IN REVIEW: Small Wonders Labs DSW-40 Transceiver
Special to The ARS Sojourner
Being a fan of all the Dave Benson, NN1G, rigs and not having built one in some time I was excited to hear the announcement of the DSW-40 from Dave. So it is with a great deal of pleasure that I write a review on one of the first ones to get built from a group of beta versions sent out by Dave.
Some background information for those new to what goes on in the engineering world. The development of any new piece of equipment requires a great deal of time, energy, money, and effort on the part of the designer and company in order to insure that the final product works and can be built by many hams worldwide. Variations in parts tolerances, etc. can make for some serious headaches for a company if care is not taken up front to insure stability and reproducability in results for a large number of builders. So there is the alpha-version which the designer builds/prototypes/etc. The alpha is the prototype that usually isn't seen outside the walls of the lab or off the workbench. Then there comes the beta-version where a few chosen individuals get an early version of the parts and boards and instructions and do a detailed report back to the owner on suggestions/problems/comments that hopefully add value to the equipment and end up in the final version or go to make the final version better. Beta builders get to find the typos and suggest some wording changes that might be a problem for the new builders. You old timers don't read the manuals anyway. :-)
This review is written from the building of a beta-version. There may be some changes that Dave comes along with from what is described here, but only minor changes probably as I find the rig worked just fine and is on the desk in use as this is being written. In fact, I just finished a QSO with K5ZTY in Houston with the rig running at 1W and using a long wire on 40 meters.
First of all let me list the appropriate information from Dave and Small Wonder Labs (http://smallwonderlabs.com).
Rig: DSW-40 Bands: 40 and 20 meters with others to follow Mfgr: Small Wonder Labs, 80 E Robbins Ave, Newington, CT 06111 dave@smallwonderlabs.com http://smallwonderlabs.com/ Avail.: At Dayton Hamvention and thereafter Price: About $90 for the board and board parts. Optional case to be priced by Dayton and shown at Dayton. Size: Board is about 7.0cm x 11.0cm which is the same size as the SW-40+ board Output: 2.5W nominal at 13.8V Supply Voltage: 8-15V Current: About 32mA on receive Sidetone: 800Hz, fixed SFDR: >60dB over specified frequency TTDR/MDS: equal to SW+ IF: ~500Hz for any band
All this information found on Dave's web page.
I won't bore you with all the gory details of building it. It went together in about 6 hours for me, so don't hold me or anyone else to that timeline. Take your time and do it right. I'd say allow yourself about 8 hours or so for the board. If you did the Elmer101 project and have built the SW-40+ or other similar single band rigs you can build this one. In fact, it is much simpler as I will explain.
The SMT parts for the DDS are already mounted and checked out by Dave. This is a neat start as even with my expertise and supply of building equipment I know that I don't have an soldering iron in the house to do the DDS chip (AD9835) which is the Direct Digital Synthesizer chip which has very small lead spacing. You do get to do two SMT inductors but these are done first and are not difficult. Just be careful and follow the instructions.
The layout is very similar to the SW-40+ and you'll recognize the similarity to it from comparison of both schematics if you have them. (View from the top. View from the side.)
There is a PIC processor (PIC16C622 from Microchip) which does the control of the DDS and is the keyer and freq-mite and RIT all in one small package....
I got the package from Dave in the afternoon but delayed until 9pm at night to start as I know how I am when I get started on these things. Sure enough at 3a.m. I have the board completely done. I find a gel-cell, the ear-buds, and a wire clip lead to hook up the antenna and then I power it up. No noise so I check and by clipping the antenna to the first point past the coupling cap from the PA filter I can hear noise. Tuning around I hear an EA8, DL7, G4, and a KH6 on 40 meters. OK, receiver is working but only for DX.... :-) I know I have a problem in the Cheby filter section so it's off to bed.
Next morning I get back to the rig and do a signal trace with the VE3DNL marker generator. Sure enough it looks like a short in the Cheby filter section. Turning the board over and looking carefully I find a connection between one of the filter caps and ground. As it turns out the board mfgr did this. Email to Dave confirms this as the board layout figures show no such land, so good thing manufacturers have a few boards made to find these kinds of errors and in this case it was not Dave's fault. Exacto knife fixes the problem in short order. OK, peak the transformer on the front end and the receiver is completely tuned.
Setup the paddle and check the transmitter and it is working. Only 0.9W out. I get the voltmeter out and the gel-cell is at 11.08V. Why is it that I always do this? But hey, it's working so keep on going. No chirp or any other problems and the voltage is within Dave's specs. I then make the only other adjustment you have to make and that is to set the receiver offset to match the transmitter frequency and I'm all done. It's just too simple.
As a side note on the low voltage. This is good news for the gang that likes to play in the fields with radios. If you battery gets low you can still keep working on the air but just at a reduced output power....
Then I take the rest of the day to take a TenTec TP-19 case and prime it with grey paint and drill the holes for everything. Pictures at the web site (http://www.qsl.net/k5fo) of all this work in progress. Give me an hour after you see this posting to get the pics out there and pointed to....
So, it now sits on the operating desk and on the first night I worked TX, AZ, and FL. The FL station being Mac, KF4KSM/QRP, a member of this group. All three Qs in response to my CQs. This was the 3rd of May and today, the 5th, I got a response from Bill, K5ZTY, in Houston to another CQ on 7.040MHz with no QRM/SSB/... A miracle unto itself. The other night around 7.040MHz I had a digital station very loud purposely follow me around causing QRM. There were no other stns nearby and when I moved he moved for about 5 times. Then I really moved up into the Novice band. This rig tunes from 7.000MHz to 7.3000MHz with no problems. See the notes below. As I write this with the earphones on I can hear digital stations pinging away on 7.040MHz and it's not RTTY...
So what we have here is a complete rig in a small package requiring only an ant, battery, and paddle externally and you are up and running. Neato.
So, in summary here are some notes from K5FO.
o Tuning range the entire 40 meter band 7.000MHz to 7.3000MHz Note that this takes a significant number of turns of the dial. :-) This at 6KHz per turn and I don't want to do the math for 1.5KHz/turn... o Constant output for the entire 40 meter band with just a minor degradation into the Novice band like 0.90 at 7.120MHz vs 0.95W or so at 7.040MHz and this probably due to ripples in the Chebyshev filter o On power up the rigs comes up at 7.040MHz on the nose. I suggested that Dave add a feature to come up in the Novice band if the RIT switch is on at power up. He'll note that if he decides to do it. o Keying speed 5 to 50wpm and it comes up at 15wpm o Freq is announced by pushing tuning knob in for an instant. The thing that I love is that if you are using the keyer at 40wpm then the freq is announced at 40wpm. No waiting on slow speeds here. o RIT is unlimited!! You can tune to the frequency you want to transmit on then tune the receiver to the DX station and that station can be 50KHz away. You really wouldn't go that far but you get the point. o Front panel layout that I used was TUNE, GAIN, RIT, Keyer Button, and PHONES. See web pics. o Rear panel layout PADDLES, POWER, and ANT. That's it. o Initial stepping is at 200Hz with the tuning dial using an encoder. But pushing knob in and holding it a sec or two you can get the finer 50Hz steps. I was asked how I like this and I like it fine. Now if you wanna move a long distance from 7.040MHz then you have some dial twisting to do, but it is fine as most of us won't move that far anyway. o 6KHz per dial revolution at 200Hz step size. Hint: put the knob that you are using, the usual Mouser in my case, with any dot in the vertical (12 o'clock) position when you power up the rig. This will be 7.040MHz. If you make 3 revolutions going up, then you'll be at 7.058MHz the FISTS calling frequency. :-) Neato. I did find that if I spin the dial rapidly then you can cause the micro to miss some increments but this is to be expected so don't tune too rapidly. o Filter is about 500Hz and you can miss a signal by tuning too rapidly. I've seen this before in the Uniden 2510 on 10 meters and all digital rigs I expect do this. o Keyer controls are speed, reverse to change dit/dah paddles to opposite paddles, tune, straight key. The thing that I like that Dave did here is that when you push the keyer button it starts cycling through the options and you don't have to sit and hold the button down to get to the next option. When the one you wanna change comes up you hit one of the paddles to stop and do whatever it is you want. o Equipment needed for alignment. None. You use your ears for peaking the front-end. The receiver is plenty sensitive. o There may be enough power to drive a small speaker but it is spec'd only for earphones and there is plenty of drive. o Some noise in the audio but this gets masked by atmospheric noise when the rig is connected to an antenna and the gain increased. o ONLY 4 toroids to wind and they don't have that many turns. o Only 2 tuning adjustments. o No test equipment needed if you don't run into problems. o Excellent receiver and I can't hear any digital generated noises that I recognize. o A thing that I learned in playing with the radio. Say you spin the dial and destroy the fact that 7.040MHz is with the dot vertically upward on the dial. Well, just tune in at 7.040MHz, enable RIT, move dial to desired position and then disable RIT. You immediately go back to 7.040MHz and the dial will be back where you want it. I don't know if Dave planned this feature but let's go ahead and give him credit for it. An undocumented feature. :-)
So, when guys and girls start getting this rig I expect to start seeing a lot of posts on QRP-L about how neat it is. As noted earlier $90 for the board and parts and I don't know what the entire rig with case and case mounted connectors, knobs, etc. is going to be. Dave will determine that later on. For those lucky individuals that are going to Dayton be sure to see the rig at the QRP gathering at the Days Inn South. (http://www.qrparci.org/) This URL from a faulty memory so don't hold me to it. Dave reports that he will have a number of the 40 meter board kits at Dayton, so be sure to take some extra funds this year if you are interested in getting one.
So, an excellent rig and one that I will enjoy for some time to come. One of things that people have to get used to is that this rig uses a lot of software developed by Dave and that won't be available at any price. So unless someone spends a lot of time and energy you won't be seeing this rig done HB style as in previous cases with other rigs that don't use micros....
My sincere thanks to Dave for letting me be in on the first group to build and test this little rig. I have placed an order for the 30M version at this time.
* * * * * * * * * *
Chuck Adams, K5FO, is an avid builder, QRP operator and administrator of the QRP-L Internet Mail Group. He lives in Dallas, TX.