开云体育

Home brew 5 band BitX

Richie Chambless All Messages By This Member #34365 Hey guys,? I’m finishing up my build of a multi band BitX and need two more bands to fill my filter board. I already have 80, 40, & 20 meters in place. Has anyone gone higher than 17 meters without an I.F. amp? (like the Hycas or similar)
Joe All Messages By This Member #34382 Nice to see someone working on multiband. Wondering if you might share project.? I have built bandpass filter for 20m and rx is good. The lpf for 20 is causing an issue. High swr. Tried several lpf? filters with same results. Joe VE1BWV toggle quoted message Show quoted text On Thu, Nov 9, 2017, 11:36 AM Richie Chambless <rchambless7@...> wrote: Hey guys,? I’m finishing up my build of a multi band BitX and need two more bands to fill my filter board. I already have 80, 40, & 20 meters in place. Has anyone gone higher than 17 meters without an I.F. amp? (like the Hycas or similar)
Mike All Messages By This Member #34465 Richie, I have been thinking of this same issue. My BITX40 on 20 meters is slightly deaf on 20 meters. I own an Hendricks BITX20 ( ) which does much better on 20. If you look at the schematics you will find an IF amp using 2 transistors. On the bench this amp seems to put out almost 2 times as much. Hope this helps Mike WA3O
Dexter N Muir All Messages By This Member #34466 You might take a look at Bill N2CQR's (he of SolderSmoke fame) rigs. He runs these up at 17m and I think 12m, and their (essentially Bitx) amps have an extra transistor. Pete N6QW ((his chat-partner) has some good explanations too ... 73 Dex, ZL2DEX
Ryan Flowers All Messages By This Member #34469 I'm doing multi-band as well, with modular filters. So far it works great on 80 and 40 (and presumably 60 once I build the filter!) but on 20, output is reduced from 5-7w to about 1w. I'm definitely interested in hearing others experiences. The receiver is definitely not deaf on 20m, as I've used it on 20m after cranking the bias WAY up on the PA (and dealing with the subsequent heat) but it seems like there's some other issue. The IRF510 works fine at 14mhz AFAIK. Input is welcome. -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
[email protected] #34470 开云体育 Great to see the work you guys are putting in to multi band trancievers .Personally I find the units so small that when I want another band I ?would buy another unit , my choice is 20M. I am wondering if any one is making one ? Brian VK3AAO ? Sent from for Windows 10 ? toggle quoted message Show quoted text From: [email protected] <[email protected]> on behalf of Ryan Flowers <geocrasher@...> Sent: Monday, November 13, 2017 7:44:35 PM To: [email protected] Subject: Re: [BITX20] Home brew 5 band BitX ? I'm doing multi-band as well, with modular filters. So far it works great on 80 and 40 (and presumably 60 once I build the filter!) but on 20, output is reduced from 5-7w to about 1w. I'm definitely interested in hearing others experiences. The receiver is definitely not deaf on 20m, as I've used it on 20m after cranking the bias WAY up on the PA (and dealing with the subsequent heat) but it seems like there's some other issue. The IRF510 works fine at 14mhz AFAIK. Input is welcome. -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
Gordon Gibby #34471 开云体育 This is sort of early, but I'd like to use the BITX40 to help more people in our local ARES group get on HF (many of them were VHF-only) -- and in the longer run I think they are going to need a bit more power.? ?I have filters waiting for me at home to try and add another band or two the rig (already ran it at very low power on 80, just enough to reach my WINLINK RMS on the other side of the house) ---? ? AC2CZ/VE2 's 50-watt dual IRF-510 amplifier caught my eye, particularly because he put in some protection circuitry after lots of experience with the system....and I'm going to try a printed circuit board of his design: ? ,_
Jerry Gaffke All Messages By This Member #34474 My first guess would be poor performance on 20m is due to the board layout around the IRF510. So maybe replace those parts with a final built ugly style on a small piece of copper clad FR4, then glue or solder that board in place.? IRF510 source lead should go directly to the ground plane, as short as possible. Perhaps add a 10 ohm resistor in series with the gate to avoid oscillations. Allison recommends clipping the drain lead from the IRF510 and soldering any drain connection directly to the tab, that keeps the AC at the drain from coupling into the gate. ? And of course, give it a good heat sink. Let us know what you find out. Jerry toggle quoted message Show quoted text On Mon, Nov 13, 2017 at 12:44 am, Ryan Flowers wrote: on 20, output is reduced from 5-7w to about 1w.
Ryan Flowers All Messages By This Member #34504 I'm not well versed in RF engineering, but I did some testing to try to understand power output of the bitx40v3 PA. I installed my 20M LPF, and bypassed the BPF so I'd get output on all frequencies. Measurements were done with my MFJ-949E Versatuner II, with output going into the dummy load, and watching the SWR meter on the 30W setting. Accurate enough for comparison, but maybe not for absolute accuracy. Either way, the results are obvious: Output drops sharply at the 8mhz mark, and continues to drop until it levels off around 2w output between 9.5 and 11mhz, and then goes down to 1w output at 12.5mhz. Conclusion: The BItx40v3 PA just isn't optimized for anything over 7mhz!?Of course this only confirms what you said Jerry. I wonder if any modifications can be done to the PA as it is to cause it to perform better. I'm not averse to building my own PA. I've never done it before, so it would definitely be an adventure! -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
Dexter N Muir All Messages By This Member #34509 Sorry, I haven't really been following this thread ... have you removed C91 and C92 (I presume you're using a DDS of some sort (the Raduino's SI5351?) via the DDS connector, so those are not needed (I get decent Rx feeding the Raduino direct to Q7 base, bypassing Q8 and Q9 - essentially emitter-followers - so I presume reasonable drive). The near-linear slope leads me to suspect increasing Xc somewhere. 73 de ZL2DEX
Ryan Flowers All Messages By This Member #34515 Yep, C91 and C92 have been removed for a long time so I could use USB on my QRP Labs VFO. I am not cool enough to know what Xc is or how to deal with it :P? -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
Jerry Gaffke All Messages By This Member #34516 You're plenty cool to be digging into this stuff so deeply. Xc is capacitive reactance. ? Which is to say that Dexter thinks there's too much capacitance somewhere. But could be lots of reasons for power out to drop as frequency rises. I'm still betting on just a bad layout around the IRF510. Capacitance could certainly play a role there. Try this: Remove power from the IRF510 drain by unhooking PA-PWR1. Now measure the RF voltage at the IRF510 gate while attempting to transmit using a scope or perhaps just a diode RF probe and DVM. If that RF voltage remains roughly the same at 20m as it is at 40m, then the issue is in the board layout right there at the IRF510. The original Bitx was for 20m, so we know this can be made to work properly. With the IRF510 drain not powered we are no longer concerned about IRF510's miller capacitance, or outside coupling from drain to gate, or parasitic oscillations. The primary load presented to T6 is now just the 47 ohms at R150 into the AC ground provided by C151. I think. Jerry, KE7ER ? toggle quoted message Show quoted text On Mon, Nov 13, 2017 at 03:20 pm, Ryan Flowers wrote: I am not cool enough to know what Xc is
Richie Chambless All Messages By This Member #34530 My BitX has Ubitx modules but arranged like the current 40 meter version. I’m using a QRP Labs vfo and a relay switched crystal bfo/carrier osc for USB/LSB. So far 80, 40, & 20 have plenty of audio, but I’m not sure if I have enough receiver gain for 15. For transmit I’m building one of those $15/45 watt amps from EBay (so far I don’t like SMD), so I’ll be happy if it makes 20-25 Watts. I will post those results when I’m done.
Ryan Flowers All Messages By This Member #34552 Richie, it sounds like yours will work fine at higher frequencies since it's designed that way- or at least I'd assume so.? Jerry, thanks to your help I have extended myself just a little bit further! I built an RF probe last night (which took FAR longer than it should have, since I spent an hour deciding which diode to use) and was able to get at least relative measurements of the IRF510 gate voltage across the same spectrum as I did the output measurements.? How I tested: Disconnected PA power, fed a 1500hz tone from my laptop speaker directly into my Bitx40 mic, and pushed PTT while measuring the voltage at the gate side of R150. I soldered my probe tip (a bit of 16ga copper wire) to R150 so that I didn't have to fiddle with it. That helped a lot! Enough about how I did it (which I explain in case my test was flawed) and on with the results of the test.? At 3mhz, it put out 3.54v (on my ammeter, not calibrated against anything) and at 14.5mhz put out 2.86v. I've added it to the previous graph for reference: It varies along the spectrum but only drops a little at about 9mhz, then spikes at 11.5, and evens out at 13.5mhz. The gate voltage only drops 19%, while the PA output drops 89%. I think this proves your theory, Jerry, that the PA is just not well designed for anything above 40m.? Also, Richie, sorry to hijack your thread here! I didn't realize you were using radio modules you'd built yourself. -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
Jerry Gaffke All Messages By This Member #34554 Ah, good job! Very thorough. Here's Farhans original Bitx from 2004: ? The stuff around the IRF510 is drawn up much differently than the Bitx40v3 schematic, but it is the exact same circuit. ? Hundreds of those have been built and I assume work well enough. Though that older one was built with coils using parts available locally, and are thus quite different. But this looks to me like a board layout issue. ? Jerry toggle quoted message Show quoted text On Tue, Nov 14, 2017 at 09:58 pm, Ryan Flowers wrote: the PA is just not well designed for anything above 40m
Ryan Flowers All Messages By This Member #34555 Thanks, Jerry. I appreciate that a lot! I suppose the next thing for me to figure out is: Do I continue trying to modify or solve the issues here, or do I call it "good enough" since I can use 80/60/40, and possibly even 160m. I'd?really like to get it working on 20m, as that's been one of my goals from the very beginning. I think it's time to fork this into another thread. Thanks again! -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
Raj vu2zap Member Profile All Messages By This Member #34556 Have a go with RD16HHF instead of the IRF.? Works for me, The leads are different so beware.. I have added a socket to one board with 4 holes. The 4th extra is soldered to the gate line closest to 2219. There is a convenient track there. Only that I have to plug in the RD16 reverse and one pin to the right. I'm sure you will get a better response at the output. Raj At 15-11-2017, you wrote: toggle quoted message Show quoted text Richie, it sounds like yours will work fine at higher frequencies since it's designed that way- or at least I'd assume so.
Ryan Flowers All Messages By This Member #34558 Thank you Raj! I have done some research on this and have decided to give it a go, so I'm going to order one or two of them and try it. My board is pretty well shot in that area, as I've had to replace the 510 so many times the vias in the board are done for (I overheated it, clearly) and so rewiring it won't be a big deal. Thanks for the suggestion!? -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
Gordon Gibby #34573 开云体育 So Ryan Flowers made the excellent measurements below, indicating that the gate drive voltage remained around 3 V AC while the output power dropped sort of continuously with increasing frequency. The forward transconductance of the device (at one point on its operating curves) is said to be 1.3? (amps per volt) so the 3VAC gate drive would have resulted (theoretically) in about 4Amps AC of drain current, which would be quite a bit of power....[which we certainly aren't seeing in his 20 meter results presented below] EXCEPT that there are inductances and capacitances in the output circuit...and I haven't analyzed all of them but the Source wire (internal to the device) has 7.5 nH of inductance and the Drain (internally)? has about 4 nH of inductance....the drain impedance isn't large compared to the? 12 ohms of the output circuit....but that source inductance + any wiring inductance would be a problem.... That source inductance by itself contributes 0.66 ohms of inductive reactance at 20 meters (14 MHz) and? IF you had 4A of AC flowing through it, you'd have 2.64V of NEGATIVE FEEDBACK dramatically eating away at your effective input drive voltage....ignoring the phase relationships, a simplistic view might be that the input drive is now reduced from 3 VAC down to 0.4 VAC....a 17dB loss....and gee whiz, no wonder we have so much less output power!!! Looked at another way,? the gain of most feedback amplifiers is sort of equal to their output terminal impedance (here, about 1/4 of 50 ohms or about 12 ohms) divided by their emitter or source resistance (here,? at a minimum the 7.5 nH of internal inductance, and probably a lot more due to lead lenths etc.? ? At 3 MHz that suggests a gain of? ? 12ohms / 0.14 ohms inductive reactance = approx 38dB At 14 MHz that suggests a gain of? 12 ohms / 0.66 ohms inductive reactance =? approx 25 dB [this is assuming a device with much higher internal gain, which isn't the case here, but it gives the theoretical effect] or 13 db lower gain.... and by his measurements he came up with 9 W output at 3 MHz and 1 W output at 20 meters --- a 13 dB reduction in gain.? ?(same reduction suggested by the theoretical analysis, perhaps a coincidence)? So I think possibly it is the CIRCUIT (especially source inductance) that needs to be dealt with.? ? In the 50 MHz high power iRF510 amplifier previously mentioned (http://frenning.dk/OZ1PIF_HOMEPAGE/50MHz_IRF510.htm)? you can see that the author soldered the source leads IMMEDIATELY??to huge ground plane PCB copper planes, and likewise the drains (off the metal tab not the wire pin) to low? inductance wiring.....I think all that may be the reason he got much better performance at higher frequencies --- and then one commenter had to add 5 ohms of SMD resistor in the gate lead to stop an?oscillation (perhaps because the gain of the circuit at VHF was now enough to sustain an oscillation). toggle quoted message Show quoted text From: [email protected] <[email protected]> on behalf of Ryan Flowers <geocrasher@...> Sent: Wednesday, November 15, 2017 12:58 AM To: [email protected] Subject: Re: [BITX20] Home brew 5 band BitX ? Richie, it sounds like yours will work fine at higher frequencies since it's designed that way- or at least I'd assume so.? Jerry, thanks to your help I have extended myself just a little bit further! I built an RF probe last night (which took FAR longer than it should have, since I spent an hour deciding which diode to use) and was able to get at least relative measurements of the IRF510 gate voltage across the same spectrum as I did the output measurements.? How I tested: Disconnected PA power, fed a 1500hz tone from my laptop speaker directly into my Bitx40 mic, and pushed PTT while measuring the voltage at the gate side of R150. I soldered my probe tip (a bit of 16ga copper wire) to R150 so that I didn't have to fiddle with it. That helped a lot! Enough about how I did it (which I explain in case my test was flawed) and on with the results of the test.? At 3mhz, it put out 3.54v (on my ammeter, not calibrated against anything) and at 14.5mhz put out 2.86v. I've added it to the previous graph for reference: It varies along the spectrum but only drops a little at about 9mhz, then spikes at 11.5, and evens out at 13.5mhz. The gate voltage only drops 19%, while the PA output drops 89%. I think this proves your theory, Jerry, that the PA is just not well designed for anything above 40m.? Also, Richie, sorry to hijack your thread here! I didn't realize you were using radio modules you'd built yourself. -- Ryan Flowers W7RLF ? ?<-- Learn how to go digital on the BITX40
K9HZ All Messages By This Member #34574 开云体育 Interesting.? I too use the RD16HHF1 or RD15HVF1 series parts every time I see a IRF510 or similar part used. ?Much easier parts to work with… ? ? ? Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch – K9ZC Staunton, Illinois ? Owner – Operator Villa Grand Piton – J68HZ Soufriere, St. Lucia W.I. Rent it: *Like us on Facebook!* ? ? email:? bill@... ? toggle quoted message Show quoted text From: [email protected] [mailto:[email protected]] On Behalf Of Raj vu2zap Sent: Wednesday, November 15, 2017 1:06 AM To: [email protected] Subject: Re: [BITX20] Home brew 5 band BitX ? Have a go with RD16HHF instead of the IRF.? Works for me, The leads are different so beware.. I have added a socket to one board with 4 holes. The 4th extra is soldered to the gate line closest to 2219. There is a convenient track there. Only that I have to plug in the RD16 reverse and one pin to the right. I'm sure you will get a better response at the output. Raj At 15-11-2017, you wrote: Richie, it sounds like yours will work fine at higher frequencies since it's designed that way- or at least I'd assume so.

Previous Topic Next Topic