- Amateur Radio (Ham Radio) Community

Call Search

New to Ham Radio?
My Profile

Friends Remembered
Survey Question

DX Cluster Spots

Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement

[Articles Home]  [Add Article]  

The Good Bad and Ugly on PSK31

from N6CRR Steve on December 12, 2005
View comments about this article!

The Good, Bad and Ugly on PSK31

I've been working PSK31 for a little over a year now and have encountered signals on the bands which range from really good, to bad to really ugly. In this article I'd like to share a few examples and explain what I think is going on to make the signal the way it is. No calls are associated with the signals, as doing so would not be in keeping the helping each other out sprit of Amateur Radio.

The signals in this paper were all captured on 20 meters one Sunday morning around 14.070 with propagation being only moderate. There was a CW contest going on, so my receiver was getting pumped a bit by contesters operating nearby.

The equipment used to capture the signals in this paper from 20 meters:

Kenwood TS 480S

Hustler BT5V, vertical ground mounted

MixW2.17b running on an AMD 3.2 GHz computer under Windows XP

BuxCom Interface connected to TS480 data in/out

Screen captures


0x01 graphic

Zoom = 4X 10 Hz/each mark

This is an example of a pretty good signal which is not supper strong, but is nicely modulated. Notice that the signal is about 10 Hz wide, and has a nice railroad track look when there is an idle or no character sent space. There's a tad bit of extra modulation next to the main signals, but overall a pretty nice signal from a weaker signal.

0x01 graphic

Zoom = 4X 10 Hz/each mark

Another not so strong signal, but one that is pretty clean. You can't see the “rail road” tracks as the band and the signal is not helping out.

0x01 graphic

Zoom = 4X 10 Hz/each mark

Another not so strong signal, but one that is pretty clean. You can see the “rail road” tracks pretty clearly here, they are more noticeable when the station is idling or not sending characters. Bad

0x01 graphic

Zoom = 4X 10 Hz/each mark

This is an example of what I call a bad signal, it's not terrible, but it is overdriven. If you look at the display, this signal is chewing up about 150 Hz of spectrum. The modulation is not a clearly defined railroad track, it more like two strong lines with lots of echo. This station is also wasting some of his modulation by spreading out the RF energy across more spectrum than needed, so the desired BPSK modulation is less efficient.

0x01 graphic

Zoom = 4X 10 Hz/each mark

The snap did not catch the worse of the signal, which at times was quite a bit more showing in the sidebands. This signal was about 100 Hz wide including the sidebands. You can see some ringing in this signal at idle, where there is only the idle, not data which looks like just simple rail road tracks.

0x01 graphic

Zoom = 4X 10 Hz/each mark

This signal even sounded “funny”, with a raspy sort of background sound.


0x01 graphic

Zoom =1.5 X 100Hz/each mark

This is an example of about as bad as it gets. This signal was so bad that I could not decode it! I'm not sure why this signal was so bad, but I suspect the station was not using an interface between radio and computer, or the speech processor was on and clipping or something was very out of whack on this station. This signal was making something like 700Hz of spectrum unusable to others!

0x01 graphic

Zoom =1.5 X 100Hz/each mark

0x01 graphic

Zoom = 4X 10 Hz/each mark

This is about as bad as it gets. Funny thing to the station with this signal was having a hard time copying others, go figure. This signal is about 500 Hz wide, compared to a properly modulated signal of 10 Hz.

The two pictures are of the same station by the way.

What's Going on

BPSK31 uses a base data signal which is more or less a square wave. Square waves and amplifier circuits do not always mix well because the sharp transition used in digital modes where data changes from one state to another contains lots of frequency components; the number of frequency components is related to how fast the transition occurs and how big the difference in signal level between one state and the other is. Amplifiers try to keep the signal level at some set gain value which the amplifier thinks is right, on a too strong a data signal it can overshoot, undershoot, overshoot… another way to describe ringing. Another way to think of a digital signal is that at some rate it is you are sending no data, then some data in an endless stream.

So what do amplifiers have to do with putting out a clean or a bad BPSK31 signal?

First off consider how you are feeding BPSK31 signals into the radio. If you are feeding data from your interface into the radio via the microphone jack, you are feeding an amplifier circuit that is designed for a wide dynamic range with a fair amount of gain to take voice signals up to some average state. The state of where the gain of the microphone amplifier is set for normal voice operations is measured using the ALC setting of your meter on the radio, with adjustments of the microphone gain so you have some peak ALC reading.

In the case of driving your radio from your computer sound card, you are setting the gain of the microphone amplifier for a constant level which is based on the digital data stream coming in as a stream of 0's and 1's. If the gain of the microphone setting is such that the amplifier is trying to hold a constant level shown on your ALC, you may be getting ringing because the amplifier circuit is not working well with the digital signal which causes sidebands.

Tips on adjusting for a clean signal

  1. Always turn the signal processor off when running BPSK or any other digital modes. Processors change the gain profile of the amplifier, and doing very strange things to the pass band which is very bad for digital modes.

  2. Adjust your computer sound card output while adjusting microphone gain for very low ALC readings.

TRICK: Set your radio mic gain to some small value say 1/8 to 1/16 of full scale, then adjust your computer sound card “Master Volume” and “Wave Out” to while watching ALC. Think of the Master Volume as the “main” knob and the wave output as the “RIT”. There is some tip point where you are getting all the RF out of the radio you are going to get, and more ALC does not make a difference so try not to use the ALC reading to set the RF power out. On my TS480S, I usually show only one “Dot” of ALC and get good reports.

  1. Adjust for an ALC reading which as close to 0 or just showing a slight movement.

  2. Drive your radio via data input ports.

TRICK: Many modern radios have data input ports which bypass some of the microphone amplifiers and are less prone to being overdriven.

  1. Ask for, and give real signal reports. It's not a bad thing to tell another operator that there is something wrong with their signal, they might not know and would appreciate your help!

  2. Lower power works very well on PSK31. This is because you are putting all the RF energy you are producing into a very narrow, 10 Hz, spectrum when you are properly adjusted on modulation.

The good news

When I started writing this paper I thought that it would be very easy to find examples of bad and ugly signals, well I was wrong! For the most part signals now days are pretty good, with few bad signals and ugly signals are now very hard to find.

Member Comments:
This article has expired. No more comments may be added.
The Good Bad and Ugly on PSK31  
by K3AN on December 12, 2005 Mail this to a friend!
When you see those broad signals, it may not be the other station, it may be your computer's sound card!

I noticed that whenever a PSK signal was strong enough to reduce or eliminate the noise trace in the waterfall, it was always broader than weaker signals. I could often discern 4 railroad track lines where there should have been 2 during idle periods. Then one day I tried to tell one of these stations that his signal was too broad, and we ended up doing some tests. He confirmed that his processor was off, and that his output power was so low that there was no ALC voltage being generated in his transmitter. Puzzled, I then I turned my receiver's AF gain to a lower than usual setting, and lo and behold, his signal "cleared up."

I noticed in the author's images that the "bad as it gets" signal is strong enough that there is no noise trace in the waterfall. I think the sound card is being overdriven.

The Good Bad and Ugly on PSK31  
by N6CRR on December 12, 2005 Mail this to a friend!
Maybe you are right about sound card causing some of this, but the one I use has a pretty good dynamic range and the ALC in the radio tends to keep things pretty level. I have seen very strong signals that are very clean, and very strong signals that are very poor, go figure.

For the most part, the extra lines in the Bad and Ugly shots are coming from the transmitting station being over-driven. By the way, as I stated in the article, there are very few Bad and Ugly signals lately, although I expect quite a few more OM's on the air on this mode after Santa puts a few new interfaces under the tree who might need a hand getting things squared away. It's OK to offer to help another station adjust his station for a good signal and they might just appreciate the help!

The Good Bad and Ugly on PSK31  
by NA4IT on December 12, 2005 Mail this to a friend!
Very good article Steve. While PSK31 is an easy mode to use, there is a little bit of a learning curve with it. I believe good articles with pictures (aka screenshots) will help everyone.

I also agree we should all spend time to help another newcomer to PSK. That's about the only way to make sure you have it set right.
RE: The Good Bad and Ugly on PSK31  
by WX5NCO on December 12, 2005 Mail this to a friend!
Thank you for posting this article. I ordered the interface cabling last week for my rig. Sunday afternoon, I got on the various bands and called CQ. I did get a few responses back on 20. I don�t know if I was doing something wrong or not, but after they answered my CQ, we did one exchange each (qth, rig, etc) the other station I guess "disappeared". Like I said, I don�t know if I did something wrong or if the band was crashing. Either way I was going to read up on PSK today and join some PSK yahoogroups, if there are any.
I knew enough to know to keep the power down and to have the ALC meter read 1 or 2 bars at the most. That being said, I hope I wasn�t one of those who had a �bad� signal.
Hope to talk�errr� type to y�all soon on the bands
RE: The Good Bad and Ugly on PSK31  
by AA4PB on December 12, 2005 Mail this to a friend!
It's the AGC in the receiver that can cause distortion of strong signals just like ALC in the transmitter can. A fast AGC action will "follow" the amplitude of the PSK31 signal and attempt to level it out - that is distortion. A slow AGC action cannot react fast enough to "follow" the signal and so will just maintain an average without distorting the signal. Turning down the RF gain will reduce the signal below the level that causes any (or much) AGC action. Always use your slowest AGC setting.

PSK31 has amplitude changes. Any non-linearity anywhere in the system from the transmitter input to the receiver output will cause distortion which results in poor IMD readings and wide signals. This can be the result of ALC or AGC action or overdriving an amplifier anywhere in the system.

I've done a considerable amount of testing to look at this issue. On an IC761 with the AGC set to fast I could have as much as 10dB of IMD caused by the AGC with an S-9 signal strength. I measured the output of my test transmitter an knew that I was transmitting a clean test signal. Switching the AGC to slow, reducing the RF gain to weaken the signal, or reducing the signal strength to S2 or S3 would clean up the signal.

For transmit ALC testing I used a K2 because I could easily change component values inside the transmitter. I found that with some ALC action I could actually see a copy of the PSK31 signal on the ALC line and of course the IMD reading at the receiver was poor. By changing the ALC decay time (adding a capacitor) the waveform on the ALC line went away and the IMD improved considerably. I found that with the proper selection of decay time the ALC could be used to level the average transmitter output without distorting the PSK31 signal.

The time constants for most of our receiver AGC and transmitter ALC were designed for typical voice signal rather than PSK31.
RE: The Good Bad and Ugly on PSK31  
by AA4PB on December 12, 2005 Mail this to a friend!
I should add that unless you have done some testing to determine the time constant of your ALC circuit you should keep the audio input levels down to where there is absolutly zero ALC reading if you want to have a clean signal. I back my gain down until there is no ALC reading and then back it down a little more. This usually results in an output of 20-30 watts from a 100W transmitter.
RE: The Good Bad and Ugly on PSK31  
by N6CRR on December 12, 2005 Mail this to a friend!

You are quite correct about any amplifier stage, either in the transmit chain or the receive chain can contribute to a perception of "ringing". Something I had not gone into either in my testing or in the paper.

My test technique was not all that sophisticated and after reading your comments I might have found a flaw in the way I have my station set up which could have had an effect on the measurements. One other note, the measurements and screen shots were taken over the course of one Sunday afternoon with a pretty cruddy band but with most signals in the S4 to S7 ranges but not recorded and signal strength was not correlated to the screen shots.

The potential flaw you point out is related to how I have the gain schedule set on my receive chain. On my receiver (TS480S) controls, I can set the output level of the composite audio signal which is fed to the sound card. I have that set fairly high, and use the record setting on the sound card to get the level right into the computer. This approach assumes that the linearity, and as you point out the AGC time constants are better on the radio side as the majority of gain is on the receive chain is in the radio side v the computer sound card. If I set the radio audio output lower, I am using less receiver audio stage gain and more sound card gain. It's not clear to me which has a better dynamic range and linearity over range, radio or sound card.

I may go back and adjust the gain schedule of my station such that the receiver is providing less audio gain by turning down the audio output drive from the receiver and using more sound card gain and re-run some tests. A very good input from you!

RE: The Good Bad and Ugly on PSK31  
by AA4PB on December 12, 2005 Mail this to a friend!
Steve, I don't think the audio levels will have a major impact unless of course you are actually overdriving some audio stage. What I do is to set my computer controls to about half way (between 1/3 and 2/3 of maximum) and then adjust the receiver output level for the correct drive as indicated on the software program.
The Good Bad and Ugly on PSK31  
by N5UV on December 12, 2005 Mail this to a friend!
Tnx. for the PSK31 info...My biggest problem w/PSK31 on my radio is that my IC707 has no ALC meter on it, so I have to just wing-it and run as little mic gain as possible to put out a 20-30 watt signal. This works okay for day-to-day activity on HF, but it's absolutely no good when I set up at Field Day. I basically have to use a different radio, mine just causes too much interference to other stations at the site.

I did notice the data port on the back of my radio, though, so I think I'll look into that in the future. It doesn't use the same kind of jack that my MFJ interface does to go thru the microphone/audio circuit, but I'm sure I could create an adapter or just wire it up to go thru the data port directly.

The Good Bad and Ugly on PSK31  
by W8JJW on December 12, 2005 Mail this to a friend!
Good Article.
You don't see much real-world help anywhere. Early on, I found it a challenge to make sure I was not overdriving the audio, since you can't see your signal on the waterfall, You had to rely on good operating practice and as mentioned, reports from others. I'm thankful for a couple ops that stuck with me during my adjustment excercise.
I've found using the data jack helps. Mine has a fixed audio output, so I don't have to worry about adjusting the AF gain. Once I set things to be clean, on the interface and sound card, it normally stays that way.
By the way, I had once had a signifcant hum caused by a shield ground that lifted off my interface card. I had no way of knowing it, the ALC looked "normal". I kept getting poor signal reports and couldn't figure out why. Finally someone actually listened to my signal and told be of the hum. (I don't normally listen to the warbles, I turn my speaker down)

When I come across a bad signal on PSK, most of the time the op was fairly new with PSK or had some other technical issue. I try to pass along help in a contructive way the way others have helped me.

Thanks for the pointers.
73, John
RE: The Good Bad and Ugly on PSK31  
by WI7B on December 12, 2005 Mail this to a friend!


ThanX for the article and images!

I was confused, however, by you reference to BPSK31 as having a 10 Hz bandwidth and your image captions with 10 Hz/div when they seem on eHAM to be 100 Hz/div. Given 100 Hz./div, your BPSK31 "good" signals are then observed with ~31 Hz bandwidth - which they should be (it is BPSK31 not BPSK10).

This insight from Peter Martinez, G3PLX, and BUXCOMM (below) was helpful for me to understand the nature of BPSK31 audio shaping so I reprinted here..

73 and see you on the waterfall,

---* Ken


There is a problem with PSK keying which doesn't show up with FSK, and that is the effect of key-clicks. We can get away with hard FSK keying at moderate baud rates without generating too much splatter, but polarity reversals are equivalent to simultaneous switching-off of one transmitter and switching-on of another one in anti-phase: the result being key-clicks that are TWICE AS BAD as on-off keying, all other things being equal.

So if we use computer logic to key a BPSK modulator such as an exclusive-or gate, at 31 baud, the emission would be extremely broad. In fact it would be about 3 times the baudrate wide at 10dB down, 5 times at 14dB down, 7 times at 17dB down, and so on (the squarewave Fourier series in fact).

The solution is to filter the output, or to shape the envelope amplitude of each bit which amounts to the same thing. In PSK31, a cosine shape is used. To see what this does to the waveform and the spectrum, consider transmitting a sequence of continuous polarity-reversals at 31 baud. With cosine shaping, the envelope ends up looking like full-wave rectified 31Hz AC. This not only looks like a two-tone test signal, it IS a two-tone test signal, and the spectrum consists of two pure tones at +/-15Hz from the center, and no splatter. Like the two-tone and unlike FSK, however, if we pass this through a transmitter, we get inter-modulation products if it is not linear, so we DO need to be careful not to overdrive the audio.

However, even the worst linear will give third-order products of 25dB at +/-47Hz (3 times the baud rate wide) and fifth-order products of 35dB at +/-78Hz (5 times the baud rate wide), a considerable improvement over the hard-keying case. If we infinitely overdrive the linear, we are back to the same levels as the hard-keyed system.

There is a similar line of reasoning on the receive side. The equivalent to "hard-keying" on the receive side is a BPSK receiver which opens a gate at the start of a bit, collects and stores all the received signal and noise during the bit, and then "snaps" the gate shut at the end. This process gives rise to the receive-side equivalent of key-clicks, namely side-lobes on the receiver pass band. So, although this "integrate-and-dump" method is 100% efficient in the task of sorting out signal from noise, it will only reject signals by 10dB at 3 times the baud rate wide and so on, the same spurious rejection figures that we got as spurious emission figures for the transmit side.

The PSK31 receiver overcomes this by filtering the receive signal, or by what amounts to the same thing, shaping the envelope of the received bit. The shape is more complex than the cosine shape used in the transmitter: if we used a cosine in the receiver we end up with some signal from one received bit "spreading" into the next bit, an inevitable result of cascading two filters which are each already "spread" by one bit.

The more complex shape in the receiver overcomes this by shaping 4 bits at a time and compensating for this inter-symbol interference, but the end result is a pass band that is at least 64dB down at +/-31 Hz and beyond, and doesn't introduce any inter-symbol interference when receiving a cosine-shaped transmission.
Our thanks and with credits to Peter Martinez, G3PLX for PSK31.

RE: The Good Bad and Ugly on PSK31  
by N6CRR on December 12, 2005 Mail this to a friend!

Right you are, 31 Hz wide, not the 10 Hz that I mistakenly stated. Also the article you added is quite useful on the technical aspects of PSK31. Thanks for adding it to the dialog.

So anyone want to help me write up how to calibrate your sound card to prevent the dreaded "I chase you, you chase me?"

73 Steve
RE: The Good Bad and Ugly on PSK31  
by G0SLP on December 12, 2005 Mail this to a friend!
Good thread. I'm still only a beginner on PSK31, despite having an Isoterm from G3LIV in the shack for 8 months now (no time for radio hi hi!); there are some really interesting topics being brought up.

Regards to all.

Mark G0SLP
RE: The Good Bad and Ugly on PSK31  
by LNXAUTHOR on December 12, 2005 Mail this to a friend!
- now this is a GREAT article!

- thanks very much to the author and posters...

(having a blast w/PSK31, gMFSK, xlog, a Sony C1XS, and a noMIC)
The Good Bad and Ugly on PSK31  
by KK9H on December 12, 2005 Mail this to a friend!
I operate PSK31 with an Icom 706 MkIIg, a RigBlaster Plus interface, Digipan and a 2.6 GHz Dell desktop computer. The radio is basically left exactly as it would be for an SSB QSO. The RigBlaster is connected to the MIC jack. When I adjust the transmit audio from the computer through the RigBlaster, I feed only enough audio with the WAVE slider to drive the radio's output to around 20 watts out and no more. There is no ALC indication on the radio's ALC meter at this level of power output. Using this method, I have had excellent IMD reports from -20 to -30 or more.
RE: The Good Bad and Ugly on PSK31  
by WA6HYQ on December 12, 2005 Mail this to a friend!
Do you or anyone know if a company like JPS makes a specifc DSP audio card for our computers that would work better for PSK31 or any of the other digital modes like Throb or ?
RE: The Good Bad and Ugly on PSK31  
by N6AJR on December 12, 2005 Mail this to a friend!
I find the best thing to do on PSK 31 and associated modes is to start at about 30 watts and turn it DOWN from there. I played psk for a while several years ago and found the ability to work almost any where, at 30 watts or less. we would start at 30 and run down till it started "skipping" on the rx. I worked lots of folks at 5 watts, ( low as my 756 would go..)

so use less, and make a better signal.
RE: The Good Bad and Ugly on PSK31  
by N6CRR on December 12, 2005 Mail this to a friend!
"Do you or anyone know if a company like JPS makes a specifc DSP audio card..."

Interesting question, but I would bet you already own a pretty good DSP based signal processing system.

Your computer!

You see your sound card is doing a Digital to Analog conversion, in some cases all in software, and then your computer is chewing on the bits like any good DSP should do.

If you want to read more about the role of sound cards doing DSP like work, check out the Software defined amateur transceiver by Flex @

They have some good write-ups about sound cards and how they work in processing signals.

RE: The Good Bad and Ugly on PSK31  
by RobertKoernerExAE7G on December 12, 2005 Mail this to a friend!
Nice prose.
Nice pixs.
Thanks for taking the time to write the article.
Merry Christmas from sunny Tucson
RE: The Good Bad and Ugly on PSK31  
by W4CNG on December 12, 2005 Mail this to a friend!
My sound card has a level bar that shows input level with green good, yellow almost too HOT and RED for Too HOT. Reducing the level into the computer with an external Audio Pot (Volume Control for others) and keeping everything in the green, makes for very good copy on PSK31. TX levels are also like described, No Processor, Mic/Ext gain to where Power Output levels are peaking about 30-40 Watts is good. I was testing my PSK31 set-up 4 years ago on my FT-817 at 2 watts output and was answered by an IK4 from my Attic Dipole. Great mode, not hard to do the right way. If your laptop does NOT have a serial interface for PTT, use the VOX PTT mode. Set the Vox gain a little bit higher, and the delay a bit longer, works great for newer Laptops without serial interfaces (Comm1).

Steve W4CNG
DEDs are DEAD  
by KA4KOE on December 12, 2005 Mail this to a friend!
Thanks to the fine folks at Eham, you will not see any future bonafide DEDs from me. Note that DED 20 was quickly pulled as it was not written by me, and no permission was asked of me in the first place by its author.

Any DEDs you see that are not written by myself, WA6BFH, WA2JJH, K4JSR, or KC8VWM (good friends), are done so without my permission, and are in violation of creative commons rights.

I am undergoing radiation treatments at the current time for tumors on my neck, and frankly have not had the wherewithal to write any more DEDs.

The DEDs are DEAD.

Thank you. It was a pleasure doing these articles for you fine folks. I have gotten many emails from you, and I do appreciate the kind thoughts.

73 and Goodbye

Electrical Engineer
DWG, Inc.
RE: The Good Bad and Ugly on PSK31  
by N3HKN on December 13, 2005 Mail this to a friend!
One other note - some operators set their rig's Power Output control to create the usual 50 watts, or less, of output. I have always found that setting the rigs power output to max and using the audio controls (rig, interface, or PC) gives me the best dynamic range. Right now I use an old Kenwood TS-520S. No interface. Audio goes directly to/from the "old" phone patch ports on the back and VOX does the keying. I tune it for max output and set the xmit level to 50 watts of tone modulated USB carrier using the microphone gain control. I have no ALC action.

Now if I could only figure out how to change the frequency of the analog VFO and turn the bandswitch from a PC....
The Good Bad and Ugly on PSK31  
by AC7ZG on December 13, 2005 Mail this to a friend!
A few additional hints:

Your IMD tells everything about your signal -- unfortunately you cannot read your own IMD without additional help:

1) Get a good psk meter to tell you what your IMD is:

- Clint makes a good one that samples the radiated signal from your antenna.

I've not used the other supplier -

(I have no financial ties with either of these, just a satisfied purchaser of Clint's meter)

2) Ask for a signal report but specifically request the IMD be sent back. Idle for a bit to ensure a reliable reading.

Generally anything better than -20 is a good signal, and better than -30 is a great signal. (>absolute(-20) )

3) Be cautious but don't dismiss the report of a wide signal. Get a second station to give you a IMD report. If both report poor IMD, start debugging your setup. If the wide/narrow reports differ widely then likely the station reporting the wide signal has some problem with their recieve path (they may need a pad in their input circuit), or some other propagational effect may be in play.

Example -- I had a report of a "wide signal" but several others broke in to report that the signal was narrow and not an issue. Don't discount any report of being wide, but that IMD report is quite useful (I note that the person reporting me wide gave me an IMD of -24 when I asked....).

4) If your signal is reported as wide, the first place to check is that you don't have any compression on your audio (turn off the DSP -- I've had this happen to me. Run a SSB QSO then switch to PSK31 and forget to turn off the compression. Ooops!). Second, lower the audio drive from the PC into your microphone input.

5) Don't assume that a poor IMD is a result of too much power, or that QRP will solve all the ills of a reported poor IMD.

Try this experiment in your next QSO -- ask the other station if they can report your IMD, then lower and then raise your transmit power, while asking for IMD readings. You may see a slight improvement in IMD when you are at 50-90% of max tx power over QRP signals. You shouldn't see much of a difference (I don't...YMMV depending upon your rig).

What about amps? Are they IMD offenders?
I've had reports that my IMD jumps from -24/-28 to -34/-38 when switching in an amp (unnecessary, but an interesting experiment).

6) If you see someone else's signal that appears wide -- don't be a LID by sending a nasty message sans your call (remember FCC part 90 regulations on station identification??).

The offending station does not see their own signal, wide or not. This leaves both parties frustrated -- the wide station has no way of diagnosing the problem, and the LID station keeps wondering why the wide station isn't magically fixing their problem.

Most of the ops on psk31 are truly interested in staying within 31kHz. Use standard calling procedure and work with the offending station in a them debug the problem. You'll make a friend.

7) A picture goes a long way to showing the other station that they are having a problem. You can always capture (windows -- ALT+PRINTSCREEN and email an image. This works best if you are working with them (sending a flame-o-gram by email isn't very effective. See #6). It does no good to just send it if you are not willing to work with them on the air to solve the problem.

Thanks for the interesting article. See everyone on the bands.


The Good Bad and Ugly on PSK31  
by N2JSO on December 13, 2005 Mail this to a friend!
One of the cleanest PSK31 signals I've ever seen was transmitted by an OM who stated that his other Ham interest was Hi-Fi SSB and AM. I'm no expert on the subject, but it seems to me that good audio is good audio, whether it's used for voice or data modes. Some of the techniques used by the Hi-Fi guys will help when creating a good, clean PSK signal.

BTW, the OM was running 100 watts on a TS-850. Lower power alone is clearly not sufficient for a clean signal.
The Good Bad and Ugly on PSK31  
by KB9YGD on December 13, 2005 Mail this to a friend!
Hello Steve & everyone.Thanks for and excellent well thought out article and also one that needs to be addressed.There are way to many hams sending out these very bad overmodulated signals on digital and i was one to when i 1st started.Its so simple to adjust the soundcard line in and volume and leave ur power all the way up then control the power output of the rig with the mic gain and get a nice clean signal.Thank you again for an article that needed to be published.73,Norm.
RE: The Good Bad and Ugly on PSK31  
by N6CRR on December 13, 2005 Mail this to a friend!

You are correct about power level, but one of the sort of traditions, if you can all it that, is for most folks to run lower power on PSK31. There's nothing wrong with running 100 watts, and you are very correct about power having nothing to do with how clean the signal is.

I can't count the number of OM's that I have worked that were QRP with a noodle for an antenna on this mode, all good copies. 5 to 10 Watts into ~30Hz gives some pretty good power spectral density!

The Good Bad and Ugly on PSK31  
by G0GQK on December 13, 2005 Mail this to a friend!
Since I've been using PSK31 I have never seen transmissions as bad as those you featured being produced by radio operators in Europe. In fact I would say the ones I've seen have been very clean and not overdriven.

73, Mel G0GQK
RE: The Good Bad and Ugly on PSK31  
by KC0KBH on December 13, 2005 Mail this to a friend!
All PSK ops have had bad signals I imagine at one time or another. Forget to turn of the processor, messed up the soundcard, misadjusted mic gain, for no apparent reason (;D), etc. I have been in the middle of QSO's, and on my next xmission, the alc meter jumps way up for no reason. Usually after I just turn the rig on and it hasn't quite warmed up. I have helped quite a few people adjust their signal from an absolutely terrible one down to a nice, clean signal. Most often it is their first psk qso, and I don't blame them. I thought you were supposed to run 100 watts at first. Then I learned that it is more like 50. Depending on the band, I can get my ic-765 up to 70 watts with a clean signal. I have a nice soundcard in the computer. If I can't do anything about a person's bad signal, I go to the opposite end of the waterfall and turn on the notch filter and block it out. What is really bad is when 2 overmodulated signals are close, and forming a big area impossible to operate in.
The Good Bad and Ugly on PSK31  
by NA4IT on December 13, 2005 Mail this to a friend!
Remember also, it does not take a KW to work PSK. It is a low power mode, meaning 50W or less (less IS more here!).
RE: The Good Bad and Ugly on PSK31  
by W9OY on December 14, 2005 Mail this to a friend!
I use a little applet called QuickMix that allows you to develope profiles for your sound card settings. It will set the various soundcard levels automatically. I write a script that turns on the program (eg hamscope) and then initializes the sound card to the correct settings using a quickmix profile. Click the script and the computer does the rest.,fid,6486,00.asp

RE: The Good Bad and Ugly on PSK31  
by K0LEJ on December 14, 2005 Mail this to a friend!

I have tried PSK-31 with a Dell notebook computer. The Dell only has a microphone in and no line in jack. The microphone interface overdrives the audio to a point where the waterfall in Digipan is WAY over driven (solid yellow). I did find an area in the audio control panel to cut back the gain by turning off the microphone boost, and that helped slightly. I have built a 40dB pad to cut back on the audio, but it's still not enough. I have a very old IBM Thinkpad that has a line-in jack, and that works okay, but it has a very slow 233Mhz. original Pentium processor. It would be nice to find a way to use the new Dell as I use that computer for my logging.

RE: The Good Bad and Ugly on PSK31  
by KF6RDN on December 16, 2005 Mail this to a friend!
I use a Dell for psk/mic in with a USB-Serial converter. I built my own interface, so was able to use enough resistance (via a POT) to cut down the drive where it wouldn't overdrive the mic line.

Maybe you can put more resistance to drop the signal in yours.
RE: The Good Bad and Ugly on PSK31  
by N6CRR on December 17, 2005 Mail this to a friend!

You can also buy "doggles" which are external sound cards that plug into USB ports on computers. They are supposed to be quite good, and add adjustments for setting the wave level and the master level on the output.

These things are pretty cheap being in the $30 US to maybe a high of $60. Check out for a list of them.

RE: DEDs are DEAD ???????????  
by KG5VK on December 20, 2005 Mail this to a friend!

what the heck is a DED ?

what the jeves are you talking about ?

The Good Bad and Ugly on PSK31  
by WB7OND on December 21, 2005 Mail this to a friend!

THe PSKMeter works pretty good, it controls the mixer level(drive) in the computer in response to your RF signal.

The Good Bad and Ugly on PSK31  
by K8AG on January 9, 2006 Mail this to a friend!
I think that a lot of the bad signals come from stations that have badly matched audio impedances or audio levels. The problem is, outputs from sound cards and laptop sound systems are not standard. Unless your system has a line out (and you are using it) phone or speaker outputs can cause a lot of trouble. They are supposed to look into something like 8 ohms. When hooked up to the microphone input (or data input) from the radio, we have, perhaps a 1000/8 ohm mismatch. This can cause significant audio distortion. In addition these output signal levels are probably too high to feed a mic input.

If you are using a phones or speaker output from a sound card, try dropping a suitable resistor across the terminals. Something like 10 ohms at the appropriate power rating should do. It will burn up some of the excess power that would cause the mic input of the radio to be overdriven. It will also present the sound card with a manageable impedance that it can use to provide less audio distortion.

Paying some attention to the highs and lows or audio impedance and signal level, can be the difference between a great signal and an overdriven nightmare.


Email Subscription
You are not subscribed to discussions on this article.

My Subscriptions
Subscriptions Help

Other How To Articles
High Capacity Battery Pack for Site Use Portable Operations
A Radial Plate for Cheapskates