Published:2011/8/10 1:24:00 Author:Phyllis From:SeekIC
By Burkhard Kainka
The Blueberry Board
For use in the lab the Blueberry Board (CY30703) with the CY27EE16 is particularly interesting. The double ’E’ in the part number gives away the fact that the chip contains an EEPROM. This has space to store 2 kbits for register contents and a further 16 kbits for general program use. No special hardware is necessary to program the device. The clock configuration parameters can be stored either in RAM or EEPROM. Storing to RAM allows the clock to be configured ’on the fly’. After a power down/up cycle the clock will be reconfigured to the parameters stored in EEPROM.
This chip has only one internal PLL but with careful programming of the dividers many different clock frequencies can be generated. To increase flexibility the output dividers can be switched to either the PLL output or to the oscillator frequency and an output cross point switch matrix routes the clocks to the user defined output pins.
RF applications
These signal generators are essentially intended as clock sources for microprocessor systems and their peripherals but we are all in favor of a bit of experimentation so why shouldn’t we try it out for RF applications? The main concern is whether the generator can meet the more stringent requirements of an RF environment. In particular the increased level of phase noise in a PLL system compared to a Direct Digital Synthesiser (DDS) could impact on a receiver’s performance. The acid test would be to use it in place of the synthesised signal at the first injection oscillator input to the digital (DRM) receiver published in March 2004 of Elektor Electronics. For example, to receive DRM transmissions from the German station RTL (broadcasting on 6095 kHz) we require a first oscillator of 6550 kHz (with an IF of 455 kHz, 6095 + 455 = 6550 kHz). After substituting the PLL clock for the DDS clock the resulting performance was surprising good: The DRM signal was completely clean and indistinguishable from the quality achieved with the DDS clock source. Based on this result there seems to be no good reason why the programmable oscillator should not be considered for use in critical RF applications also.
Anyone who has worked with PLLs will be aware of the usual problems: A VCO is controlled by the output signal from a phase comparator and a loop filter, this configuration gives rise to relatively high levels of phase noise and the effects of loop resonance. In critical RF applications LC oscillators with high Q characteristics are used to reduce these tendencies but never the less PLL based receivers usually tend to have higher levels of phase noise than for example, DDS controlled receivers. The performance achieved here is therefore remarkably good.
The VCO Calculator in CyberClocks details how 6.550 MHz is derived from the 10 MHz clock. The Q divider is set to 4, giving a fairly high comparison frequency of 2.5 MHz. The PLL (the P divider) multiplies the frequency by a factor of 131 to produce 327.5 MHz. The VCO output is divided by 50 in the output dividers to give an output of: 10 MHz / 4- 131 / 50 = 6.550 MHz.
The secret of the high quality signal lies in the fact that the VCO and comparison frequency are relatively high and the VCO phase noise on the output signal is effectively divided down by the output divider chain (N). By comparison a conventional PLL system with a channel spacing of 10 kHz would typically use a comparison frequency of 10 KHz with the output frequency sourced directly from the VCO output, this set-up makes it far more difficult to produce a clean signal.
The Blueberry board can generate signals higher than those produced by the DDS of the Elektor Electronics DRM receiver so it is now possible to explore the short wave region around 26 MHz. The receiver design uses a wideband first mixer stage connected directly to the aerial so it is also possible to push the oscillator frequency up a bit more and receive signals in the VHF band. It is perfectly possible to cover transmissions in the aircraft and 2 m amateur radio band; all we need now are some DRM stations up there to listen to.
Reprinted Url Of This Article: http://www.seekic.com/blog/project_solutions/2011/08/10/USB_CYBERCLOCK___For_low_noise_VCOsPLLs__(3).html
Print this Page | Comments | Reading(553)
Author:Ecco Reading(32820)
Author:Ecco Reading(3526)
Author:Ecco Reading(3246)
Author:Ecco Reading(3801)
Author:Ecco Reading(5331)
Author:Ecco Reading(3315)
Author:Ecco Reading(3462)
Author:Ecco Reading(3604)
Author:Ecco Reading(4009)
Author:Ecco Reading(3807)
Author:Ecco Reading(3748)
Author:Ecco Reading(3780)
Author:Ecco Reading(6094)
Author:Ecco Reading(3764)
Author:Ecco Reading(4579)