QYT KT-8900 > Amateurtele.com

The built quality is rather bad, for example the crystal oscillator wasn't placed on the printed circuit board but a couple of millimetres off the board and also placed askew. The built quality is therefore rather messy. The design quality is also rather bad. The data wires for communication with the microphone is placed directly next to the microphone wire. Therefore the pule width modulation of the digital square waves an be heard easily. These wires should be shielded individually.

The good thing is that the rig is rather small and therefore doesn't take up much space in the garbage bin.
My advice is: "Don't buy one, it isn't worth the money. Spend a little more on a "real" radio and you won't regret it."

package

1 x QYT KT8900 radio

1 x DTMF microphone

1 x Car Power Cable

1 x Fuse

1 x Mounting bracket

1 x Screw set

1 x User"s Manual

specifications (by the manufacturer)

TX/RX frequency range: 136...174 MHz (VJF) and 400...480 MHz (UHF)

Memery channels: 200

Channel Spacing: 5 / 6,25 / 10 / 12,5 / 15 / 25 KHz

Supply voltage: 13,8 VDC +/-15%

Squelch way: CTCSS / DCS / 5 Tone / 2 Tone / DTMF

Frequency Stability: ±2.5ppm

Operating Temperature: -20...+60 'C

Dimension: 98 mm wide, 35 mm high and 118 mm deep

Weight: 408 grams

Bandwidth selection: wide and narrow

Sensitivity: ≤;; 0,25 μ;;V (wide) / ≤;; 0,35 μ;;V (narrow)

Channel choice: ≥;;70 dB (wide) / ≥;;60 dB (narrow)

Intermodulation: ≥;;65 dB (wide) / ≥;;60 dB (narrow)

Spurious Rejection: ≥;;70 dB (wide) / ≥;;70 dB (narrow)

Audio response: +1...-3 dB @ 0,3...3 kHz (wide) / +1...-3 dB @ 0,3...2.55 kHz (narrow)

Signal to noise ratio: ≥;;45 dB (wide) / ≥;;40 dB (narrow)

Audio distortion: ≤;;5 %

Audio output power: ≥;;2 W @ 10%

Output power: 25 W / 20 W (VHF/UHF)

Modulation Mode: 16KΦ;;F3E / 11KΦ;;F3E

Channel power: ≥;;70 dB / ≥;;60 dB

Signal to noise ratio: ≥;;40 dB / ≥;;36 dB

Parasitic harmonic: ≥;;60 dB / ≥;;60 dB (not...)

spurious radiation on VHF

According to the Dutch "Gebruikersbepalingen amateurfrequentiegebruik" number 212-3.3 of the 20th of august 2012 part 3.1 it's not allowed to radiate spurious signals of -36 dBm or -60 dBc or more between 50...1.000 MHz wherever the highest number is applicable. The highest harmonic (at 435 MHz) is -29,55 dBc (decibel relative to the carrier). This is approximately 30 dB (1.000 times) worse than allowed. The transmit power during the test is 20 Watts. At 435 MHz this is approximately 30 dB less of 20 Watts. This is approximately 20 mW / 13 dBm. The maximum limit is -26 dBm. Therefore the spurious radiation doesn't meet the legal limits. The third harmonic is the worst one which is represented at the 70 cm band. (144 MHz * 3 = 432 MHz and 146 MHz * 3 = 428 MHz) When transmitted at the 70 cm band, the -60 dBc limits are met at a first glance and therefore shouldn't be a problem.
To meet the regulations on VHF an additional low pass filter should be used! I tried to replace the filter inside the radio, but without success...

image is loading...

measurement setup
image is loading...

The used equipment is:
- Rigol DSA-815 TG spectrum analyser.
- Radio System AB RS-9694 measuring coupler.
- Spinner BN 52-77-66 400 W dummy load.

It seems there are two (8-pole?!) low pass filters on the circuit board. The antenna signal is split in two ways and switched by PIN diodes. Therefore the VHF and UHF path should be able to suppress harmonics. I used the computer program ELSIE to determine what suppression should be able to get on VHF. I designed a 7-pole Chebycheff low pass filter for 148 MHz with a passband ripple value of 0,1. The component values are rounded to the nearest real world values. The suppression for 290 MHz is 55 dB, at 435 MHz 82 dB and at 580 MHz is the suppression 100 dB. If there is no loss at the carrier, the dB values can be considered as a negative dBc value. Since the third (theoretical strongest) harmonic is the strongest for real, it's possible the applied low pass filter doesn't work what it's designed for.
My plan was to bridge the VHF low pass filter en measure the results and do the same measurement without the filter bridged (original configuration). Than it could be determined if the filter is working at all. I tried this and without success. It's likely the replacement of inductor (L) and capacitor (C) components will fix the problem. Since the crappy quality I decided not to spend more time on this rig. But I guess a replacement filter will work and solve the spurious radiation problem.

Replacement filter sketch:
image is loading...

Het (voort)bestaan van deze website wordt mede mogelijk gemaakt door aanbieden van onderstaande reclame.

CTCSS test

I tested the CTCSS sensitivity. I set the receiver to the desired 77 Hz tone. I set the R&S SMT-02 signal generator to a 1 kHz audible tone with 2,5 kHz deviation and a second 77 Hz (sub)audio tone with a deviation of also 2,5 kHz. The -100 dBm signal from the generator is fed into the antenna port and the received 1 kHz tone is audible. The 77 Hz tone of the generator is changed to determine how critical the CTCSS circuit is. The squelch is opened between 76,3...78,1 Hz. This means that there is a tone span of 1,8 Hz for the 77 Hz tone. This seems good.

digital noise on audio

There's a flaw in the design of the radio, or should I say in the microphone? When the volume is turned off (knob fully CCW), and the [MONI] button is pressed to open the squelch a digital noise is coming from the speaker. When the microphone is unplugged, the noise is gone. This flaw was reported earlier by IK1ZYW. It's likely this noise is generated by the digital signals between the microphone and the radio. The buttons are connector to logic that "translates" the button pressing into a digital signal that can be handled by the CPU. This saves a lot of wiring. Adding capacitors toe reduce harmonics of the digital block signals doesn't work since the digital signals are affected. Every wire from a board should have filtering capacitors or should be shielded. Due to the lack of shielding, the microphone wire should be replaced by one with shielded wires. Due to the quality of this rig it isn't the time and money worth repairing this design flaw...

Het (voort)bestaan van deze website wordt mede mogelijk gemaakt door aanbieden van onderstaande reclame.

reverse engineering

output stage amplifier
The final stage of the power amplifier is based on a wideband AFT09MS015NT1 N-channel RF power LDMOS transistor. This component is designed for mobile two way radio equipment with frequencies from 136...941 MHz to generate 16 Watts of RF power at 12,5 VDC. The LDMOS is fed with a RF signal of 230...380 mW (depending on VHF of UHF use) and has a gain of 16...18,5 dB. The output power is approximately 15...16 Watts @ 12,5 VDC. It has integrated ESD protection and has high lineairty for TETRA and SSB use. Since the KT-8900 is designed for 13,8 VDC, the output stage amplifier is running on higher voltages and therefore has more output gain and more heat generation...

radio microcontroller
The 'beating heart" of the KT-8900 is a RDA1846 chip. This is a highly integrated single CMOS chip microcontroller for two way radio applications like portable transceivers. This chip handles the translation from RF carrier to voice in the RX path and from voice to RF carrier in the TX path completely. The RDA1846 has a digital signal processor, which makes it have optimum voicequality, flexible function options and robust performance under varying reception conditions according to the RDA datasheet. The chip can handle signals of 134...174 MHz and 400...500 MHz. This chip is also used in the Baofeng UV-5R portable radio. Altough it is a powerfull chip with a lot of processing power, it's only 5 by 5 mm (QFN package). There's an onboard VCO, supports AFC, AGC, pre/de-emphasis, VOX, squelch, CTCSS, DCS, DTMF, 8 GPIO's for external controls, I2C bus interface, 8 dBm RF "power" amplifier and analog /digital volume control possibility. That's a lot of smart things stuffed in this small package...

Het aanbieden van de kennis via deze website kost tijd, energie en geld. Het is heel leuk om kennis te kunnen delen en anderen hiermee te kunnen helpen, maar ook alle kosten worden uit privé middelen voldaan. Er zijn onder andere kosten voor de hosting aanbieder om de site te kunnen laten bestaan. Mijn doel is om de inhoud aan te kunnen blijven bieden zonder reclame, maar de kosten worden hoger waardoor financiële middelen nodig zijn. Dus een bijdrage van €3 wordt zeer op prijs gesteld. Het is mogelijk om op een veilige manier een financiële bijdrage van €3 (of een veelvoud hiervan) te kunnen leveren. Klik op de knop hieronder om via Ko-fi een donatie te verzorgen.

open de afdrukbare pagina door hier te klikken
klik hier voor de mobiele versie. (Beta test.)