very high BER (~13%)

Help with DMR issues
Post Reply
KG7PAR
Posts: 28
Joined: Mon Aug 19, 2019 3:20 am

very high BER (~13%)

Post by KG7PAR » Mon Apr 27, 2020 1:41 am

Pi-Star:4.1.0 / Dashboard:20200421
MMDVM Hardware - Custom designed board (RPI GPIO Hat style).
---Derivative of this design http://www.repeater-builder.com/product ... DVM_v3.pdf, same firmware
Radios: External SM-50 RX/TX (Duplex)
HT: TYT-UV380

Only trying to do DMR at the moment, just want to get something to work. If this belongs in a different channel, my apologies, just looking for some guidance

In the firmware I have no "f/F" options, so I can't use the mmdvmcal with that option set to debug

Tried changing the RX offset from 1000 down to -1000,
Tried changing the RX level from 5% to 100%,
Tried adjusting the audio input (noise) level from 100mVpp to 3.3Vpp

nothing seems to move the need from the outputs you see below

What other knobs can I turn to get this under control?

Transmission lost, total frames: 26, bits: 3666, errors: 476, BER: 12.98418%
Transmission lost, total frames: 26, bits: 3666, errors: 477, BER: 13.01146%
Transmission lost, total frames: 26, bits: 3666, errors: 480, BER: 13.09329%
Transmission lost, total frames: 26, bits: 3666, errors: 475, BER: 12.95690%

Dan, kg7par

If useful, here are my current settings
I: 2020-04-27 01:40:17.315 This software is for use on amateur radio networks only,
I: 2020-04-27 01:40:17.315 it is to be used for educational purposes only. Its use on
I: 2020-04-27 01:40:17.315 commercial networks is strictly prohibited.
I: 2020-04-27 01:40:17.315 Copyright(C) 2015-2020 by Jonathan Naylor, G4KLX and others
M: 2020-04-27 01:40:17.315 MMDVMHost-20200326_Pi-Star-v4 is starting
M: 2020-04-27 01:40:17.315 Built 15:46:26 Mar 26 2020 (GitID #90c5b24)
I: 2020-04-27 01:40:17.315 General Parameters
I: 2020-04-27 01:40:17.315 Callsign: KG7PAR
I: 2020-04-27 01:40:17.315 Id: 3132763
I: 2020-04-27 01:40:17.315 Duplex: yes
I: 2020-04-27 01:40:17.315 Timeout: 240s
I: 2020-04-27 01:40:17.315 D-Star: disabled
I: 2020-04-27 01:40:17.315 DMR: enabled
I: 2020-04-27 01:40:17.315 YSF: disabled
I: 2020-04-27 01:40:17.315 P25: disabled
I: 2020-04-27 01:40:17.315 NXDN: disabled
I: 2020-04-27 01:40:17.315 POCSAG: disabled
I: 2020-04-27 01:40:17.315 Modem Parameters
I: 2020-04-27 01:40:17.315 Port: /dev/ttySC0
I: 2020-04-27 01:40:17.315 Protocol: uart
I: 2020-04-27 01:40:17.315 RX Invert: no
I: 2020-04-27 01:40:17.315 TX Invert: yes
I: 2020-04-27 01:40:17.315 PTT Invert: no
I: 2020-04-27 01:40:17.315 TX Delay: 100ms
I: 2020-04-27 01:40:17.315 RX Offset: -1000Hz
I: 2020-04-27 01:40:17.315 TX Offset: -450Hz
I: 2020-04-27 01:40:17.315 RX DC Offset: 0
I: 2020-04-27 01:40:17.315 TX DC Offset: 0
I: 2020-04-27 01:40:17.315 RF Level: 50.0%
I: 2020-04-27 01:40:17.315 DMR Delay: 0 (0.0ms)
I: 2020-04-27 01:40:17.315 RX Level: 50.0%
I: 2020-04-27 01:40:17.315 CW Id TX Level: 50.0%
I: 2020-04-27 01:40:17.315 D-Star TX Level: 50.0%
I: 2020-04-27 01:40:17.315 DMR TX Level: 50.0%
I: 2020-04-27 01:40:17.315 YSF TX Level: 50.0%
I: 2020-04-27 01:40:17.315 P25 TX Level: 50.0%
I: 2020-04-27 01:40:17.315 NXDN TX Level: 50.0%
I: 2020-04-27 01:40:17.315 POCSAG TX Level: 50.0%
I: 2020-04-27 01:40:17.315 RX Frequency: 146640000Hz (146639000Hz)
I: 2020-04-27 01:40:17.315 TX Frequency: 146040000Hz (146039550Hz)
M: 2020-04-27 01:40:17.315 Opening the MMDVM
I: 2020-04-27 01:40:19.340 MMDVM protocol version: 1, description: MMDVM 20190130 (D-Star/DMR/System Fusion/P25/NXDN/POCSAG) 12.0000 MHz GitID #ff7a9fd
I: 2020-04-27 01:40:19.360 Display Parameters
I: 2020-04-27 01:40:19.361 Type: None
W: 2020-04-27 01:40:19.361 No valid display found, disabling
I: 2020-04-27 01:40:19.361 DMR Network Parameters
I: 2020-04-27 01:40:19.361 Address: 74.91.114.19
I: 2020-04-27 01:40:19.361 Port: 62031
I: 2020-04-27 01:40:19.361 Local: random
I: 2020-04-27 01:40:19.361 Jitter: 360ms
I: 2020-04-27 01:40:19.361 Slot 1: enabled
I: 2020-04-27 01:40:19.361 Slot 2: enabled
I: 2020-04-27 01:40:19.361 Mode Hang: 20s
I: 2020-04-27 01:40:19.361 Info Parameters
I: 2020-04-27 01:40:19.361 Callsign: KG7PAR
I: 2020-04-27 01:40:19.361 RX Frequency: 146640000Hz
I: 2020-04-27 01:40:19.361 TX Frequency: 146040000Hz
I: 2020-04-27 01:40:19.361 Power: 1W
I: 2020-04-27 01:40:19.361 Latitude: 47.979851deg N
I: 2020-04-27 01:40:19.361 Longitude: -122.202003deg E
I: 2020-04-27 01:40:19.361 Height: 0m
I: 2020-04-27 01:40:19.361 Location: "CN87vv"
I: 2020-04-27 01:40:19.361 Description: "United States"
I: 2020-04-27 01:40:19.361 URL: "http://qrz.com/db/kg7par"
M: 2020-04-27 01:40:19.361 DMR, Opening DMR Network
I: 2020-04-27 01:40:19.361 RSSI
I: 2020-04-27 01:40:19.361 Mapping File: /usr/local/etc/RSSI.dat
I: 2020-04-27 01:40:19.363 Loaded 14 RSSI data mapping points from /usr/local/etc/RSSI.dat
I: 2020-04-27 01:40:19.363 DMR Id Lookups
I: 2020-04-27 01:40:19.363 File: /usr/local/etc/DMRIds.dat
I: 2020-04-27 01:40:19.363 Reload: 24 hours
I: 2020-04-27 01:40:21.135 Loaded 160281 IDs to lookup table - /usr/local/etc/DMRIds.dat
I: 2020-04-27 01:40:21.135 Started the DMR Id lookup reload thread
I: 2020-04-27 01:40:21.135 DMR RF Parameters
I: 2020-04-27 01:40:21.135 Id: 3132763
I: 2020-04-27 01:40:21.135 Color Code: 1
I: 2020-04-27 01:40:21.135 Self Only: no
I: 2020-04-27 01:40:21.135 Embedded LC Only: no
I: 2020-04-27 01:40:21.135 Dump Talker Alias Data: yes
I: 2020-04-27 01:40:21.135 Prefixes: 0
I: 2020-04-27 01:40:21.135 Call Hang: 3s
I: 2020-04-27 01:40:21.135 TX Hang: 4s
I: 2020-04-27 01:40:21.135 Mode Hang: 20s
I: 2020-04-27 01:40:21.135 OVCM: off
I: 2020-04-27 01:40:21.135 DMR Roaming Beacons Type: off
M: 2020-04-27 01:40:21.135 MMDVMHost-20200326_Pi-Star-v4 is running
M: 2020-04-27 01:40:31.329 DMR, Logged into the master successfully

DG0OFZ
Posts: 13
Joined: Thu Jun 07, 2018 6:53 pm

Re: very high BER (~13%)

Post by DG0OFZ » Mon Apr 27, 2020 3:50 pm

Hi,
KG7PAR wrote:
Mon Apr 27, 2020 1:41 am
Tried changing the RX offset from 1000 down to -1000,
Tried changing the RX level from 5% to 100%,
With the design of the repeaterbuilder boards you can't change RX or TX frequency by the modem.
The modem is pure audio processing.

Please check your external radio for correct Rx and TX frequencies!
Also check the RX bandwidth of the external radio.
73s de Thomas, DG0OFZ

KG7PAR
Posts: 28
Joined: Mon Aug 19, 2019 3:20 am

Re: very high BER (~13%)

Post by KG7PAR » Mon Apr 27, 2020 4:06 pm

Thanks good to know on the frequency adjustments, this is all new to me

The HT TX is at 146.0391MHz (had to use in analog mode to keep it active long enough to get a reading), the RX is programmed for 146.040 MHz/156.7. I am not able to adjust the settings in the RX radio at the moment, waiting for the programming box to get delivered later this week. I have picked off the raw audio inside the radio so I don't think the PL will be an issue for my purposes.

Working with a local club member, I have added an HT whip antenna to the RX rig and I have also tried to adjust the RX invert with the following results

RX Invert: On
DMR audio seq. 3, FEC BER % (errs): 9.929% (14/141)
DMR audio seq. 3, FEC BER % (errs): 9.929% (14/141)
Transmission lost, total frames: 26, bits: 3666, errors: 472, BER: 12.87507%
Transmission lost, total frames: 26, bits: 3666, errors: 478, BER: 13.03873%
Transmission lost, total frames: 26, bits: 3666, errors: 485, BER: 13.22968%
RX Invert: Off
Transmission lost, total frames: 4, bits: 564, errors: 77, BER: 13.65248%
Transmission lost, total frames: 4, bits: 564, errors: 77, BER: 13.65248%
Transmission lost, total frames: 4, bits: 564, errors: 77, BER: 13.65248%

DG0OFZ
Posts: 13
Joined: Thu Jun 07, 2018 6:53 pm

Re: very high BER (~13%)

Post by DG0OFZ » Mon Apr 27, 2020 4:37 pm

Hi,
KG7PAR wrote:
Mon Apr 27, 2020 4:06 pm
The HT TX is at 146.0391MHz (had to use in analog mode to keep it active long enough to get a reading), the RX is programmed for 146.040 MHz/156.7.
So you are 900Hz off. I'd think that's to much.

Please make sure, that your TX and RX are on the correct frequencies. This means measure the frequencies!

With (pistar-)mmdvmcal you can set a constant carrier without modulation. So you are able to connect the TX to a counter (please: not directly!) and see if it is TXing on the right qrg.
For RX you can't measure it so easy. But you may check for lowest BER while your repeater is in RX mode an you are TXing with another device.
73s de Thomas, DG0OFZ

KG7PAR
Posts: 28
Joined: Mon Aug 19, 2019 3:20 am

Re: very high BER (~13%)

Post by KG7PAR » Mon Apr 27, 2020 6:01 pm

For the first reading on the HT, I used a cheap frequency counter.

I have dug out my Agilent 53131A (225MHz capable) and used an antenna for the input. When I key the HT in analog mode, I get a reading of 146.040.0xx.

Using a similar technique, I am measuring the TX frequency on the receiver (assuming the same oscillator effects will be in play at RF frequencies for Rx and Tx) and I am getting 146.640.1xx.

AF6VN
Posts: 466
Joined: Fri Jul 20, 2018 1:15 am

Re: very high BER (~13%)

Post by AF6VN » Mon Apr 27, 2020 10:14 pm

KG7PAR wrote:
Mon Apr 27, 2020 4:06 pm
The HT TX is at 146.0391MHz (had to use in analog mode to keep it active long enough to get a reading), the RX is programmed for 146.040 MHz/156.7. I am not able to adjust the settings in the RX radio at the moment, waiting for the programming box to get delivered later this week. I have picked off the raw audio inside the radio so I don't think the PL will be an issue for my purposes.
For a dual radio (repeater) configuration, those frequencies look rather close together -- such that the Tx could be feeding back on the Rx radio.
KG7PAR wrote:
Mon Apr 27, 2020 6:01 pm
I have dug out my Agilent 53131A (225MHz capable) and used an antenna for the input. When I key the HT in analog mode, I get a reading of 146.040.0xx.

Using a similar technique, I am measuring the TX frequency on the receiver (assuming the same oscillator effects will be in play at RF frequencies for Rx and Tx) and I am getting 146.640.1xx.
... while that appears to reflect common 2m repeater offset /between/ Tx and Rx phases. I'd think one of the radios should be listening on 146.6xx while the other is transmitting on 146.0xx -- as the system user would be sending on 146.6xx and listening on 146.0xx (or the opposite configuration depending on + or - repeater shift setting).

--
AF6VN
Dennis L Bieber

KG7PAR
Posts: 28
Joined: Mon Aug 19, 2019 3:20 am

Re: very high BER (~13%)

Post by KG7PAR » Mon Apr 27, 2020 10:19 pm

I think there may be some confusion. There are 3 radios in the mix,

1) the HT (TYT UV-380) which I have measured the TX frequency of, which matches the RX (SM-50).
2) The Rx (SM-50) which I have measured the TX frequency of to check the accuracy of the oscillator.
3) The Tx (SM-50) which so far has been completely powered off.

The Rx and Tx do obey standard 0.6MHz offsets when they are both in use as a repeater setup (ultimate goal)

Post Reply