Hello,
I have an HDHR5-4DT (firmware 20260326) in the UK, working perfectly on Freeview. I'm also a radio amateur experimenting with narrowband DVB-T2 on the 70cm band, and I have three questions I'd rather ask than guess at.
1. Are the narrow DVB-T2 modulations implemented in the demodulator?
get /sys/features reports:
t8dvbt2 t7dvbt2 t6dvbt2 t5dvbt2 t4dvbt2 t3dvbt2 t2dvbt2 t1dvbt2 (and the same ladder for DVB-T)
Are the narrow entries (t1 to t5) actually implemented in the demod, or accepted by the command parser without corresponding demod support?
What I've observed: the tuner accepts arbitrary frequencies in Hz, which is very useful. set /tuner0/channel t1dvbt2:437000000 is accepted and reported back correctly. With a 1.7 MHz DVB-T2 signal present at 437 MHz, signal strength reads ss=90, so the front end clearly hears it, but snq stays at 0 and there's no lock, even after a minute. Identical behaviour at t2dvbt2 with a 2.0 MHz signal, and across frequency offsets of ±200 kHz.
For comparison, a Freeview DVB-T2 (Crystal Palace) mux at 546 MHz locks within about five seconds: ss=100 snq=100 seq=100 bps=40427520. So the unit is healthy, and the narrowband signal is known good, it's from a BATC Portsdown 4 transmitter and locks reliably on both a Sony CXD2880 receiver and a Knucker/Ryde receiver.
The complete absence of snq movement suggests the demod isn't attempting acquisition in these modes rather than trying and failing. Is that right?
2. Is there a newer model that does support them?
If this unit's demodulator doesn't do the narrow bandwidths, is there a current or forthcoming model whose demod does? I'd happily buy one if so. Equally, if the answer is that no HDHomeRun supports them, I'd rather know than experiment further.
3. ATSC 3.0 — how flexible is the tuning?
Separately, I'd like to begin ATSC 3.0 experiments on UK amateur allocations, where we have some freedom to try modes that aren't broadcast here. I'm interested in an ATSC 3.0 model as a reference receiver, but I'd need to know:
Does an ATSC 3.0 model accept arbitrary frequencies in Hz, as the DVB models do, or is it constrained to the US channel plan?
What is its tuning range? I'd be working in the 70cm band (430–440 MHz) and potentially 23cm vi a downconverter.
Is 6 MHz the only bandwidth, or are narrower options exposed?
Would the demodulator care about anything US-specific in the signalling, or would it lock to any conformant ATSC 3.0 transmission?
The appeal of ATSC 3.0 for us is LDM, a rugged layer for weak-signal and DX work alongside a high-capacity layer for normal use, in one carrier. That's an attractive proposition for amateur television, and a network-attached reference receiver would be ideal for developing against.
For context, I'm the author of an open-source DVB-T2 receiver project for amateur use and a trustee of a UK amateur television repeater, GB3OO so this is ongoing development work rather than idle curiosity. Happy to share findings back.
Thanks for any insight.
Justin, G8YTZ
Narrow DVB-T2 bandwidths (t1–t5) — demod support, and ATSC 3.0 tuning flexibility
Re: Narrow DVB-T2 bandwidths (t1–t5) — demod support, and ATSC 3.0 tuning flexibility
>Are the narrow entries (t1 to t5) actually implemented in the demod, or accepted by the command parser without corresponding demod support?
Yes, supported by the demod.
>2. Is there a newer model that does support them?
No, so far it is only supported on the HDHR5-2DT/4DT.
We are considering making a model specifically for this.
>3. ATSC 3.0 — how flexible is the tuning?
ATSC 3.0 models support arbitrary tuning by frequency in Hz.
Channel width is always 6MHz.
More than happy to support the amateur radio community.
Yes, supported by the demod.
>2. Is there a newer model that does support them?
No, so far it is only supported on the HDHR5-2DT/4DT.
We are considering making a model specifically for this.
>3. ATSC 3.0 — how flexible is the tuning?
ATSC 3.0 models support arbitrary tuning by frequency in Hz.
Channel width is always 6MHz.
More than happy to support the amateur radio community.
Re: Narrow DVB-T2 bandwidths (t1–t5) — demod support, and ATSC 3.0 tuning flexibility
Thank you and you were right.
Going back to the bench with your answer, the problem was mine. I'd assumed t1dvbt2 meant 1.7 MHz rounded down. It doesn't: it's a literal 1 MHz, and the ladder is exact. My transmissions were all at 1.7 MHz, which has no matching setting, so the demod was never being asked for the right thing while the front end sat there hearing the signal perfectly.
Retransmitted at 1 MHz and it locked immediately:
The 2m result is the one that surprised me most. Amateur television has been effectively a UHF and microwave activity because of the bandwidth involved, and a 1 MHz DVB-T2 signal changes that arithmetic — your tuner demodulates it at 146.5 MHz with perfect figures, well below anything it was built for. I'm not aware of any other receiver, amateur or commercial, that will do narrowband DVB-T2 on 2m and 70cm out of one box on the network.
The full 1–8 MHz ladder matters more than it might appear, too. Amateur repeaters worldwide use very different bandwidths: several German DB0 repeaters transmit wide DVB-T, and VK3RTV in Australia runs 7 MHz. So one receiver spanning 1 through 8 MHz covers narrowband experimental work and established wideband repeater outputs alike, which nothing in our world currently does. For the record, 1 MHz is entirely adequate for what I do; 1.7 would be a nice-to-have rather than a requirement.
I've added the QUATRO to Lynx as a network tuner alongside the usual amateur hardware, and it'll reach the beta channel shortly; the project is at https://github.com/G8YTZ/lynx-datv-receiver if you'd like to see how the device is being driven. Building that took an afternoon, which says a good deal for the control interface: arbitrary frequencies in Hz, explicit modulation, program selection, UDP target. Nothing needed working around.
Which brings me to three things I'd value your view on.
First, which model should I be recommending? This is about to reach a fair number of amateurs and I'd rather point them somewhere definite than have them guess. You've said the narrow bandwidths are on the HDHR5-2DT and 4DT — is the 4DT the one to recommend for Europe or Australia, and is it still current? And is there any route to the same narrowband capability for amateurs in other regions, or is the DVB model the only one today?
Second, on a model built for this. Thank you for raising the possibility. What we'd value is largely what the 4DT already does: the narrow DVB-T2 bandwidths, the full 1–8 MHz range, and tuning across at least 144–148 and 430–440 MHz. Above all, the network-attached multi-tuner architecture unchanged — a receiver that needn't be in the same room as the antenna, and several channels from one box, is transformative for a repeater site and is the single biggest advantage over everything else we use.
Third, ATSC 3.0. Thank you for confirming arbitrary Hz tuning at a fixed 6 MHz; that fits our 70cm and 23cm allocations comfortably. The standard interests us a great deal, particularly layered division multiplexing a rugged layer for weak-signal and DX working alongside a high-capacity layer in one carrier is something no mode we currently use offers. Before I buy one:
What is the tuning range of the ATSC 3.0 models? Does it reach down to 2m as the DVB tuners evidently do, and up towards 23cm?
Does the API expose the LDM layers separately, can a client select the core layer, the enhanced layer, or both?
What does the device output? ROUTE/DASH or MMT over IP rather than a transport stream, presumably; is the IP multicast delivered as received, or repackaged?
Are signalling and quality figures exposed the way snq and seq are for DVB, so a receiver can display lock quality?
Is there anything US-specific in the signalling a non-US transmission would need to satisfy, or will it lock to any conformant ATSC 3.0 signal?
Genuinely appreciated. Answering a detailed question from one amateur, within a day, and telling me plainly where I'd gone wrong, is not something every manufacturer would do.
Justin, G8YTZ
Going back to the bench with your answer, the problem was mine. I'd assumed t1dvbt2 meant 1.7 MHz rounded down. It doesn't: it's a literal 1 MHz, and the ladder is exact. My transmissions were all at 1.7 MHz, which has no matching setting, so the demod was never being asked for the right thing while the front end sat there hearing the signal perfectly.
Retransmitted at 1 MHz and it locked immediately:
- 437.000 MHz (70cm), 1 MHz DVB-T2: lock=t1dvbt2, ss=91, snq=100, seq=100, 770 kb/s
- 146.500 MHz (2m), 1 MHz DVB-T2: lock=t1dvbt2, snq=100, seq=100, level 100, 0.77 Mb/s
The 2m result is the one that surprised me most. Amateur television has been effectively a UHF and microwave activity because of the bandwidth involved, and a 1 MHz DVB-T2 signal changes that arithmetic — your tuner demodulates it at 146.5 MHz with perfect figures, well below anything it was built for. I'm not aware of any other receiver, amateur or commercial, that will do narrowband DVB-T2 on 2m and 70cm out of one box on the network.
The full 1–8 MHz ladder matters more than it might appear, too. Amateur repeaters worldwide use very different bandwidths: several German DB0 repeaters transmit wide DVB-T, and VK3RTV in Australia runs 7 MHz. So one receiver spanning 1 through 8 MHz covers narrowband experimental work and established wideband repeater outputs alike, which nothing in our world currently does. For the record, 1 MHz is entirely adequate for what I do; 1.7 would be a nice-to-have rather than a requirement.
I've added the QUATRO to Lynx as a network tuner alongside the usual amateur hardware, and it'll reach the beta channel shortly; the project is at https://github.com/G8YTZ/lynx-datv-receiver if you'd like to see how the device is being driven. Building that took an afternoon, which says a good deal for the control interface: arbitrary frequencies in Hz, explicit modulation, program selection, UDP target. Nothing needed working around.
Which brings me to three things I'd value your view on.
First, which model should I be recommending? This is about to reach a fair number of amateurs and I'd rather point them somewhere definite than have them guess. You've said the narrow bandwidths are on the HDHR5-2DT and 4DT — is the 4DT the one to recommend for Europe or Australia, and is it still current? And is there any route to the same narrowband capability for amateurs in other regions, or is the DVB model the only one today?
Second, on a model built for this. Thank you for raising the possibility. What we'd value is largely what the 4DT already does: the narrow DVB-T2 bandwidths, the full 1–8 MHz range, and tuning across at least 144–148 and 430–440 MHz. Above all, the network-attached multi-tuner architecture unchanged — a receiver that needn't be in the same room as the antenna, and several channels from one box, is transformative for a repeater site and is the single biggest advantage over everything else we use.
Third, ATSC 3.0. Thank you for confirming arbitrary Hz tuning at a fixed 6 MHz; that fits our 70cm and 23cm allocations comfortably. The standard interests us a great deal, particularly layered division multiplexing a rugged layer for weak-signal and DX working alongside a high-capacity layer in one carrier is something no mode we currently use offers. Before I buy one:
What is the tuning range of the ATSC 3.0 models? Does it reach down to 2m as the DVB tuners evidently do, and up towards 23cm?
Does the API expose the LDM layers separately, can a client select the core layer, the enhanced layer, or both?
What does the device output? ROUTE/DASH or MMT over IP rather than a transport stream, presumably; is the IP multicast delivered as received, or repackaged?
Are signalling and quality figures exposed the way snq and seq are for DVB, so a receiver can display lock quality?
Is there anything US-specific in the signalling a non-US transmission would need to satisfy, or will it lock to any conformant ATSC 3.0 signal?
Genuinely appreciated. Answering a detailed question from one amateur, within a day, and telling me plainly where I'd gone wrong, is not something every manufacturer would do.
Justin, G8YTZ
Re: Narrow DVB-T2 bandwidths (t1–t5) — demod support, and ATSC 3.0 tuning flexibility
You are right, I don't think there are any demods that support narrow bandwidths by default other than 1.7MHz T2. We reverse engineered how the driver was configuring the channel width on the demod then tested the performance with a new set of custom bandwidths.
One thing to note is that the tuner (RF to IF) can't filter down below 5 or 6MHz.
Currently only the HDHR5-2DT and HDHR5-4DT will work and both models have been discontinued. I am paying attention to what you are doing because I would love to make a new model for this using the same demod we used in the HDHR5 DT models.
The tuning range for the HDHR5 DT models = 44MHz to 866MHz.
The tuning range for the ATSC3 models = 44MHz to 1002MHz.
One thing to note is that the tuner (RF to IF) can't filter down below 5 or 6MHz.
Currently only the HDHR5-2DT and HDHR5-4DT will work and both models have been discontinued. I am paying attention to what you are doing because I would love to make a new model for this using the same demod we used in the HDHR5 DT models.
The tuning range for the HDHR5 DT models = 44MHz to 866MHz.
The tuning range for the ATSC3 models = 44MHz to 1002MHz.
Re: Narrow DVB-T2 bandwidths (t1–t5) — demod support, and ATSC 3.0 tuning flexibility
Thank you, that's genuinely useful, and it corrects two things I had wrong.
On the models being discontinued: that changes what I tell people. I've been writing this up for the amateur television community and pointing them at the 2DT and 4DT, which I'll now amend to say they're second-hand only. Is there stock still moving through distribution, or is it eBay from here?
On a new model: there is real interest, and I'd be glad to help however is useful; testing, documenting, or putting you in front of the people who'd buy one. For what it's worth on market size: the BATC has well over 1,400 members, there are active amateur television groups across Europe and in Australia, and repeater groups would want several units each rather than one. Not a large market, but an engaged one that talks to itself, and there is currently nothing else that does this.
What we'd want is essentially what the HDHR5 DT does today: the narrow bandwidths, the full 1–8 MHz ladder, if lower B/W is possible 333 kHz and 500 kHZ would open up the 50MHz and 70MHz bands and give it an even wider appeal. Arbitrary frequency tuning, and the network-attached multi-tuner architecture unchanged. That last part is the real differentiator a receiver that could be near the antenna, and several channels from one box, is transformative at a repeater site.
On the IF filtering: that's worth knowing and I hadn't considered it. So at 1 MHz there's demodulator selectivity but not front-end selectivity, and a strong signal within a few MHz of the wanted one still reaches the demod. At a repeater site with transmitters nearby that matters, and it argues for external filtering ahead of the tuner. I'll document it rather than let people discover it.
On tuning range: 44 MHz at the bottom is more interesting than I expected — that covers 6m and 4m as well as 2m and 70cm. Neither fits a 1 MHz channel in its amateur allocation, but it's useful to know the front end reaches there.
And it corrects my own assumption. 6 MHz ATSC 3.0 would fit our 23cm allocation with a down converter, which starts at 1240 MHz here with TV Repeater outputs in the 1300-1325 MHz bands. So ATSC 3.0 experiments would need to be on 23 cm
Thank you for the detail, and for taking an amateur enquiry seriously. If a new model does happen, I'd be glad to be an early test site and further develop the Lynx Receiver to support it.
https://github.com/G8YTZ/lynx-datv-receiver/tree/beta
Justin, G8YTZ
On the models being discontinued: that changes what I tell people. I've been writing this up for the amateur television community and pointing them at the 2DT and 4DT, which I'll now amend to say they're second-hand only. Is there stock still moving through distribution, or is it eBay from here?
On a new model: there is real interest, and I'd be glad to help however is useful; testing, documenting, or putting you in front of the people who'd buy one. For what it's worth on market size: the BATC has well over 1,400 members, there are active amateur television groups across Europe and in Australia, and repeater groups would want several units each rather than one. Not a large market, but an engaged one that talks to itself, and there is currently nothing else that does this.
What we'd want is essentially what the HDHR5 DT does today: the narrow bandwidths, the full 1–8 MHz ladder, if lower B/W is possible 333 kHz and 500 kHZ would open up the 50MHz and 70MHz bands and give it an even wider appeal. Arbitrary frequency tuning, and the network-attached multi-tuner architecture unchanged. That last part is the real differentiator a receiver that could be near the antenna, and several channels from one box, is transformative at a repeater site.
On the IF filtering: that's worth knowing and I hadn't considered it. So at 1 MHz there's demodulator selectivity but not front-end selectivity, and a strong signal within a few MHz of the wanted one still reaches the demod. At a repeater site with transmitters nearby that matters, and it argues for external filtering ahead of the tuner. I'll document it rather than let people discover it.
On tuning range: 44 MHz at the bottom is more interesting than I expected — that covers 6m and 4m as well as 2m and 70cm. Neither fits a 1 MHz channel in its amateur allocation, but it's useful to know the front end reaches there.
And it corrects my own assumption. 6 MHz ATSC 3.0 would fit our 23cm allocation with a down converter, which starts at 1240 MHz here with TV Repeater outputs in the 1300-1325 MHz bands. So ATSC 3.0 experiments would need to be on 23 cm
Thank you for the detail, and for taking an amateur enquiry seriously. If a new model does happen, I'd be glad to be an early test site and further develop the Lynx Receiver to support it.
https://github.com/G8YTZ/lynx-datv-receiver/tree/beta
Justin, G8YTZ