September Trike and Welder Update

The BITX40 is dead. Long Live the BITX40!

DSB Digital: Friend, or Foe?

I recently became aware that there are companies selling Double Side Band radios for use on digital HF frequencies (FT8 specifically) on 40m, 30m, and 20m. In this post we’re going to take a look at what these rigs are, what they were originally designed for, and then you can decide whether these should be used for digital communications on HF.

Disclaimer: I don’t own a digital DSB rig. This post is intended to discuss the realities of such a design so that somebody looking to purchase a starter radio is completely informed of what they are getting into.

What is DSB?

Double Side Band describes an AM radio transmission which has had its carrier suppressed (aka filtered out). That means that only the sidebands are amplified, which in turn means that the transmission is 1/3 more efficient, because 1/3 of the useless signal isn’t transmitted. SSB (Single Side Band) goes a step further and filters out the redundant half of the transmission yet again, and so 2/3 of the original waveform is cut out, leaving a Single Side Band. This is what we’re accustomed to when communicating in most HF modes. Double Side Band foregoes the added filtering and still puts out everything but the carrier.

Why DSB digital?

Among the vendors who are selling kits for DSB Digital transceivers, at least one of them references Peter Parker’s (VK3YE, not Spiderman) Beach40 design as a design inspiration, But let’s look at what Peter actually says that the Beach40 was for. This is taken from his website and is for reference only. I hope you don’t mind, Peter!

It’s a project for the tinkerer… if you’re not quite ready to assemble an SSB transceiver, give this simpler DSB rig a shot.

Peter Parker VK3YE on the Beach40 transceiver project

You’ll notice that it’s not intended for anything more than a project for a homebrewer– somebody who intends to build radios from scratch- to get on the air with without having the complexity of a SSB transceiver, and ostensibly to eventually graduate from. So, why are vendors putting out DSB digital transceivers that are solely intended for digital communications?

These digital DSB rigs are apparently intended to lower the cost and complexity of a transceiver by omitting the sideband filter and the alignment that goes along with it, and just transmitting both sidebands regardless of whether this is needed.

Is DSB good for Digital modes?

I’m not going to come out and tell you whether you should think DSB digital is bad or not. Instead lets look at some facts about DSB, and then you can determine this for yourself.

Extra Data Transmission

A SSB transmission is 3khz wide, but a DSB transmission is twice that at 6khz. When an audio signal is transmitted at say 1500hz, its opposite is transmitted at -1500hz. Thus you have two signals being sent but only one is needed.

Take for example 7074khz and 7078khz. These are the FT8 and JS8Call frequencies commonly in use. If you’re transmitting a signal on 7078khz, at 1500hz, then you’re outputting TWO signals: One at 7079.5khz, and one at 7076.5khz. Only the Upper Side Band can be decoded, because the Lower Side Band at 7076.5khz is technically “reversed” and cannot be decoded. So, if you were using one of these rigs for JS8Call, the FT8 crowd might not appreciate it very much. But there’s a catch.

You can’t use these rigs for JS8Call, because they are crystal controlled. They are not frequency agile. They can only be used for FT8. This is of course unless you go to the trouble of finding other crystals and figure out a way to switch them out.

How does power output look? They tout about 3W of DSB output on 40m, but that’s DSB output. That’s 1.5W per side band. And since only one sideband is useful, effective wattage is half that listed.

There is one other thing to consider: The receiver lacks filtering as well. That means that you’re effectively doubling the receiver bandwidth, which reduces sensitivity by at least 3db.

Counting the Cost

Lastly, there is cost. These digital DSB rigs are selling for about $40. That’s definitely cheap, but is it a value? For $59 and a little bit of hackery you can have a BITX40 that is frequency agile and can be used on any digital mode or with voice, is Single Side Band and puts out 5-7W.

Now that the BITX40 is gone, the perceived value does go up a bit. It is an inexpensive entry into ham radio, and as an owner of QRP Guys pointed out below, it’s good for experimenting and learning with.

What do you think? To digital DSB rigs belong on the air? Let us know your thoughts in the comments below.

13 comments

Skip to comment form

    • Terry Bendell on October 22, 2019 at 9:37 PM
    • Reply

    I’m building this one Ryan:

    https://qrpguys.com/digital-transceiver

    1. Hey Terry, please report back your results. I’d love to hear your impression of it!

    • David Goodrow on October 23, 2019 at 7:57 AM
    • Reply

    Actually you can use it for JS8CALL, if the frequency is not currently in use. No one owns a spot on the dial. Most hams can find and change out a crystal, especial one they put on the board themselves. I have watched the BITX groups for a while and the BITX40 kit has a lot of issues and a lot of fixes. It’s a hobby people, let’s stop getting excited and thinking ever new thing that comes alone is going to be the end of the hobby. I am sure the cw people cursed the am crowd when they showed up.

    I have purchased two of these boards. I planning on testing them side by each into dummy loads at first. Then it looks like probably 7.110 for JS8CALL.

    1. You’re right, it’s possible to use JS8Call, but not on the default frequencies. Also the BITX40 has far fewer issues than the uBITX and is cheaper too. It’s a great radio. I’d love to hear how your experiments turn out.

    • Artemus on October 24, 2019 at 8:48 AM
    • Reply

    Probably illegal in the USA in the CW and narrow bandwidth data bands. It would have to be used in the voice bands. It’s putting out a 6khz wide AM suppressed Carrier. AM is not legal in the Narrow data bands.

    1. The general consensus is that the legalities aren’t an issue, but this post is rather focused on the technical aspects: Low power output, poor sensitivity, etc.

    • Doug Hendricks on November 4, 2019 at 5:02 PM
    • Reply

    We put out the kit as a low cost way to try ft8. It’s not intended to be a full featured rig that you will use all the time, but as a platform for experiments and tinkering, just as Peter said. It is entirely legal to operate dsb digital in the cw band, just like it is legal to operate ssb digital there. There will be far less interference from a 1 Watt DSB signal running FT8, than there will be from a 100 Watt SSB station running FT8. If you want to experiment and try the mode of FT8, and have a choice of 3 bands, the Qrpguys kit is a bargain.
    Disclaimer. I am a co-owner of Qrpguys.
    Doug KI6DS

    1. Hi Doug,

      I applaud you for commenting here. I want to be clear: There’s two sides to this story, and both sides have merit. It’s obviously caused a lot of controversy. As a learning tool, I agree that it has quite a bit of value. You also make a great point about 100W FT8 signals. They splatter so badly!

  1. Hi Doug, I have seen the kit in person as you showed it to me! The production quality is excellent, regardless of the debate over the mode. My comments on this now, as previously, are not intended to refer to yours or any particular kit, they are purely technical in nature relating to DSB and digital modes DSB generally.

    Regarding your comment “There will be far less interference from a 1 Watt DSB signal running FT8, than there will be from a 100 Watt SSB station running FT8.”… I think this is wrong. 1W is 20dB down on 100W. I have never heard of an SSB transmitter whose unwanted sideband is at only -20dBc. Typical would be -60dBc or better. In that case, 1,000,000W of SSB would be required to produce 1W in the unwanted sideband. “Splatter” also cannot be an issue since the usual IMD3 performance of linear power amplifiers is around -30dB; even this requires the presence of two tones, but FT8 is only a single tone so any transmitter non-linearity has nothing else to intermodulate against.

    73 Hans G0UPL

    • Doug Hendricks on November 5, 2019 at 10:28 AM
    • Reply

    Hans, I am referring to the amount of bandwidth taken up by the 100 Watt signal. There is a ham close to me who totally overloads my receiver when he transmits at 100 watts. He never hears me, but he certainly affects my reception. I don’t see the need for anyone ever to use 100 Watts for FT8, yet I know guys who use a Kilowatt. My point is that I just don’t see DSB causing a problem at 1 Watt. We did the transceiver as an experimental project, for hams to be able to try FT8 without a huge expenditure. We chose DSB because it was cheap, easy and LEGAL to do. Most of the complainers don’t seem to realize that it is 100% legal to transmit DSB FT8 according to FCC rules and regulations in effect today. That may change, and if it does, we would abide by those changes. It is a fun project, it works, and you will make contacts with it.

    1. Yeah I think the main point is that the folks who plug in a rigxpert to their radio and just start operating are not doing any favors with their highly overdriven audio splattering all over the place. It’s possible to run 100w properly, but most don’t.

  2. Hi Doug,

    I have no doubt about legality and am not commenting on your transceiver in particular. Just making technical discussion on DSB and other misunderstandings.

    Watts don’t take up bandwidth. The mode being transmitted is what takes up bandwidth. An unmodulated carrier has zero bandwidth regardless of the power being transmitted. In this regard FT8 (done properly on SSB) is fairly innocent as it occupies only 50Hz which is relatively narrow. I’d estimate around 90% of QSO partners I encounter on CW are using around 100W and CW is also quite narrow.

    Improperly done SSB, if the volume sent into the SSB transceiver’s input is too high, will cause clipping and the audio harmonics will tend to inter modulate with each other. This is an improperly setup transmitter – and is nothing to do with the power level nor the mode.

    Properly adjusted equipment having suitable performance should be able to operate in reasonably close proximity to each other, such as on the same site at field day.

    I have no doubt your 100W nearby ham is much more likely to cause you trouble, than your 1W causes him. Assuming you are both using FT8 on the same band, you are both within the same SSB channel so splatter has nothing to do with this. Two possible causes exist for the interference he causes you: his transmitter and your receiver. Your receiver is the place to start, both because you have most easy access and control over it, and because it is the most likely cause of the problems. Your close-in within band dynamic range is the crucial performance characteristic here. If the receiver you are using is your SA602 then you can totally expect to have problems: the SA602 has well known limitations on dynamic range and IP3, and your filtering ahead of it is also very limited. Poor phase noise performance of oscillators is also a possible culprit, in this regard in sure since your SA602 transceiver is crystal-controlled it probably has good phase noise performance. The nearby ham’s transmitter phase noise can also be a problem if it isn’t at a good low level.

    Even if everything is excellent… I suspect operating near a 100W transmitter spaced only some hundreds of Hz away from you is always going to be challenging. The same would apply to CW or any other mode. But at least in CW, SSB etc you could move many kHz away from each other. Then this is a limitation of FT8 generally perhaps. The fact everyone is close to each other, makes it difficult for nearby stations to coexist on the same band. Every mode has advantages and disadvantages and I guess this is one if the DISadvantages of FT8.

    Vy 73 de Hans G0UPL

    • Jeff on November 6, 2019 at 9:36 AM
    • Reply

    You wrote: “That means that only the sidebands are amplified, which in turn means that the transmission is 1/3 more efficient, because 1/3 of the useless signal isn’t transmitted.”

    It is actually better than that. In regular AM with a modulation index of 1 (100% modulation), the carrier has 66.67% of the power, and the sidebands have 16.67% each. With lower modulation indices the power in the carrier goes up at the expense of the sidebands.

    Jeff aa6xa

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.