Can I just say that was the most boring thing I have ever built and I still have 2 more 3db and one each of 2 and 1db boards to build. HAHAHA.
100w 50 ohm resistors, overkill i know. HAHA. For such a simple circuit that an swr bridge is, i really do not understand its operation. Yeah i get the diode rectifies AC to DC and we twiddle the knobs till we get an output voltage that suits the meters or ADC we are using. Its the bit above that which does not make much sense to me, like how does the transformer separate out forward and reverse powers. That is beyond my level of comprehension. And it seems none of my books know either and seem to take it on faith that it works.
Kind of obvious how it all works, pulling the BNC to ground changes over the amp relays from receive to transmit. 12v to the antenna relays to turn on the appropriate antenna port.
I will be controlling the antenna ports from a micro controller that is also changing over relays on the low pass filter board, which is also what I intend on building next.
When big black wires start to replace your beautiful traces, YOU FUCKED UP! HAHAHA
Other than the wiring error and the wrong footprint for the RCA jack, I think this amp tx/rx switching with 3 antenna ports should do the trick. Its been designed to handle 400w plus, but will get mated to a revision of the 100w amp I designed. I probably won’t even respin the board with fixes on it, I am only ever going to build 1 of these I do not need more amps than that HAHAHA
Started to design up the next piece of the amp puzzle. transmit and receive switching as well as various antenna ports for all the antennas. While the graphic shows BNC connectors, they are in fact So239’s. On transmit the exciter comes in the radio port, gets sent to the amp input, the output of the low pass filters goes to the amp out and gets routed to one of the antenna ports. On receive the antenna port bypasses everything and gets routed to the radio port and receiver.
The transmit and receive relays are controlled by a PNP switch which when its base is pulled to ground, switches the relays over. The entire thing will be controlled by a micro controller with either a touch screen or some buttons. Have not decided on which yet, that part is still in the what am i going to do stage 🙂 If i really wanted to go poverty, i could use a 3 position switch on the antenna relays. But who really wants to be agricultural when you can go all ferrari.
So today I learned a thing. Like why linear amp manufacturers spec their AC requirements so high. Like the Elecraft amp calls for 20A at 240v and a bunch of others say 15A etc. The reason why I am thinking about this is because I have been thinking about upgrading my license and then doing some sums to work out the costs of adding an amp to the line up.
If these things really did require ALL THE AC AMPS, I would have to pull a dedicated 20A line from the sub panel in the garage. Lucky there is room on the board to do this, but, its going to cost about $1500 to have a elechicken come and take half the roof off the house to pull a cable to where it needs to go and that is on top of the costs of an amp and the like.
My radio room already is on a 20A circuit that it shares with the water pump, bathroom and laundry, so leaving some overhead for the freezer and washing machine, there is room to spare, but not 15A, maybe 10A and then you would not want the wife to dry her hair in the bathroom. LOL
So why do these things need so much current? Inefficient power supplies. If they are using linear power supplies, because hams are scared of switch mode supplies, they might only be at best 60% efficient compared to say 80% efficient for even the worst switcher from MeanWell. And there is where I had my eureka moment. Math will give me a rough idea of the actual AC current requirements for any particular amp.
AC AMPS == DC Power (watts) / (Power Factor * AC Volts * Efficiency%)
So for your typical LDmos amp these days with 50V at 30A a maximum of 1500 watts, with linear power supplies 1500/(0.8*240*60%) you are looking at 13 Amps AC (guessing the power factor here). Compare that to a fairly efficient switch mode supply with power factor correction and your total AC current draw is going to be around 8 Amps or better.
This got me thinking, legal limit here is 400w and those numbers are for something closer to US power limits. So in theory, without knowing what the actual DC Power requirements are, but lets say 50V at 10A, thats 500 DC Watts and with a 90% efficient switch mode power supply we are looking at only around 2.5A AC.
That is not going to stress the existing household circuitry at all, in fact there is more than enough capacity to not even stress things even a tiny little bit. So maybe this upgrade and build and amp, yeah build one not buy it because I have the skills to design the boards and write the code to control it all, could be done for a reasonable cost, not in the 5 to 10K kind of mark.
So maybe this stupidity is actually within reach and I should give a little more thought to upgrading the license. Currently I am really close to finishing off 40m and 10m dxcc which will give me 4 bands and for the 5th band we are talking 80m and for that having 400w on tap with crappy dipole antennas might make dxcc a lot less painful.
So who knows, but watch this space. 🙂
Ignore all the burned flux residue, thats just there because I managed to toast all the fets when I accidentally tried to drive a 100w amp with 100w. HAHAHA, dumb arse move i know but shit happens. The good thing is, blowing up $4 worth of parts does not smart like blowing up $200 worth of LdMos.
But as far as things go, being a proof of concept, I really could not have been more happier. 100w out on 40m dropping down to 50w out on 10m, with little to no thought going into the design, I am sure I can improve the higher frequencies with a bit more thought on the design, input matching and general layout.
And on the subject of layout, I have worked out a lot of things i can do to improve things, like rotating the fets 180 degrees and using flood fills for the output traces. I also need to add in Tx/Rx switching and the like to make it perfect.
As for its performance, it draws about 6A at 13.8V, even with the sub optimum heatsink that is passively cooled, after 30 minutes of 50% duty cycle, the case of the fets did not get beyond 65 Deg C. All in all, i am very pleased with the performance and now get to redesign it, add in the improvements and build it all together into a box with filters and call it a thing 🙂
A work in progress, not going to say much for now, other than its working, made it tune the 40m 1/4wave onto 80m. Need to make some resistance standards to get its full tuning range, but I am hoping for something in the 5 to 5000Ohm range. Frequency wise its good for HF and maybe 6M, but who cares about 6M LOL.