Halbert.PA@XEROX.ARPA (09/18/84)
W3IKG's message about inaccurate frequency readout on the TS-430S is somewhat misleading. I speak as an owner of a TS-430S who knows the circuitry fairly well. The 430 never gives an incorrect frequency readout on the frequency display. The tuning knob uses an optical shaft encoder and quadrature encoding to send pulses to the microprocessor. The microprocessor counts these pulses. If you spin the knob quickly, a few pulses get lost somewhere, perhaps by the microprocessor which may be busy doing other things. So the tuning rate (kHz/revolution) is not always exactly the same. This means that you can't rely on the black marks on the tuning knob for calibration unless you don't spin the knob at high speed. On the other hand, you can always look at the frequency display, which is always accurate, since it is updated by the microprocessor and reflects the numbers the microprocessor is sending to the frequency synthesizer. So think of the marks on the knob as a styling feature, and not as calibration marks. (I have some mods by W6EED for the 430 to improve the performance of the noise blanker and correct possible AGC overshoot. They just involve changing a couple of resistors. I don't have them here, but will get them soon and send a message next week.) --Dan, N6ICX
jhs%Mitre-Bedford@sri-unix.UUCP (09/18/84)
Well, either the TS-430S DOES give incorrect readings, or else a respected friend of mine doesn't know what he is talking about. He went through the alignment procedure quite carefully, especially after my cousin N4AZY and I both chided him for being nearly 200 Hz off frequency relative to our true frequency counters on older TS120-S and TS-130S rigs. He came back and told us that after much puzzling over the schematic diagram, he had concluded to his extreme annoyance that Kenwood's design does NOT provide true counter operation, but merely reads out its own idea of what the frequency OUGHT to be, as in your "numbers it is sending to the synthesizer". (Clearly, a readout of numbers SENT TO the synthesizer is an "open loop" measurement, not a true readout of synthesizer output frequency. For one thing, what if there is a logic failure in the synthesizer? Where would a 430 owner be THEN? Why, probably out of band, while reading out a nice comforting frequency within the band.) My friend aligned his rig for precise readout accuracy at one point on the 20 Meter band, as the manual told him to, and then found that it was off by about 160 Hz at the other end of the band, as read with a well-known and highly accurate brand of frequency counter. That is to say, the digital readout on the 430 and the counter agreed exactly at the calibration point, and differed by 160 Hz at the other end of the band. Which obviously cannot be attributed to "dial slippage" or "lost pulses". Sorry to disillusion you 430 owners out there! I rest my case. If you still think your 430 gives exact counter readouts, I suggest you get a REAL counter and check it at several points on the dial. If it checks out everywhere, then I would be interested to hear about it and so would N1BHI who pointed out the problem to me originally. (Maybe Kenwood has seen the error of their ways and redesigned the unit to get rid of this very annoying "feature".) I bet you will find a 100 - 200 Hz error at a few points, which should be readily detectable, since their digital readout shows 10-Hz increments. Unless you find out that my information really is incorrect, BEWARE of operating right at the band edge, and BEWARE of the possibility of a logic failure which could put you out of band. To be legal, you need a separate means of checking the 430's output frequency! 73 and apologies for being the bearer of bad news... de W3IKG
karn@mouton.UUCP (09/22/84)
I don't know about the TS-430, but on synthesized VHF radios such as the Kenwood TR-9000 (which I have) there is a mechanism which explains small frequency calibration errors. Very few radios are actually synthesized to such small frequency steps such as 100 hz; this makes the phase lock loop design much more difficult. Instead, they typically have two oscillators which are mixed to generate the final LO signal: a "coarse" synthesized generator which operates on, say, 10 khz steps, and a "fine" conventional analog VCO which is operated over a relatively small range (10 khz) with a control voltage generated by a D/A converter. This means that while the frequency accuracy over a wide range will be fairly good, there may be small (several hundred hertz) errors which cyclically repeat due to nonlinearities in the D/A converter or the VCO (which is often a crystal oscillator and a variable capacitance diode.) There can also be nonlinearities when tuning through a "coarse" frequency step, e.g., a 10 khz boundary on the TR-9000 when the PLL synthesizer changes its output by one step and the analog VCO must sweep from one end of its range to the other. This is particularly noticeable on the TR-9000 in lower sideband. Phil Karn