Review: RCI 2950DX Dual Band Amateur Radio
The RCI 2950 has been around for many years. The first radio of this type from Ranger Communications was the RCI 2900. The 2900 radio was a CPU type radio with an analog S/RF meter. Comparing the 2900 to the 2950, the meter was the only redeeming quality of the 2900. The 2900 radio was a complete failure. There were modifications on top of modifications to attempt to make these radios work.
One was so extensive, it required numerous parts and circuitry changes, taking a couple of hours to complete. After all the upgrades, the radio was still not very good. In fact as soon as the 2950 was released they no longer recommended the upgrade, they suggested radio replacement. I imagine some 2900 owners received a free 2950 as a replacement if they complained loud enough.
Ranger Communications has continued to improve their engineering and manufacturing techniques over the years. They manufacture most of the 10 Meter radios on the market today. Their factories manufacture radios for many other US based companies including Galaxy. They have been known to copy or clone radios like President Jackson and Grant, this information is only included to inform, not to praise them for doing so.
In fact, the 10 Meter chassis used in all their radios have evolved from that original Uniden design. The dual final, AM modulator/regulator, mixer, receiver stages are all variations of that original 120 channel Grant radio. The designs quickly included circuitry to increase bandwidth to cover the increasing frequency range. Other variations were to adapt to obsolete components as years passed.
Now Ranger Communications has improved the 2950 with completely redesigned CPU and main circuit boards. From the outside, you may think it’s the same old 2950. All the controls and buttons are in the same location as in its predecessor. The face is gray instead of black and of course, it’s labeled 2950DX. Many operators thought the only difference was the dual-band coverage. This wouldn’t have taken a complete redesign; there were Mirage 2950s that covered 24.0000 MHz to 30.0000MHz. The only difference between the RCI and the Mirage was the CPU.
The mainboard was pretty much the same. However, the real news is that the 2950DX is improved in many areas. Not too many people are aware of why the 2950 radio was really redesigned. The driving force was Sony discontinuing the PLL chip used in this radio. The chip was discontinued some years ago. RCI must have made one last large purchase of the chips to keep them going as long as they did.
SMT or surface mount technology is used in this new generation Ranger. Many radio shops haven’t been receptive to the surface mount technology used in these radios. They are more difficult to repair, if not impossible for some shops. To combat this Ranger changed the warranty period from one year to two years, in order to show their faith in this newer technology. It seems to be working, sales of these radios have increased in recent months.
RCI is using this CPU and mainboard combo in a number of radios. The 2950DX-30 Watt mobile, 2970DX-150 Watt mobile, 2985DX-30 Watt base station, the 2990DX-150 Watt base station, and I’ve been told the 5054DX 6 Meter mobile. The board even has provisions for producing an AM/FM only version, which has caused this issue of CB World Informer to run late. This will be explained later in the review.
The first look inside is memorable. This doesn’t look like any of the other RCI HF radios. There is RF shielding on the main PC board for VCO circuits, TX mixer circuit, and RX front end, mixer and initial IF stages. All the tin shields are punched for adjustment access and are stamped with the adjustment designators.
Quite impressive, this hasn’t been done since the late 1970s. Two radios come to mind that had this level of shielding, the CPI CB radio line, and the Stoner PRO-40 SSB rig. These, of course, were CB radios, but they were top of the line rigs and operated in the sister band of 11 Meters.
A look at the 2950DX schematic revealed a double-balanced mixer stage in the receiver. This is found in HF rigs costing many times the price of the radio. This design is used for better intermodulation rejection. RCI claims other receiver enhancements that improve receiver sensitivity. Also revealed, the schematic contains a seven-transistor noise blanker. We’ll put that baby to the test at the shop’s noisy location.
The Ranger specification sheet indicates that the meter is capable of reading modulation, no function found to perform this measurement. This is a misprint. The frequency stability is listed as .001%. If this specification is true, this makes it one of the most frequency-stable 10 Meter radios available.
It’s now time for the stock radio bench test. The first results are with the RF power control set to the full power clockwise rotation. The AM and FM output is at 9-Watts. The AM modulation swing is 22-Watts peak. The sideband power is a 27-Watts PEP. Now the output readings with the RF power control in the low power counter-clockwise rotation. The AM/FM power dropped to 1-Watt. The AM modulation swing is a 2.5-Watts PEP. The sideband power dropped to 4 –Watts PEP.
Testing the receiver indicated that the sensitivity was quite good on all modes. Even very weak signals on AM and sideband were cleaner than on the old 2950 and on most 10 Meter radios for that mater. Now satisfied with the sensitivity of the receiver, the selectivity was then checked. My crude method is to crank up the signal generator to full RF output, modulated to 100% with a 1KHz tone. This registers 30 dB on a calibrated “S” meter or is equivalent to a signal of someone less than a ¼ mile away.
On AM & FM, the radio performed much better than expected. The old 2950 didn’t do very well with this test. Sideband, however, displayed a signal of approximately 4 bars on the LCD meter 200KHz on ether side of the center frequency before the strength started to decrease. This is to say, if someone was coming in at 30dB on channel 20, the signal from that transmission would still be received at four bars on channels 1 and 40.
Being puzzled and thinking this must be a defective radio; a second 2950DX was pulled from its box and tested. The results were identical. To confirm the test equipment was operating properly, two radios were tested, a Uniden Grant and a Galaxy 88. Both radios tested fine. Now the story gets more interesting, a call to JR at Ranger service didn’t resolve the issue. JR said they could not reproduce the results at their service lab.
He went further to say the ARRL tested the RCI 2970DX and found no problem of this type. As concern grew, a decision to contact the author of the RCI 2970DX review in QST magazine was made. Contacting Wayne Irwin was a pleasant experience. After explaining that he wasn’t responsible for the lab testing, he offered to find what he could and reply. Wayne agreed the transmitter unwanted sideband figure of 39dB (50dB is a minimum figure one would expect) could be due to the design issue findings.
Here is the response from Wayne:
Hi Again Bob,
I just checked with Joe Bottiglieri in the editorial office. It appears that the problem you found with the Ranger was not apparent in our lab tests. Since you have seen it in a couple of different radios, he suggests that you consider submitting a little piece for possible publication in Hints and Kinks.
Again, many thanks for your feedback.
Wayne K. Irwin, W1KI
Assistant to the ARRL VEC Manager
Tel: (860) 594-0305
The E-mail states the problem wasn’t apparent in their lab tests, but they don’t deny a potential problem. Looking at the receiver test results, a test for selectivity was done on FM, but no SSB selectivity test results were published. The assumption may have been made, that if FM were tight SSB would be better. I wrongly made this assumption myself on all the radios tested prior to this article. I will cover what I feel is a design flaw and what is needed to correct the potential problem in the next article, titled Image Rejection Modification.
I have also tried to contact Gordon West about his findings while reviewing the RCI 2970DX for Popular Communications. I haven’t received a response yet.
Continuing the review with a modified IF stage showed the 2950DX to achieve excellent results on the sideband selectivity test. Compared to the other 10 Meter radios, the RCI 2950DX performance was outstanding, and this one covers 24 MHz to 32 Mhz. Until now, radios selectivity suffered more as frequency coverage increased.
The RF power control works well and tracks well with the modulation and output limiter circuits very well as long as the limiter isn’t removed altogether. This radio sounds so good, I don’t recommend eliminating the limiter circuit. Great results can be had by turning up the SSB power, SSB limits, and AM modulation controls up fully.
If you were looking for a radio to make a lot of noise with, many other radios would be a better choice. This is a great all-mode radio, for operators who are looking for good clean communications, don’t spoil it with the old tricks to get every last milli-watt out of it. Square wave audio isn’t becoming of this one.
Testing the receiver on the air dramatically demonstrated the difference between the RCI 2950DX and its predecessor. All modes sounded clear and crisp. Even very weak signals were easy to understand. Single sideband is especially natural-sounding, with greater sensitivity and lower noise than any other 10 Meter radios I’ve tested to date. The only fault with the radio is the meter. It flickers on AM and SSB not holding long enough to achieve accurate readings.
This was corrected by the addition of a 4.7uf capacitor connecting the positive lead to the MT signal or the band side of D14 and the negative side of the capacitor to the ground. This only affects the meter portion of the radio, both the incoming signal, and outgoing power
Frequency stability is super. The radio tested drifted only 30 Hz from power on to one hour of running. The clarifier is very easy to tune whether it’s unlocked or not.
The display is unchanged with its large six-digit frequency readout. The backlighting has changed from amber to green. This green lighting is very evenly distributed and it almost appears to be of the electroluminescent type used in the Cherokee and Cobra faceplates, but it’s not. The schematic indicates two incandescent lamps, but the lamps are in a sealed unit and an ohmmeter reading leads me to believe there may be banks of LED in the light panel. I hope this to be true, LEDs will last much longer.
If you’re familiar with the old 2950, you’ll know how all the functions work, they haven’t changed much. The following figure indicates the front panel controls, switches, and indicators.
Press PRG, then press MEM, then enter frequency, then press ENT. To select any of the 0-9 memories, press MEM each time to step to the next desired memory channel.
Press PRG, then press SCAN (SCAN+ will appear on the display), then enter frequency for the high scan limit, then press ENT, then press SCAN (SCAN- will appear on the display), then enter the frequency for the low scan limit, then press ENT. Unlike the 2950, setting the scan limit doesn’t limit the frequency operation outside the scan mode, even though the DX user manual states it will.
Press PRG, then press SPLIT, enter the desired frequency split, then press ENT.
Memory channels are accessed by pressing the MEM button. Each time the MEM button is pressed, the memory advances one channel. To exit the memory channels, press the MAN button.
Two scan options are available, memory scan and frequency scan. Frequency scan will scan between the high and low limits programmed. If no limits are programmed, the radio will scan the entire frequency range of the radio. Squelching the radio starts the radio scanning.
Once a signal breaks through the squelch, the radio pauses until the signal no longer breaks the squelch for about a second. Pressing the SCAN button once indicates a SCAN(+) on the display and causes the radio to scan up. Pressing the SCAN button again indicates SCAN(-) on the display and causes the radio to scan down.
The split frequency is selected by pressing the SPLIT button. Pressing the SPLIT button once indicates SPLIT(+) on the display and causes the amount of frequency split programmed to be added to the receiver frequency during transmission. Pressing the SPLIT button again indicates SPLIT(-) on the display and causes the amount of frequency split programmed to be subtracted from the receiver frequency. Pressing the SPLIT button a third time shuts off the split feature.
If you’ve never used 2950 before, it comes with a user manual that Wayne from QST correctly regards as adequate that will help you master the functions. If you liked 2950, you’ll love the 2950DX. This is a great dual-band radio for the money; I recommend it for anyone that takes radio seriously. If you weren’t impressed with the old 2950, like myself, I think you’ll have a different outlook on Ranger Communications Inc. after trying one of these! Bob F