- This topic has 14 replies, 3 voices, and was last updated 14 years, 10 months ago by
.
-
Posts
-
Hello. I built this transceiver. The receiver works, but I have a problem with the performance level. I can't get more than 0.5W from it, because it starts to bind at higher excitation.
http://sklep.avt.pl/photo/_pdf/AVT2310_1.pdf?sess_id=301efcf419994b7b86918bf8460e3fbb
I tried turning the coils to change different capacitors ,dust and ferrite BFO and VFO cores are normal .
Some of you also built this Polish transceiver ? How did you solve the final stage ?
if by chance you haven't studied it like the first one, I'll attach the Czech translation as a background: http://vitkov.xf.cz/amateri/Transceiver%20SSB.htm
I don't know how you set the output voltage, but I didn't have resistor R31 in the kit (1R) which is used to set the idle current. and after setting it is shorted with a jumper (which can be used and measurement, and I personally don't trust the jumper very much. Try to install it as a precaution (the jumper is quite far from the transistor)
The final amplifier is in “standard” connection and its variation can also be found on these pages. (https://www.cq.sk/modules.php?name=News&file=article&sid=37) Your excitation levels are fine? a IRF? Because only the bc211 could give half a watt. According to the article on these pages, it likes to oscillate if you connect the adjusting transformers incorrectly.
Which is not a problem as it is marked on the kit. (it is not significant at all. And the most likely connection is incorrect 😉 )As for the stabilization of the end, I have the scheme of SP2JQR, and he solves it with a CR circuit between Drain and Gate (D – 100n – 500R – G) but he installed an IRF530 there and increased the overall power to 10W
I don't know where the error is, I will make a new flat plate and solder it there again. I've been lurking there so much that my paths are coming off.
Nepoznate ine SSB QRP transceivre pre Novice pasma ? I thought about Forty one from F6BQU, but that is on the 40m band. I know that it could be adjusted, but obviously I don't have enough experience for that.
MKARS80:
https://www.cq.sk/modules.php?name=News&file=article&sid=1797
https://www.cq.sk/modules.php?name=News&file=article&sid=1687Another option is an SDR transceiver, very simple is the telegraphic TCVR EA3GHS Mosquito: http://sites.google.com/site/ea3ghs2/
TCVR Softrock are also simple, but it is quite difficult to get them.
SoftRock probably won't be right anymore just because of the need for a PC with two speakers (pre SSB)
As for novice pastures. So it can be done from Antek. It is more or less enough to replace the output filter, and the two bands + oscillate (ideally there data DDS)
Every “specialist” co to spravyl: http://sq9gl.nq.pl/TRX_ANTEK/sq9gl_antek.htm
Otherwise, most of the novice pastures are kits and freely available things only for 80m or so 160 (but there, novices have it in a rather strange way). on 10 and 15m is very little TRX.
Even TRX that knows more than one pass (mainly kits) end at 20m or at 15m
No Antek. I keep reviving him because it is not possible ,that it worked for so many people and not for me. The trimmer that is located at the power transistor IRF510 and goes to the Gate transistor , so at first when I play with it, I can hear perfectly in ODRA reception. Instead of a wattmeter, I have a 12V light bulb connected 250 mA, which is at the nominal current and power 3W a 48 ohm (I don't have any other performance indicators or benchmarks). The bulb is only faintly smoldering. When we get to a certain point with the trimmer, it starts to distort, and when I move again, the bulb shines at its maximum even without pressing the key. There is a hissing sound in the Oder and I can't hear it at all. So I tried to go slowly through the amplifier stages. 1. stupen s 2N2222A. I connected the oscilloscope behind the tuning transformer instead of the 2N2219. On the oscilloscope, after pressing the PTT and holding the keys, a nice sine carrier frequency will appear. I tried spinning the cores and I found a place where this first stage starts to spin, which is strange because the coils are shielded and connected to GND.. So I went further, I connected the 2n2219 and instead of the toroid I connected a light bulb (collector-power supply). After pressing the PTT and the sea of keys, I had to reset the coils in the first stage again because it started to weave. Well, after connecting the last stage with the IRF 520 it's a disaster. I bought the coils from GM electronics, and in addition to the fact that they are classic coils with a core, there is also a ferrite pot that slides onto the coil and increases the inductance. I put it there, now I'm thinking about trying it without it.
The only thing that occurred to me is that you have a poorly designed flat. Alternatively, you filter the individual voltages poorly and weakly. Otherwise, I can't imagine how he can carry it.
In addition, I will only take T5 back (2n2219)
T4 (umna 2n2369) is an ordinary zosik in A class without feedback. Only problem with filtration (c51,r40) which could theoretically establish a return link.
Are you using the components according to the documentation or are you experimenting?
I haven't set the end yet. Because I have a problem with the signal strength from DDS (which is enough for reception but not for broadcasting. A classic problem when a person starts reaching into the structure ;-))
But the documentation deals with revival rather casually and it seems like it's a piece of cake. (zero noise setting, of the pulsating current of the IRF) And if you want, you can play with the filter for maximum performance. And the end.
p.s. Otherwise, the bulb is not a good idea. If you measured the resistance of the fiber when it is fully retracted, it is much higher than when the bulb is cold. Rather, there are sudden resistances without indications. (In an emergency, a 50R terminator is also sufficient) Napetia zmeras on osciloscope.
Of course, I made the flatware myself , I actually have sociastics(even for such crap as coils and NE 612s ,)according to the documentation. I made the PCB with photopath, I made the template from the documentation. They published an insert that I edited in a pdf editor, mirrored, etc… I tried to avoid induction resistors, I use metallized and carbon ones. I chose the capacitors according to the photo. Where there were ceramics, I have ceramics, where there are scrolls, I have scrolls(especially the 100n) turned so that the last twist , which is located near the capacitor casing, directed to the ground if the capacitor has a connection to the ground and thus it seems to shade it. 50 Ohm resistance I also have BNC around 2W I think.
Method:
– balance the carrier (e.g.. with an oscilloscope)
– connect a 50-ohm resistor with an oscilloscope to the output
– on the R31 (1 ohm) connect a voltmeter and measure the current of the IRF520 in this way
– switch to TX (without excitation) and trimmer R33 to test quiescent current regulationHF voltage must not appear at the output, possibly only a small level from the carrier. The quiescent current must be able to be smoothly regulated and must not jump, e.g. z 0,2A na 2A.
IRF transistors are not classic RF transistors. It must be cooled already during the setting (see. e.g. AL block in my MKARS). It is possible to try another piece. And on what material and design are TR1 and TR2?
P.S. One sound card is enough for SSB SDR + relay
for OM0AAO : If I understood correctly, carrier balancing is done using the CW trimmer ? If so, I shot with it and when I have it in the middle I have the least tension (but not zero )originating from SA612 . Maybe I'll add some photos of my bizarre progress. on an oscilloscope. Because the SA612 already emits a key when pressed, like a carrier frequency and probably a VFO in it (or BFO ? I don't know, the signal is really small, so I have no idea what it is). It looks like the course of a higher frequency that oscillates at the carrier frequency. Of course, when it passes through the bandpass of the first stage at the output, the higher frequency does not appear.
I have ferrite toroids. There is also blue flak on them. Ale urcite su feritove. I have them from another Datal TRCVR that I received and it was in a working condition. It was on 80m, just like my Antek. 1cm outer diameter, 10of bifilar e winding with a lacquered wire of about 0.3 mm.
IRF 520 I have a quite large cooler with fins that is soldered on the flat connection to GND. He hasn't broken a sweat yet when I play with him.
For OM2AMB : Well, according to the first link I posted on this topic. If you open it right away, you have a photo of the built device. Of course, I also looked at SP5DDJ's site, who got 7W SSB from it. He has photos there, he also has such capacitors.
Yes, trimmer R10 is used for balancing. During TX, IO US2 mixes the NF signal with the BFO signal. The desired SSB sideband is selected by the crystal filter (and suppress unwanted products). Subsequently, the SSB signal will be mixed with the signal from the VFO in IO US3. The desired frequency is amplified by a series of T4 amplifiers, T5 a T6.
Parallel to C41, I would add an electric capacitor of at least 470uF and 1000uF/25V. C39 is in the power supply, but may be located quite far from T6.
With the help of an oscilloscope, it is thus possible to trace the signal levels and their course nicely. Specific measurement points: R44, collector T4, R39, collector T5, gate T6, drain T6 and output (everything towards the earth).
In the first step, do it without excitation, in the second with excitation. if it is possible, write all the measured voltages plus a description of the process (if it is other than sine).
I would not add a filter capacitor to the end. It is true that by default the condicator is far away, but the spinal relay 12V to 1mF condicator will not do any good. In addition, the TX branch will be powered for a long time even after connecting to the RX
I would rather connect R31 directly to the supply voltage.
1mF will not hold the TX branch for long. If the quiescent current were 0.2A, it will discharge in a few tens of milliseconds, while the TX circuits also stop working after the voltage is reduced. One of the modification options is to change R31 to a choke.
Ad R31.: 1R is there because of the kludge current setting. Then it is shorted (in the scheme there is an acquaintance ZWORA, in the list of components it also has the value ZWORA)
It and the voltages mentioned in the documentation are a bit strange. R36 should have 1.4V at its value 4.7 it was doing a good job (~0.3A) I get ~30mA there.
One way or another, he would come up with me and start over “standardnym” I am involved.
And if the original has a flat plate, then it is possible that the inlet and outlet taps are badly connected, which causes one basin to oscillate..
