DMR Amplifiers

Digital ATV - The latest generation, cutting edge ATV - Please discuss it all here.
Forum rules
This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.

Thank you
Post Reply
g0mjw
Posts: 2330
Joined: Sat Sep 20, 2014 9:15 am

DMR Amplifiers

Post by g0mjw » Sat Sep 25, 2021 11:13 am

I have had a few questions now about the DMR amplifier I showed on the BATC QO100 net a while ago.
Picture1.png
Picture1.png (611.7 KiB) Viewed 1693 times
These are usually based on the RA modules, with the 80W UHF version based on the 60W Mitsubishi module. Once you have got over the specification inflation, they are actually fairly decent amplifiers.
Picture2.png
Picture2.png (785.18 KiB) Viewed 1693 times
There are a few problems. Firstly they are RF sensed which may not be what you want and often not sensitive enough for DATV use. Secondly they are prone to oscillate. Thirdly there is a significant through loss on receive and finally, there is no low pass filter. Oh yes, and they have an SO239 output connector that nobody should not replace.

The first thing to do is replace the SO239 with a proper RF connector, usually an N-type. Probably the next thing to do is fit a low pass filter to the output. There is space inside the case, or this can be fitted externally.

These amplifiers are designed for DMR and switch from transmit to receive every 30mS. They use RF sensing and PIN diode switching. For DATV use, where the output power needs to be kept under 10W for an acceptable spectrum, the sensitivity of the RF detection circuit may not be sufficient to switch to TX. It starts to work at around 50mW input.

I have reverse engineered the circuit, more of less, of the one I have.
Schematic.JPG
Schematic.JPG (292.72 KiB) Viewed 1693 times
On inspection, PTT can be enabled via Q1, a general purpose NPN transistor, through grounding it’s collector or alternatively providing positive bias to its base via a suitable resistor.

The amplifier is designed to be bi-directional. The isolation between input and output depends on the isolation provided by the PIN diodes, mainly D3 and D5. Poor isolation will result in feedback oscillations, especially if the output match is not great, which will probably be the case if a filter is added for frequencies above the filter passband. Much higher isolation is needed when the module is biased for linear operation as the gain increases. I found with my unit, the feedback via the RX path pin diodes was sufficient to cause oscillation.

This penchant for oscillation could be cured at the expense of breaking the receive path, by lifting one side of D3. As the through loss on receive is over 1 dB, a proper RF relay would be a better solution for TX/RX separation and a small relay can be driven from the 12V on TX supply, which is available from the drain of Q4. If you do this remember to include a diode across the relay coil to absorb the back EMF. Also remember to remove D3 to reduce the feedback.

Typically for DATV use there is no requirement for bidirectionality, rather the TX and RX ports often need to be separated for a Pluto or Lime SDR. In that case, removing D2 and L2 and feeding the transmitter at C3 would appear beneficial if there is no need for RF sensing. In that case it is probably best to also bypass the output diode switch and take the output from the module. This isn't necessary though, unless you use a LimeSDR where you definitively do not want any RF sensing!

Do not be tempted to modify for split RX TX using the existing pin diodes. It looks promising but the presence of sufficient feedback when the module is linearly biased to cause oscillation implies hat several 10s of mW must be leaking past the RX/TX switching. After modification this leakage would instead appear at the receiver port and test the patience of any receiver connected to it.

Note all of these modifications will prevent the amplifier being used for DMR.

Mike

g0mjw
Posts: 2330
Joined: Sat Sep 20, 2014 9:15 am

Re: DMR Amplifiers

Post by g0mjw » Tue Sep 28, 2021 6:39 pm

An update - should anyone be interested. I found when biased for DATV at a few A the bias level was very sensitive to supply voltage. I have fitted a 78L05 to stabilise it. Much better. With 5V bias the current was 5A so I dropped it a little with a series 22R resistor and a 1K to ground as I didn't have multiturn POT to hand. That set the bias to 4.9V and 3A. The module is extremely sensitive to the bias voltage. A POT from the supply isn't good enough to control it.
Capture.JPG
Capture.JPG (145.53 KiB) Viewed 1550 times
Mike

Post Reply

Return to “DATV - Digital ATV”