A handy collection of answers to frequently asked questions
Enough audio from the computer?
Do you have enough audio from your computer sound card? You can check your Control Panel sound card settings by right-clicking on the ‘Speaker’ icon in the lower-right of your window. At the top of that pop-up menu is ‘Open Sound Settings’. You may see that this master volume control is not high enough.
i. Normal supply current at full ‘normal’ power out can range from 500-900 mA, depending on your board and computer. If it’s way above or below this range there’s probably something else going on.
ii. For starters, you want to make sure the components in the output low-pass filter are the correct values. If those are OK, doublecheck the values of the 1 ohm, 10 ohm and 100 ohm resistors in the Tx strip. Too many browns and blacks – It’s easy to mix those up.
i. A few people have experienced the Q4 getting so hot that the heat sink falls off and they discover that the final transistor Q4 has been smoked. In most of these cases we’ve found that user adjusted the drive control (R10) was turned all the way up, which is always a bad move. One should never adjust R10 to the fully-clockwise position! The PA device is capable of pulling two-or-more amperes of current under that seriously overdriven condition, which is the likely case for severe meltdown of the final transistor.
ii. The Q7 bias switch (MPS751) also usually glows a cheery orange before emitting smoke in this meltdown process. Bottom line: Make sure R10 is advanced no further than needed during extended 'key-down' testing or when using modes of long Tx duty cycle.
You can also do a close inspection of the components C18 (PA base) and the components C22-C25 and the toroids L5 and L6. A magnifying lens is helpful for this. The capacitor values need to be the ones shown in the instructions. Make sure that C24 and C25 are in the right location.
This is very important: Count the turns on L5 and L6. Each time the wire goes through the center hole, it counts as a turn. One extra turn on each toroid will decrease the output power a lot.
If the ‘PWR’ slider is almost at full-scale when performing the Adjustment section of the instructions (Appendix 5), you probably don’t have enough audio from your computer sound card. You should check your Control Panel sound card settings as it is possible that the master volume control is not high enough. You can always remove the cable from J3 and measure AC volts between ground (plug 'barrel') and plug tip. Experience has shown that 1-2V peak-to-peak is normally needed for sufficient audio drive from the computer.
"I had to turn the WSJT-X ‘Pwr Level’ to max and the R10 trimpot to CCW to get it."
a) There's an easy mod to raise the power output but only for Phasers on 17m, 20m, and 30m: Cut the lead on R45 (330 ohms, near L3). Simply snip the loop at the top of the resistor and separate the leads slightly. R45 is in there as a snubber. The lower bands (80 and 40M) are prone to an instability without it due to the use of a solenoidal-type inductor, rather than a toroidal-type inductor for L3.
b) Still, there may be something else going on that’s causing low power output- a misplaced component or an extra turn on one of the toroids. That extra turn will cause a surprising reduction in power output, so it's worth taking a look at.
Absolutely! There’s virtually no difference in 3.7W vs 4.0W on the receiving end, especially with the superior weak signal detection that the Digital modes bring us. Further, the fact that you didn’t have to put R10 up at its max range proves that you’ve probably got things just about right. To be sure, when you are adjust R10 you’ll note a point where the power output ceases to rise any further, which is the point you should *stop* any further increase of R10 … and then slightly back it off a little bit. You will then be assured of not overdriving your signal and thus have a much better chance of being received on the other end.
Not sure where the power level drop might be? Try this simple test. If you have a big rig to listen to your Phaser transmission with, you can do the following: Connect a clip lead (leave the other end free) first to R22. Note the signal strength. Then move the cliplead to the tab on Q3- the signal strength should increase quite a bit. Finally, move the cliplead connection to Q4's heat sink. That signal should be somewhat stronger than the Q3 reading. If it's the same- or weaker- Q4 may well be suspect.
“I wasn't expecting the heat sink on Q4 to be too hot to touch after only two CQs. Is that normal?”
a) The heatsink does indeed get warm, as it is supposed to. Even too hot to the touch after a while, but this is still okay. However the heatsink CAN get too hot if your Phaser is producing distorted signals due to overdrive. That’s why we emphasize not to crank R10 up all the way.
b) Heatsink temperatures can get pretty hot to the touch during extended TUNE or high duty-cycle transmissions. We’ve measured the temperature of Q4 and its heatsink at 170-degF after a 10 minute tune cycle, which is still within the safety window for safe device operation.
c) If you are still worried about having such a hot heatsink, or if you operate high duty-cycle ‘conversational’ modes such as JS8 and PSK31, then it is possible to install another heatsink (Digi-Key A10760-ND) ono the backside of the existing one, or a ‘heat spreader’ to the existing one.
If you need to replace the PA transistor and attach a heatsink, this procedure might work for you:
a) Use a chisel tip on the soldering iron - it provides better heat flow than a pointed tip. If you can, crank the temperature all the way up (e.g., 750-degF works well.)
b) Pre-tin Q4's heatsink tab first, then apply heat to the heat sink while it's resting on a heatproof mat, for example a fiberglass mat- the kind use by plumbers to avoid damaging wood surfaces in close quarters.) With the iron, it takes 4-5 seconds for solder to pool on the heat sink.
c) Use a piece of antistatic foam to hold the PA transistor, as the transistor gets too hot to hold and it is a static sensitive device.
d) Marry it up to the heat sink by applying the iron to the tab on the transistor. The heat sink is still hot and it doesn't take long before you see the solder flowing again.
e) Once the solder's flowing, remove the heat and use a small screwdriver to press down gently on the transistor. That holds it in place until the joint solidifies. You might be surprised at how long that heat sink will be too hot to touch!
As a practical matter, yes. The spurious requirement is no more than -43 dBc when operating below 30 MHz. Good amateur practice puts the carrier suppression at 40 dB and we don't make that number. Our harmonic content is all down 50 dB or better and there are no non-harmonic spurs. That's what matters. Short version: The Phaser is legal.
[We should point out that DSB rigs have an unwanted and unusable sideband at full strength... and probably a fair of amount of carrier leakage. Those are legal, though the Phaser does far better.]
“What ‘RADIO’ does one choose in the WSJT-X settings when using the Phaser?”
In Settings --> Audio tab. The Radio should be set to 'none'. That's because CAT control isn't applicable. To its right, click on 'VOX'. The Phaser switches to transmit whenever it senses audio from the computer sound card.
In Settings --> Audio tab. Audio settings are based on how your computer provides access to its ‘soundcard’. Many computers provide a separate stereo ‘mic input’ and a stereo ‘headphones’ output jack that may be conveniently used for connecting the Phaser. Others may need you to select 'Headset Microphone' for input and 'Headphones' for the output. You may also need to select which left/right channel is to be used for the audio, in which case you can try selecting 'left' for the microphone input and 'both' for the headphones.
External soundcards can be a convenient and inexpensive way to use for connecting to the Phaser. Oftentimes just plugging one of these little devices into the USB port will trigger Windows to find and install the driver required for it and you will be rewarded with a ‘ding’ sound when the computer is ready to use that new device. Then just go to the Settings --> Audio tab and be sure the Audio input and output selections indicate the USB source for the audio,
In general, the Phaser transceiver is capable of working any ‘linear’ amateur digital mode that transmits and receives audio tones within a 1200 Hz window in the upper sideband. Examples include FT8, JS8, PSK31, RTTY, Feld Hell, Olivia, MFSK, SSTV and more.
i) JS8 is a digital mode that uses the robustness of FT8 combined with a messaging and network protocol layer for weak signal communication on HF. Users are able to exchange arbitrary length messages using multiple 15-second frames to send/receive. JS8 uses a keyboard messaging style interface to enable hams to actually exchange normal QSO information operating under weak signal conditions. JS8 uses the FT8 protocol as its underlying structure, and its 15-second frame is comprised of 13 seconds Tx and a 2-second pause to receive an ACK from the other station. Then the JS8 protocol continues onward with transmitting its next bock of data in the next frame. (Note: This higher duty cycle mode of operation causes the Phaser’s final transistor to get warmer than when using most any other mode.)
ii) The Phasers will work with both modes FT8 and JS8 ‘out of the box’. The Phaser powers up with the green LED ‘on’ indicating the frequency is set to the FT8 window; and the user merely needs to tap the pushbutton beneath the yellow LED to switch to the ALT (Alternate) frequency, which is preset for the JS8 window in each band. While there are many FT8 users in the bands, as I’m sure you’ve seen with your WSJT-X app, there are not too many JS8 users … perhaps just too new or perhaps people tire of waiting for the full messages to come across. It’s not like the ‘conversational typing’ mode of PSK31, but JS* (and FT8) are most assuredly terrific for weak signal 2-way communications.
You can work RTTY stations with the Phaser. It's a little more work but totally do-able. One way is to use FLDIGI - the 'Swiss Army knife' of Amateur Radio. Move the ALT frequency a few kHz at a time and calling the stronger signals that you can copy. Because the Phaser receives both sidebands, half the signals won't decode. [You can swap the Mark/Space convention in FLDIGI to copy them, but you won't make contact.]
"The WSPR frequency is 7,038.6 kHz, but I can only program the Phaser’s ALT frequency with 1 kHz resolution."
i) The Phaser’s ALT frequency can be set to the WSPR frequency of 7,038.6 kHz, despite the lowest resolution being 1 kHz. Just calibrate the Phaser to be 400 Hz lower than the reference used – then when program the ALT frequency to 7039 kHz then you will actually be right on 7038.6 kHz. And when it comes time to run FT8 mode again, tapping the FT8 frequency button will put the frequency 400 Hz lower than usual for that mode, but you can then just use the upper end of the FT8 window in WSJT-X.
ii) Another approach is to use the WSPR app (WSPR2.0) rather than WSJT-x. You can set the dial frequency to 7038 and set the audio offset (under the advanced tab) to 2100Hz and the Phaser will work perfectly. Use a little math to set the TX frequency in the software where you want it, but this is lots easier than re-calibrating the Phaser.
“How can I determine when my Phaser is transmitting?”
A simple addition to your Phaser can serve as a Tx mode indicator. Just place a red LED and 3.3K resistor in series from the top of L3 (the 12V switched bus Vsw for the transmit strip) to a ground conveniently at the top of nearby R26. Then whenever your Phaser is commanded by WSJT-X to go into transmit, the red LED will come on. You might consider adding the LED to a new hole in the front panel right next to the arrow in the Phaser logo and your Phaser will look STUNning! [Note: We have added this LED-and-resistor to the Rev C pc board that started shipping in Round 4 kits.]
It's a transmitter and it's designed to radiate! <g> The imperative for shielding came in the TVI days and that's no longer a real concern. The other reason for shielding was due to 'hand-capacity' effects, and that too has been eliminated with the digitally-controlled oscillators.
The voltage at R34 actually stays very low (~0.5V) then suddenly flips up to about 4.5V when the WSJT-X ‘Pwr’ slider reaches a high level.
This operation with the 4V test is normal. Once the slider reaches the Phaser's threshold, the 4V bias will come on suddenly to switch the Phaser to Transmit. It is OK if that happens at almost full-scale on the slider, but that might indicate that you don’t have enough audio from your computer sound card.
Yes, you may see signals decoding with just a dummy load on the Phaser’s antenna port. There's gain to spare with the Phaser receiver and the computer sound card. I've gotten decodes with a dummy load connected to J1, and several other folks have noted the same thing. The traces going from J1 to U13 (the receiver mixer) are fairly long, and that's where the signal pickup is occurring.
A technique I use when replacing TO92-packaged parts like U7 and U8 is to just carefully snip off their leads close to the pc board. I usually do not try to unsolder the parts because I worry about messing up the pads by applying too much heat, lifting the pads off the board, etc. So when it comes time to add the new part …
a) I first clip the new part’s leads kind of short.
b) Then I place a nice solder blob on each of the pads.
c) And finally, I hold the part in the right position with one hand, nice and perpendicular to the board and next to one of those solder blobs, and then lightly solder one lead in position. I double-check that the part is upright, in the right orientation, and positioned with the other two leads (still unsoldered) next to their respective pads with the blobs. When everything is good, I carefully solder the middle lead to its blob, and then the last lead to its blob.
We considered 12V as the nominal supply voltage for the instructions. In reality it's more like from 10V (resulting in considerably reduced power output) to a maximum of 15V. The low end is perhaps as little as 7V, but there'll be miniscule output power if you use this supply voltage. At the high end, a 15V supply would be OK. If we had to recommend a range of supply voltages, it'd be 10-14V DC, with the understanding that it will work somewhat beyond that range. You can absolutely go to 13.5V or a little higher. 13.8V comes to mind as a standard transceiver supply voltage, and there's no harm in using that if you have it.
If your computer has a single 4-pin jack instead of separate ‘mic in’ and speaker/headset jacks, there’s a solution. There are several Y-adapters available:
a) Tripp Lite P318-06N-FMM (Digikey # TL1631-ND)
b) StarTech.com DH ‘ST 3.5mm Headset Splitter’. This one was available at a local computer service emporium.
Either of these also requires a 4-pin male-to-male cable between the computer and the Y-adapter:
c) Tensility 10-02153 (Digikey # 839-1407-ND). This one is 3 feet (0.9m) in length. Other lengths are available.
Ray AE5HN tells us that he was using a laptop with the 4-wire audio connector (TRRS),
not separate headphone/mic and wasn't getting the proper response. When he
realized that, he ordered an adapter from Amazon like
this one to solve the problem.
"PSKReporter shows my frequency as 7.0754 MHz. Should I recalibrate?"
The PSKReporter frequency indicates the actual signal, which is the AFSK shift from the base of the FT8 window up into the upper sideband. Look for the number listed next to your outgoing transmissions in WSJTX (e.g., 1400 Hz in your case) and add that to your nominal tuned frequency of 7.074 MHz and that should roughly match the PSK Reporter figure you see.
Another couple of ways to Calibrate ...
Set your frequency in WSJT-X to 1,000 Hz and you should hear a 1 kHz tone on the big rig when set to 14.074 USB. And if you 'zero beat' that tone my moving your big rig dial 'up', it'll end up being close the actual frequency of 14.074 +1 kHz = 14.075.
Or ... set the big rig dial to exactly 7075 kHz and the WSJT-X frequency to 1,000 Hz, and use the S1 or S2 pushbutton in the Phaser's Calibrate mode to zero beat the tone.
L-3, the coloring on the component is brown, green, gold and silver. Not brown,
green, gold and gold as described in the most recent assembly manual. The
‘silver’ band on the end is component tolerance of 10%, vs a ‘gold’ tolerance of
5%. Either is good for our use.
What is the audio
“I’ve run a copy of JTDX set to FT4 and successfully decoded FT8 (3.573) on WSJT and FT4 (3.575 Mhz) on JTDX at the same time. Opening up the waterfall it suggests the audio bandwidth is at least 5.5 kHz. Is this correct??
Yes- that's correct. The -3dB audio response on the receiver is on the order of
5.5 to 6 kHz.
"Running FT8 (3.573) on WSJT and FT4 (3.575 Mhz) on JTDX at the same time, looking at one water fall, is it possible to Transmit across the whole FT8/FT4 bandwidth (3.573-3.578 Mhz)?"
recommend against straying very far from the phasing circuitry's 'sweet spot' of
800 to 2800 Hz.. Unwanted sideband content comes up pretty quickly below 800
Hz. Above 2800 Hz, the unwanted sideband comes up more slowly. If I
recall, it's still down 30 dB at 3600 Hz, but I didn't carry the measurements
any further. You've always got the option of reprogramming the ALT frequency a
couple kHz higher for JT4 operation.
examples … TechRise USB External Stereo Sound Adapter Splitter Converter with
Volume Control for Windows and Mac,Plug & Play No Drivers Needed ... <https://www.amazon.com/gp/product/B01J7P0OGI>
As of the moment, a 10M PCB prototype on the bench is putting out 3Watts. The carrier is down 35 dB and the unwanted sideband is down 30 dB, as measured on a spectrum analyzer. Two significant changes from the existing Phasers are required. First- there's not enough Transmit strip gain. I've added 12 dB of additional gain using an inexpensive MMIC device. The other change eliminates the FST3257 devices (U5 and U6) and replaces them with ADE-1ASK mixers. George and I have not yet discussed plans for how and when this would hit the streets, and I'll ask folks for patience on this. We've got plenty else going on right now.
Thanks for asking! Yes- there's always that chance. I'm simply not there yet. Once we have the 10 Meter version under our belts, there are several more functional changes needed for 6M. The Si5351 won't clock at 200 MHz, so generating 50 MHz quadrature RF has to change. This'll take the form of some added low-pass filtering and a quadrature (Fisher) hybrid. It remains to be seen how much output power the Tx will develop. The FCC requirements for spectral purity are also more stringent above 30 MHz- -54dBc instead of the -43dBc for HF. That means additional low-pass filtering or a different approach for the PA…To be continued ...
“I like to measure the inductance before I install it. The inductance values do not appear in the schematic, parts list, nor PC board. “
a) There's an easy way to come up with the inductance values: Go to kitsandparts.com, select the toroid type, and it'll open a dialog box. You enter the number of turns and it'll spit out the inductance.
b) What you measure will generally not match those results- for several reasons. The winding integrity will affect the inductance. The biggest variable is the turns spacing. Inter-winding capacitance increases the effective inductance, so it pays to space the turns as evenly as possible. The inductance calculations don't take this into account, as far as I know. There's also some amount of lot-to-lot variation in the core characteristics. The Phaser does use toroids from KitsAndParts.com and Diz buys in large quantity. As such, the AL value will be pretty consistent from unit-to-unit
C) The other factor that may affect the measurement integrity is the measuring instrument itself. A meter that does a great job in the millihenry range may be less accurate in the 1-microhenry ballpark. A decent way to test this would be to use a 1uH 5% RF choke with short leads. You've probably checked this already and are confident the measurements are good.
d) All of which begs the question: What do you do with the results? They'll typically read higher than the calculated values. At some point, the measurements will differ from the calculations by enough to be meaningful. For the sake of discussion, a measurement 20% higher than the calculated value starts to become meaningful- IF you trust the instrument itself. The low-pass filter (L4 and L5 and related capacitors) will begin to roll off output power at the design frequency. The effect is like having an excess turn on one of those two inductors. If that's the case, it won't hurt to remove a single turn from the winding.
e) L3 isn't especially critical- it's part of an L-network that's based on a presumed 20-ohm impedance looking into the PA. L6 is non-critical- its impedance is on the order of several hundred ohms (IIRC). That's high enough that it doesn't materially affect the calculations for the collector impedance and low-pass filter.
23. Part number for U8 (78L33)
"In the kit I got a few days ago, U8 is listed as 78L33, the part in the kit is a 89M110E KY5033 TI .. .is this right?"
Yep! The KY5033 is what is being supplied now from our vendor as the '78L033'. Same 3.3V regulator.
L6 should have 17 turns, jut like in the previous version of the pcb. (The version C1 document has been corrected - sorry about that!)
The red Tx Indicator LED (D8) may be simply installed in the enclosure by drilling a 7/32” hole near the ‘arrow’ on the front panel (or anywhere else you might wish) and the sticking the LED in it. Any kind of glue/epoxy/hot-wax will work to hold it in place. Then solder two wires from the LED to the spot in the pcb it would otherwise go. The Enclosure Kits were designed and produced long before the idea of having a Tx indicator, so it’s up to the builder to ‘homebrew’ it in place, if desired. Otherwise just solder it in place on the pcb and things will still be fine … but you won’t see the Phaser go into Tx mode when in the enclosure.
Most of the kits now do not have FT8 and ALT noted on the outside of the enclosure front panel. There was a problem at the fab house on the labeling, but we decided that it wasn’t too much of a concern, as the operating mode is fairly obvious by the colored LEDs and their associated switches. One could place any label desired on the panel under these switches – say FT8 and PSK31 – just as their operating habits might uniquely be for their unit. Again, stay tuned on this one too and we’ll have some labels that user can download and apply if they wish.
Yes, the plastic molding around the shorter plugs prevent full insertion through the enclosure’s rear panel. For years I have standardized on using the longer power plugs in my kits (Mouser p/n: 490-PP3-002A) and we selected the hole in the rear panel accordingly. If you do have a shorter plug, you can easily make the hole bigger using a file before inserting the Phaser PCB. (We’ll make the hole larger in future orders from the pcb fab house.)
When the BNC nut is tightened to the Enclosure rear panel, the board will ‘tip up’ a little and the two nylon standoffs will be in the air about 1/4-inch at the front edge, which is just fine. The board is secure in this position.
Those using their Phasers in the field use the portable 12v battery to provide the juice. And for the bench in the shack, most just power the Phaser from the existing (beefy) 12V power supply used for powering the 'big rig'. But those who don't have a bench supply might try a good “AC adapter” that provides at least 12V at 2A. Something like this at DXEngineering.com or MFJ.com should work fine. Alternatively, you could try the Pyramid PS3KX.
Apply the 12V power but do not connect the audio cable. Then temporarily ground pin3 of U9 by briefly touching the pin with a wire which is connected to ground. That should cause:
1) U9 pin1 to toggle low [CHECK FOR IT]
2) and U9 pin7 to toggle high [CHECK FOR IT]
3) and the junction of R35 and R36 to toggle to about 1V [CHECK FOR IT]
4) and Q7 to toggle on, which supplies 12V to L3 the red LED to light. [CHECK FOR IT]
If things all work as described above …
5) connect the audio output from your computer to J3 and click on TUNE in the WSJT-X app.
6) Briefly run the ‘Pwr’ power slider up to maximum. Red LED turn on? [CHECK FOR IT]
6a) If no, the audio level from your computer either isn’t coming out or is simply too low.
6b) Plug earphones into the computer audio jack … Do you hear the tone when you click on TUNE? [CHECK FOR IT]
6c) If you hear a tone it may not be loud enough … Go to your computer Sound Settings and make sure the master volume is set to 100%. [CHECK FOR IT]
The PA transistor (Q4) is likely overheating whereby its internal resistance decreases and it draws more current- a positive feedback effect. The current limit function of Q8 is thus likely being triggered which causes a corresponding lowering of the voltage on the Vsw bus. If you monitor Vsw when the power dips, that will confirm it. This ‘current limiting’ circuit is described in the Phaser Knowledge Base.
The PA can get ‘too hot’ due to long transmissions and/or when the load (i.e., the antenna) has a high SWR. Always use an antenna that is resonant at the Phaser’s frequency, and if you engage in extraordinarily-long transmissions, you can try adding additional heatsinking to Q4 – see the Phaser Knowledge Base item #2 for examples and photos.
Current Limit Operation: Starting with revision C Phasers, the Transmit bias switching was significantly upgraded. The bias switch- a 2A-maximum PNP transistor- was replaced with a 16A-max P-channel MOSFET (Q8). Additionally- a new transistor (Q8) and ‘sense’ resistors (R47-A and -B) were added. Under normal Transmit conditions, the voltage drop across the sense resistor is not sufficient to cause Q8 to conduct. At about 0.7V (0.8 amps), though, Q8 begins conducting. When it does, it pulls the Gate bias on the MOSFET toward the positive supply. This is in the direction of turning off the MOSFET. It’s actually an example of ‘negative feedback’, and it sets a stable current-limit of about (0.7V/0.9 ohms) or 800 mA.
Why? The PA device (Q4) is rated up to 5 Amps max current. It was possible to approach this value when turning R10 fully clockwise for maximum drive. We replaced a number of PA device after their heatsinks fell off, or Q7’s ratings were exceeded. Q7 typically registered its protest by emitting smoke. The 800 mA current-limit function introduced by this change should be ‘transparent’ to normal Transmit operation.
Discussion List: Chat With The Designers ... https://groups.io/g/cwtd/topics
Questions? Email to George N2APB ...
Copyright 2019 Midnight Design
Solutions, LLC. All Rights Reserved.
Page last updated: June 30, 2020