and ?b? make a dramatic drop bigger than 15 dbuv/ m. above 10 mhz, fm technique benefits are reduced and capacitive coupling between copper tracks in pcb is dominant. table 1. emi comparison between original ?a? and ?b? configurations <1 mhz 1 mhz to 10 mhz >10 mhz original marginally pass pass enough margin ?a? pass more margin enough margin ?b? more margin more margin enough margin
docid10909 rev 2 5/13 AN2052 circuit and description of test configurations 13 2 circuit and description of test configurations figure 4. original configuration (fixed frequency at 70 khz, and no fm) figure 5. configuration "a" (switching frequency from 63 khz to 77 khz fm cycle 20 ms, 50 hz) figure 6. configuration "b" (frequency from 63 khz to 77 khz, fm cycle 1.4 ms, 700 hz) n 9gg 9gf ,& p$pd[ 26& 9,3hu +] n n n 9gg 9 ,& p$ 26& 9,3hu 2vfloodwru +] n n 9 9 9 9shdn ,& p$ ,& p$ 9,3hu 26&
circuit and description of test configurations AN2052 6/13 docid10909 rev 2 by changing the resistor value between vdd and osc pin, the frequency oscillator (capacitor value is fixed at 220 nf) can be modified. in this way the fm percentage can be set. the target fm percentage is +/-10%. in the original circuit a 10 kohm resistor is used, obtaining a switching frequency of 70 khz. this resistor value gives a charging current capacitor of 1 ma max. the external circuit works to increase the switching frequency. therefore, the resistor value should be changed between vdd and osc pin in order to reduce t h e minimum switching frequency to 66 khz. as a result the new resistor value is 11 kohm, obtaining a charging current of 0.9 ma when the external voltage is 0 v. the switching frequency variation depends on the charging current capacitor, therefore it is proportional to the external voltage. maximum value of the switching frequency is 77 khz: it is obtained when the charging current is 1.1 ma where 0.9 ma is given by v dd and 0.2 ma is supplied from the external circuit. below, the design of the resistor value, between osc pin and the oscillator, is explained: ? in configuration ?a?, maximum voltage is 374 v, so the resistance is r = 373 v/0.2 ma = 1.8 mohm ? in configuration ?b?, the maximum voltage is 10 v, so the resistance is r = 10 v/0.2 ma = 50 kohm it is important to verify the frequency obtained using the oscilloscope. this frequency value has to be tuned to the desired value. in figure 7 , 8 , 9 and 10 the typical clock signal is shown. figure 7. configuration ?a?: ac signal (50 hz oscillator) figure 8. configuration ?a?: pwm ramp signal
docid10909 rev 2 7/13 AN2052 circuit and description of test configurations 13 figure 9. configuration ?b?: oscillator signal (700 hz oscillator) figure 10. configuration ?b?: pwm ramp signal
analysis and possible further development AN2052 8/13 docid10909 rev 2 3 analysis and possible further development in figure 11 , 12 and 13 the peak value data of the three tested configurations are shown. figure 11. original peak figure 12. configuration ?a? peak
docid10909 rev 2 9/13 AN2052 analysis and possible further development 13 figure 13. configuration ?b? peak one remarkable point is that the peak value of the three tested configurations is almost the same. the major difference between peak and average measurement is in the filter and in the measurement time. wide-band filter and short measurement time are used to capture the peak value; while narrow-band filter and relatively long measurement time are used to capture the average value. through web search engines it is possible to find the definitions for the terms ?peak? and ?average? as ?spectrum peak? and ?spectrum average?, respectively. these definitions are different from the ?time-peak? and ?time-average? defined in this report. instead, the term ?frequency m odulation? can be found on the web as ?spread spectrum ? . the spectrum peak is the highest component of the frequency spectrum. in emi test data of smps, it corresponds to a low frequency. the spectrum average is the average noise over the whole frequency spectrum. the time peak is the ?peak? data measured by most emi test lab. noise, in terms of dbuv/m, is measured through a wide-band filter within a short period of time. analysing such measurements, the fact that fm does not bring benefit is highlighted. the measurements obtained are very similar to the spectrum envelope of figure 11 . the time average is the ?average? data measured by emi test lab. noises are measured through a narrow-band filter, over a relatively long period of time. note that there is no regulation on the measurement average time.
optimization of external oscillator AN2052 10/13 docid10909 rev 2 4 optimization of external oscillator targeted to lower down the average value of emi noise, the pwm must go through the whole frequency range within the testing time. in other words, the fm cycle is as short as possible. on the other hand, since the pwm signal is modulated by the fm cycle and switching cycle, to avoid audible noise, fm frequency should be lower than 2 khz. in fact the average data of configuration ?b?, 700 hz oscillator, are lower than the configuration ?a?, 50 hz oscillator. another drawback of configuration ?a? is the dependency on ac mains, which can be an issue in the transformer design.
docid10909 rev 2 11/13 AN2052 reference 13 5 reference figure 14. oscillator with 2 bjts, low cost and start-up voltage down to 3 v
revision history AN2052 12/13 docid10909 rev 2 6 revision history table 2. document revision history date revision changes 12-nov-2014 2 updated the title in cover page. content reworked to improve readability, no technical changes.
docid10909 rev 2 13/13 AN2052 13 important notice ? please read carefully stmicroelectronics nv and its subsidiaries (?st?) reserve the right to make changes, corrections, enhancements, modifications, and improvements to st products and/or to this document at any time without notice. purchasers should obtain the latest relevant in formation on st products before placing orders. st products are sold pursuant to st?s terms and conditions of sale in place at the time of o rder acknowledgement. purchasers are solely responsible for the choice, selection, and use of st products and st assumes no liability for application assistance or the design of purchasers? products. no license, express or implied, to any intellectual property right is granted by st herein. resale of st products with provisions different from the information set forth herein shall void any warranty granted by st for such product. st and the st logo are trademarks of st. all other product or service names are the property of their respective owners. information in this document supersedes and replaces information previously supplied in any prior versions of this document. ? 2014 stmicroelectronics ? all rights reserved

▲Up To Search▲

Price & Availability of AN2052

	To Download AN2052 Datasheet File
If you can't view the Datasheet, Please click here to try to view without PDF Reader .