Chapter 4 - DSP-10 Software Setup
Much of the DSP-10 Transceiver operation is controlled, or customized by variables contained in the UHFA.EXE configuration file. In all cases default values are provided for this file. Some of the variables can then be altered to tune-up the radio. Chapter 4 suggests procedures that can be used for this purpose. The following sections of this Chapter are external to this file, but available by htm link:
Automating the boot and loading
Configuration File the key to customizing the DSP-10
Pallette and Colors Methods and ideas for customizing the screencolors
Diagnostics Screen information to peek inside the DSP-10
Setting the freq offset of the EZ-Kit A minor refinement in frequency accuracy
RF Gain Configuration Calibrate your S-meter readings at reduced RF Gain
Reference Frequency Calibration - This is to software calibrate your 10 MHz reference oscillator. These calibrations do not apply to the FM mode, and it is important that, in hardware, the oscillator be within about 1 kHz at 144 MHz (about 70 Hz at 10 MHz) to prevent FM problems. If FM is not used, the correction to ref_offset is not constrained in value.
If you are using an accurate 10 MHz external reference , this section does not apply, and your ref_offset variable should be 1.0000000.
It is a number of up to 14 digits that is very close to 1.0. If you do not have a way of calibrating frequency, just set this parameter to 1.0. Here is the procedure if you have a frequency standard available with harmonics that can be received on 2-meters. First, set the parameter ref_offset to 1.0 in the uhfa.cfg configuration file. If, for instance, the standard can be received at 145 MHz (say the 29th harmonic of 5 MHz) then set the DSP10 to to 145.000000 MHz in the USB mode. Use the frequency trimmer on the 10 Mhz reference oscillator, C110, to allow the harmonic of 5 MHz to be seen on the spectral display. Now close up the DSP10 box and let it settle to the normal operating temperature. Observe how much higher than 145 MHz the received frequency is by reading the marker frequency just under the S-meter display. Suppose this is 792 Hz. The factor ref_offset will be calculated as :
145.000792/145.000000 = 1.00000546
This value should be entered into the configuration file as a line ref_offset 1.00000546
It is also possible to correct for harmonics of the standard that are received below 145 MHz. For this case, put the DSP into the LSB mode. If the marker frequency is, say, 480 Hz the factor ref_offset would be:
(145.000000 - 0.000480)/145.000000 = 0.99999669
It is important that the frequency be adjusted with C110 to within a kHz or so of 145 MHz so that the FM center frequencies will be correct. These are not corrected by ref_offset, but all other modes are.
Transmitter Audio Equalizer Setup - Over the years, there has been much discussion about microphones characteristics, and their affect on voice understandability. One must suspect that much of this has to do with frequency response, and a number of articles and products have advocated various ways to optimize the response. Sometimes multi-band "graphic equalizers" are promoted and other times a new and sometimes expensive, microphone is offered as the answer! The DSP-10 has a DSP equalizer in software to let you experiment with this concept.
The equalizer has six bands, with break points at 250, 550, 850, 1200, 1600, 2100 and 2800 Hz. Outside the 250 to 2800 Hz region, the audio is rolled off. The levels for the six bands are stored in the configuration file, under the variable aeqdb_t . There are seven values, as the region below 250 Hz can be programmed. It is recommended that this region be kept at -99, however, so there isn't really a 7th band. The values are the relative responses in dB. Raising all values by 10 dB, for instance, does not change either the response, or the gain of the audio. So one might leave the bnd from, say, 850 to 1200 Hz at 0 dB and make the other values relative to that band.
In order to make the adjustment of the equalizer simpler, there is a dialog box that shows the actual frequency response of the transmitter audio path. This box is opened by the SCRL-F5 keyboard command. The top of the box has a graph showing the microphone equalizer response. The six band relative responses, in dB, are below the graph. As is standard for the dialog boxes, you use the cursor keys to move up and down and then backspace over the number to be changed. Whenever a new line is selected, the new response is plotted on the graph. When the desired response has been achieved, the box can be closed with the ENTER key.
It will be found that the equalizer responses of the six bands tend to overlap. Changing any band tends to change its two neighbors, as well. This is also seen by the response above the 2800 Hz breakpoint that typically is not down 40 dB until about 3500 Hz. This response prevents having a "ringy" sound to the audio, and yet is abrupt enough to achieve almost any needed shaping for the audio.
One final comment on audio response. Theories have been advanced that the vowel sound information is in a narrow band at lower frequencies. The exact band depends on the particular voice. The all important consonants are at high frequencies, perhaps 1500 to 2500 Hz, and these frequencies must not be attenuated. The in-between regions may not be needed and their omission allows more power for the othe two frequency bands. Something to play with!
Transmit/Receive Sequencing - Many stations use some form of time sequencing of the Receive-to-Transmit and the Transmit-toReceive operations. This is usually protection against having the transmitter power damaging relays, receivers or transmitters. Often this is accomplished with a small box of hardware with appropriate timing and decision circuits. The DSP-10 has a programmable sequencer built in software. This should obviate the need for separate timing. However, there are any not relay drivers in the sequencer and these need to be built separately. Here are the rules.1 -The software sequencer is slightly different than the common hardware versions. Once the transition from R to T is started, it will continue up to, but not including actual RF generation. Once T to R is started, it will continue all the way to receive, before it can reverse and proceed to transmit. This should never cause damage, but it is different.
RECEIVE3 - The logic-level drive for the Amplifier and the T/R Antenna relays comes from the accessory plug. All delay amounts are Configuration File programmable in milliseconds with the variable names as shown in 2 above. The delays can range down to 0, appropriate for a barefoot DSP-10, or upward without a limit. Default values are shown after the variable names, above. There is an additional delay that is used only in the EME-2 mode to line up the Moon delay, including the switching delays.
Receiver audio muted, Turn receiver off, Switch
the Ant Relay to Transmit
Turn Amp Control to ON
Place DSP-10 into transmit
Fixed 10 millisec delay
DSP-10 transmit off
Turn Amp Control to OFF
Switch Ant Relay to Receive.
Fixed 2 millisecond delay
Turn on DSP-10 RF hardware
Fixed 20 millisecond delay to remove pops
If ANT, but not AMP or RF, "TRANSMIT" is white.
If ANT and AMP, but not RF, "TRANSMIT" is beige.
If ANT, AMP and RF, "TRANSMIT" is red.
If all three are low, "RECEIVE" is in white.
CALL STRING - Customize your panel a bit by putting your own call between the vertical lines, after call_str, in the UHFA.CFG configuration file.
Table of Contents