Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
- BITX20
- Messages
Search
|
Re: sBitx antenna-first warning
Jack, How much voltage were you running on the drains? - f On Wed, Aug 3, 2022, 11:00 PM Jack, W8TEE via <jjpurdum=[email protected]> wrote:
|
|
Re: sBitx antenna-first warning
Jack, W8TEE
We've tested the T41 with its 510s to 40W and nothing fried. 80-40-20M all can deliver about 25W without much effort. 15M is about 20W and 10M drops to about 13W. Like I said, it is what it is... Jack, W8TEE
On Wednesday, August 3, 2022 at 01:14:06 PM EDT, Jerry Gaffke via groups.io <jgaffke@...> wrote:
The IRF510 can work dependably, Allison reports years of success with HF amps at 40W push-pull, surviving everything from accidental open antenna connections to dead shorts. In posts to this forum from several years ago, Allison went into plenty of detail. She even used them on 6m at several hundred watts (with bunches of them in parallel). The QSX amp from qrp-labs uses IRF510's in push-pull, does a consistent 10W from 3 to 30mhz. ?https://www.qrp-labs.com/linear.html Clean and linear even when operating from 12v, tolerates any antenna condition. A product of collaboration between Hans and Allison.? ?Cost of $26 USD for the kit, a steal. If I were building a rig like the sBitx from scratch, I would start with that amp for a PA, perhaps add PIN diode TR switching with the HV supplied by rectified RF. HFsignals could steal that PA design for the sBitx-v2, give Hans $5 for every unit if feeling guilty. Using the IRF510 for RF power amps takes careful design.? Especially heat sinking, as the IRF510 is not very good at getting heat from the die to the TO220 tab.? The IRF510 prefers more like 24vdc for higher power and reduced distortion. Improper design can blow the gate due to exceeding Vgs max spec momentarily Needs more drive than an RD16HHF1 to get similar power levels due to moderately high gate capacitance. However, Vds max is much higher for the IRF510 than it is for the RD16HHF1, so for cases of high Vds the IRF510 would survive a high SWR better than the RD16HHF1. Allison was of the opinion that moving to the RD16HHF1 on the uBitx would not help much. Reading from??https://www.vu2ese.com/index.php/2022/07/20/the-sbitx-the-sdr-for-the-homebrewer/ "The power amplifier of Figure 6 is a conventional power chain, probably a little sub-optimal. Two stages of pre-driver amplification with metal 2N2222As boost the signal level from -15 dBm to a little above 100 mW" For "sub-optimal" we could substitute "inexpensive".? One necessary change if wanting somewhat even power output all the way up to 30mhz would be an additional stage of amplification.? Gong from -15 dBm to +20 dBm (100mW) in two stages of 2N2222's is 35 dB of gain, or 17.5 dB per stage.? That's pushing it,? Though as we saw on the uBitx, optimizing a power amp is complicated, especially when doing it for a low budget product.? If NFET's seem too hard to use and just too dang complicated, an 807 or 6146 should do fine with any SWR when operating at 40W.? ?;-) Names could be shorter 80 years ago, they didn't have 10 gazillion different electronic parts to keep track of. Jerry, KE7ER On Wed, Aug 3, 2022 at 09:29 AM, Ashhar Farhan wrote:
-- Jack, W8TEE |
|
Re: sBitx antenna-first warning
The IRF510 can work dependably, Allison reports years of success with HF amps at 40W push-pull,
toggle quoted message
Show quoted text
surviving everything from accidental open antenna connections to dead shorts. In posts to this forum from several years ago, Allison went into plenty of detail. She even used them on 6m at several hundred watts (with bunches of them in parallel). The QSX amp from qrp-labs uses IRF510's in push-pull, does a consistent 10W from 3 to 30mhz. ?https://www.qrp-labs.com/linear.html Clean and linear even when operating from 12v, tolerates any antenna condition. A product of collaboration between Hans and Allison.? ?Cost of $26 USD for the kit, a steal. If I were building a rig like the sBitx from scratch, I would start with that amp for a PA, perhaps add PIN diode TR switching with the HV supplied by rectified RF. HFsignals could steal that PA design for the sBitx-v2, give Hans $5 for every unit if feeling guilty. Using the IRF510 for RF power amps takes careful design.? Especially heat sinking, as the IRF510 is not very good at getting heat from the die to the TO220 tab.? The IRF510 prefers more like 24vdc for higher power and reduced distortion. Improper design can blow the gate due to exceeding Vgs max spec momentarily Needs more drive than an RD16HHF1 to get similar power levels due to moderately high gate capacitance. However, Vds max is much higher for the IRF510 than it is for the RD16HHF1, so for cases of high Vds the IRF510 would survive a high SWR better than the RD16HHF1. Allison was of the opinion that moving to the RD16HHF1 on the uBitx would not help much. Reading from??https://www.vu2ese.com/index.php/2022/07/20/the-sbitx-the-sdr-for-the-homebrewer/ "The power amplifier of Figure 6 is a conventional power chain, probably a little sub-optimal. Two stages of pre-driver amplification with metal 2N2222As boost the signal level from -15 dBm to a little above 100 mW" For "sub-optimal" we could substitute "inexpensive".? One necessary change if wanting somewhat even power output all the way up to 30mhz would be an additional stage of amplification.? Gong from -15 dBm to +20 dBm (100mW) in two stages of 2N2222's is 35 dB of gain, or 17.5 dB per stage.? That's pushing it,? Though as we saw on the uBitx, optimizing a power amp is complicated, especially when doing it for a low budget product.? If NFET's seem too hard to use and just too dang complicated, an 807 or 6146 should do fine with any SWR when operating at 40W.? ?;-) Names could be shorter 80 years ago, they didn't have 10 gazillion different electronic parts to keep track of. Jerry, KE7ER On Wed, Aug 3, 2022 at 09:29 AM, Ashhar Farhan wrote:
|
|
Re: Can't transmit FT8
Press on the cq decode and then, F1 for "reply to cq" On Wed, Aug 3, 2022, 10:04 PM WB2OSZ <wb2osz@...> wrote: The builtin FT8 reception works great.? I'm hearing stations from around the world. |
|
Re: sBitx antenna-first warning
Interesting discussion. Just a few very quick thoughts I think that you have to consider the average of the input power to consider the input impedance of the power device as the instantaneous power varies widely under SSB conditions according to the voice characteristics, as indicated by Farhan in the examples he?gives of varying input power. Otherwise, one can say that the low pass filter will not work correctly Furthermore, I think that it is not the case that the SWR varies according to the power, but that since you have more power output, the reflected power will be greater?with a greater indication?on the SWR meter. If the LPF is designed for 50 ohms at its output end the?SWR is not going to change with power as can be seen if the TX is terminated into a suitable dummy load, but will only change with the reactance of the antenna / feeder which does not change with power. The pi network in the old valve tx's is not a good example?because it was a matching network to tune the tx and match it to the antenna.?? What you were doing was matching the tx to the antenna impedance, whatever that was, which did not change according to the power, but what was changing was the pi network impedance transformation between the antenna and the PA according to the PA input power. Just a few quick thoughts for your consideration Regards? Lawrence On Wed, Aug 3, 2022 at 5:59 PM Ashhar Farhan <farhanbox@...> wrote:
|
|
Re: sBitx antenna-first warning
On Wed, Aug 3, 2022 at 10:58 AM, Ashhar Farhan wrote:
The design question is, should we go for more expensive and difficult to get RF transistors or just use inexpensive IRF series and keep many spares?Farhan, The answer to that question would depend on instructions or guidelines that would minimize the occurrence of the failures.? If there is no way to use a tuner without a high risk of finals failure, then we need to go to the more robust part.? If there is a method to use a tuner that reduces risk, then I would vote for the lower-cost device.? This also depends on the devices' cost difference.? A dollar or two is low enough to change.? $20 is too high to change.? Between the two is a matter of how clear the guidelines on proper operation of the radio are. We are still in the early adopter's phase of the sBitx.? That is clear on the purchase page that risks need to be accepted by the purchaser.? It is up to us to develop the guidelines and evaluate the risks.? I believe that is what I have signed up to do. I like that we start with lower-cost parts, as failures are likely even with more robust parts.? If I make a mistake, the cost to correct it is minimized. Above are my opinions, which may be wrong. 73 Evan AC9TU |
|
Can't transmit FT8
The builtin FT8 reception works great.? I'm hearing stations from around the world.
However, I'm having trouble transmitting.? Software has been updated to the latest available today.? I've entered my callsign and grid square.? FT8 mode is semi-auto. Clicking on a received CQ line doesn't seem to do anything.? Pressing the CQ button causes sBitx to exit. Any ideas? Thanks & 73, John WB2OSZ |
|
Re: sBitx antenna-first warning
The RD16HHF1s are no longer in production. There are some really good parts from Nexperia. They have no easy names like 807 or 6146. They have names like AFT05MP70. Why have short names when you can make up names that can't fit on the package? - f On Wed, Aug 3, 2022, 9:45 PM Jack, W8TEE via <jjpurdum=[email protected]> wrote:
|
|
Re: sBitx antenna-first warning
Jack, W8TEE
We went through that same questioning process. Someone dinged us a few days ago because we're using the IRF510's. Out answer may be different because we are aiming the T41 at experimenters who may well blow a few while they're experimenting. Since we bought a bunch several years ago and still have some left, that's what we went with. If enough people are upset about the power drop above 20M, I'm sure someone will replace the PA board. Also, we felt a lot of experimenters have experience with the 510's and probably have some spares. Also, they're still easy to find (Mouser have 27000+!). Your situation is different in that changing the PA would be a little harder. Also, assuming they have spares is, perhaps, less likely. Also, I searched for the RD16HHF1 at Mouser and Digikey...nada. Jack, W8TEE
On Wednesday, August 3, 2022 at 11:59:04 AM EDT, Ashhar Farhan <farhanbox@...> wrote:
I guess the best way out is to use a robust transistor pair like the RD16HHF1 which can withstand infinite SWR, match the antenna regardless of the impedance at either end with an inline swr meter. Remember the ol' 807 transmitters with a Pi network? You just did a dip and load until you came up to the full power. The design question is, should we go for more expensive and difficult to get RF transistors or just use inexpensive IRF series and keep many spares? - f On Wed, Aug 3, 2022, 8:33 PM Jerry Gaffke via <jgaffke=[email protected]> wrote: Not sure which statement that would be. --
Jack, W8TEE |
|
Re: sBitx antenna-first warning
I guess the best way out is to use a robust transistor pair like the RD16HHF1 which can withstand infinite SWR, match the antenna regardless of the impedance at either end with an inline swr meter. Remember the ol' 807 transmitters with a Pi network? You just did a dip and load until you came up to the full power. The design question is, should we go for more expensive and difficult to get RF transistors or just use inexpensive IRF series and keep many spares? - f On Wed, Aug 3, 2022, 8:33 PM Jerry Gaffke via <jgaffke=[email protected]> wrote: Not sure which statement that would be. |
|
Re: A short video of sBitx working
There's an excellent wikipedia article about this transmitter:??https://en.wikipedia.org/wiki/Grimeton_Radio_Station
toggle quoted message
Show quoted text
"The Grimeton transmitter is the last surviving example of an Alexanderson alternator, the only radio station left from the pre-vacuum tube era, and is still in working condition. Each year, on a day called?, either on the last Sunday in June, or on the first Sunday in July, whichever comes closer to 2 July, the site holds an open house during which the transmitter is started up and transmits test messages on 17.2?kHz using its??SAQ, which can be received all over Europe." The good news is that you have almost an entire year to prepare for the next transmission. Jerry, KE7ER On Wed, Aug 3, 2022 at 06:05 AM, Niels Jalling wrote:
If the receiver goes down to 10kHz you can try to catch the sendings from the Grimeton transmitter at 17.2 kHz. |
|
Re: sBitx antenna-first warning
Not sure which statement that would be.
toggle quoted message
Show quoted text
I've been making lots of them. A resistive SWR meter assumes that the transmitter wants to see a 50 ohm resistive load, but the transmitter will not necessarily work best at exactly 50 ohms. For example, assume we are operating at 30mhz and there is a long trace to the antenna connector. The antenna system would want to have a little bit of capactive reactance to counter the inductance of that trace. Not much point to worry about such minor issues. We just need a reasonably good match between transmitter and the antenna system, doesn't have to be perfect.? You could spend a career trying to understand everything that could go slightly wrong here, many people do. Jerry, KE7ER On Wed, Aug 3, 2022 at 07:45 AM, Evan Hand wrote:
Jerry, |
|
Re: A short video of sBitx working
Yes, the transformer at T8 matches the extremely low impedance coming out of the IRFZ24N's
toggle quoted message
Show quoted text
to something near 50 ohms.? There is also a discussion here about an external antenna matcher to match that 50 ohms from the sBitx into whatever impedance the antenna presents.? The sBitx transmitter is designed to work from 30mhz to 3mhz, a factor of 10 to 1. Very tough to get transformers to work well across such a wide frequency range. Since T8 has to operate at 40 Watts, that is the most critical with regard to operating at 0.474mhz. Operating at 0.474mhz would best be done by increasing the windings on all transformers that must work at the operating frequency, namely T2, T5, T7 and T8. Jerry, KE7ER On Wed, Aug 3, 2022 at 03:04 AM, Alan Cooper wrote:
I understand the need for a low pass filter. That is needed for the IC 7300 also which I built a low pass filter for it which was easy enough. It produces real nice harmonic spurs right through the AM? broadcast band without one, and gives a normal watt meter fits because one cannot tell where the rf is getting produced is in the electromagnetic spectrum. It also causes an swr meter on the radio to read incorrectly since the reflected spurs power causes the rf foldback to activate and cut the transmit power way down. |
|
Re: sBitx antenna-first warning
My first guess for having different values of SWR reported at different power levels would be that the SWR measurement device cannot accurately measure the two power levels.? This does make sense to me.? So If you use a meter that can measure down to the low power level, it should be good to tune the tuner.
NOTE that the SWR meter is not part of the tuned circuit.? It is measuring at the cable's connector to the tuner.? The "measurement plane" in S Parameter terms. Am I missing something? 73 Evan AC9TU |
|
Re: Drive potmeters?
toggle quoted message
Show quoted text
On Wed, Aug 3, 2022 at 01:52 AM, Niels Jalling wrote: The potentiometers has minimum at the "right" place. Lowest at anticlockwise position and max at clockwise position. Just like a volume pot. |
|
Re: sBitx antenna-first warning
I would tune up any antenna matching network (if you have one) at low power,
toggle quoted message
Show quoted text
that should be close.? Then increase power and tune again. Alternantely, use something like an Antuino or nanoVNA to do the initial tune up. If you have an antenna you know to be somewhat tuned, a matching network is not needed. But you don't want to start at 40 watts, only to find that a windstorm took it down and you have a 10:1 SWR. For those who actually read my comments, it should be clear now. I started out this morning by reading the most recent post and am working my way back. Jerry, KE7ER On Wed, Aug 3, 2022 at 04:50 AM, Bob Benedict, KD8CGH wrote: "Many recommend tuning at a lower power. This risky as well. That is because at lower power, the output impedance of the PA will be different. Once you are tuned at QRP level, increasing the power will result in a mismatch again! " |
|
Re: sBitx antenna-first warning
If it works well, it's good.
toggle quoted message
Show quoted text
This does not strike me as all that odd. I wouldn't worry unless it somehow caused significant distortion in the transmitted signal. Jerry, KE7ER On Wed, Aug 3, 2022 at 05:21 AM, John AE5X wrote:
On Wed, Aug 3, 2022 at 02:45 AM, Ashhar Farhan wrote: |
|
Re: sBitx antenna-first warning
There is room for doubt here.
toggle quoted message
Show quoted text
An SWR meter assumes the transmitter wants to see a 50 ohm resistive load. A uBitx/sBitx might work a bit better into a load that is somewhat different than assumed. Jerry,? KE7ER On Wed, Aug 3, 2022 at 06:20 AM, David McGaw wrote: This is not correct.? The output impedance of the PA does not change the SWR and resistance/reactance of the tuner/antenna combination. |
|
Re: sBitx antenna-first warning
David, Here is the explanation as I see it. First, the impedance. It is the ratio of voltage to current. Let's skip RF and only consider DC, then we will expand this to DC.? Ex 1: Consider a 12 supply?that is delivering 40 watts of power to a resistor across it's terminals. It's current should be 3.3A, so that 12v x 3.3A = 40 Watts. Now, which also means that the resistor is 12V/3.3A =? 3.6 ohms. This is also the case of the full power of sBitx.? Ex 2: Now, consider that we have decreased the output of the power supply's current?only such that we are producing? 12 watts of power. The current will now be 1 A (as 12v x 1A = 12 Watts) and the load resistance required will be 12 ohms (12V / 1A). Hence, the load impedance is changed. If you still put a 3.6 ohm resistor, the voltage from the power supply will drop down and become unregulated and unstable, the power supply is overloaded. Now, let's switch to an RF 'power supply', that is, a transmitter.? The output impedance of a power amplifier, derived from the above explanation is given as (Vcc * Vcc)/ (2 * Po). That is, output impedance of the power amplifier is calculated as square of power supply voltage divided by double the output power. Ex 3: Consider a QRP transmitter like the BITX20 or QRP labs QCX. The Vcc is 12v and the output power is 5 watts, hence, the PA output impedance is (12 * 12)/ ( 2 * 5) = 14.4 ohms.? Ex4: In the above Ex 3, now consider that you have reduced the power to 1 watt. The Vcc is still 12V, hence, the output impedance is (12 * 12) / (2 * 1) = 72 ohms. In case of regular SSB transmitters, they are tuned to match at the maximum output to protect the transistors from excessive power remaining in the device at the peak output, it also makes sense as for CW and other continuous carrier modes like FT8 and RTTY, the output power doesn't vary but remains the same as the peak ssb envelope power. - f?? On Wed, Aug 3, 2022 at 6:50 PM David McGaw <david.g.mcgaw@...> wrote:
|
to navigate to use esc to dismiss