Re: Wow... 15 volts in and a bunch out..
Allison, what about adding an additional stage after the last mixer in the transmitter lineup? Better choice of transistors (replace Q90 and maybe the other 3904¡¯s)... add (insert) a tiny daughter board with an additional stage, could that allow for operating
the existing stages at more reasonable gain goals with more negative feedback? ??
Further ?Having intercepted the signal it might also give a point for bandpass filters to be easily inserted¡ª-
You might laugh at it, but even a single or dual-stage preselecter with variable capacitor(s) tuning a parallel filter, (like what I had on my heat kits),?once marked for the proper positions for each band, might significantly clean up most of them except
for the 21 MHz debacle. ?¡ª tune for maximum signal output making sure you¡¯re near the correct Mark. ?? If too much to ask from one inductor, add a tap for the higher end?
?Gordon
On Aug 21, 2018, at 11:51, ajparent1/KB1GMX < kb1gmx@...> wrote:
toggle quoted message
Show quoted text
Tom,
Back when I was resorting to the full bag of tricks the goal was not more power!
The goal was even power from 80 through 10 and improved distortion levels.
Alas, the were other issues found that superseded that work.??
For the moment it was and is possible to improve the amp.? The problem is?
All the spurious product presented to the amp would be amplified as well.
That and the basic modulator/filter/if had issues as well.? The result was a
good amp would be largely wasted.? Then a collective bunch of us noted
other things like compromised output filters that allowed all that excess to
escape.? I like Warren reached the point that salvage was not possible.
I did literally cut the board up to investigate root cause issues and the
conclusions are harsh.
The amp is a disaster.? The radio is 80-10 the amp never made it to 20m.
The choice of devices and layout were big issues and the IRF510 was
blamed?for most of it and it was already a mess before we got to them.?
Plain english, the 2n3904 was a weak choice and poorly implemented
as asking it for more than 10DB at at 30mhz was impossible.? ?Layout
made for further issues with stability.
Output filters due to layout.? Again layout at HF still counts.
Lack of band pass filters before the amp.? I'll repeat what I've said?
garbage in nets garbage out, minimal filtering is wishful thinking.
IF gain issues.? Asking for 18db from 2n3904s is not engineering its wishful.
There are a few that will and many that will not.? IF one asked for 10db it
would not be an issue.? On that same boat , the high gain of the transmit
path leads to spurious signal creation.? To go from the modulator to the last
mixer no more than 16db of gain is needed across the 12 and 45 mhz if
for transmit.? We have on average 32 to 28db of gain.? Too much gain
causes problems those include poor carrier to signal and overload.
On top of that? The "12mhz" IF and filter layout made it susceptible to
the 11.997mhz carrier further reducing the carrier to modulation.?
Board layout is the major issue not circuit failure.
The above applies to transmit only but echos of that are in the
receiver as birdies and spurious responses.
The whole 12V thing was a result of the TDA2822.? My board I put in a
regulated voltage?for the TDA and end of issue. By doing that 13.8 was
the standard.? However that does not fix all the broken.? We or at least
I was not?tied to 12V.? If you think fixing that by running 20V or more?
is a solution its clearly a CB solution.
The bottom line is mod like mad for power?and your amping up a
lot of spurious trash and putting it on the band.
Allison
|
Re: Wow... 15 volts in and a bunch out..
Tom,
As a follow up I've built my own amp and also worked with Hands on his and?
as a result I have two different amps that can do 3-30mhz with nearly flat
power and very linear.? That's at 12V (not 12V is really 13.8 noninal).
It I take both of those to 20V I get 20+ watts of clean for the effort
but consider that the design was for 10W with headroom so that
10W isn't full up distorted.
As to using a boot converter...? Hope the switching noise doesn't show up.
The relays will tolerate 16V all day.? The rest should be regulated anyway.
Allison
|
Re: Wow... 15 volts in and a bunch out..
Tom,
Back when I was resorting to the full bag of tricks the goal was not more power!
The goal was even power from 80 through 10 and improved distortion levels.
Alas, the were other issues found that superseded that work.??
For the moment it was and is possible to improve the amp.? The problem is?
All the spurious product presented to the amp would be amplified as well.
That and the basic modulator/filter/if had issues as well.? The result was a
good amp would be largely wasted.? Then a collective bunch of us noted
other things like compromised output filters that allowed all that excess to
escape.? I like Warren reached the point that salvage was not possible.
I did literally cut the board up to investigate root cause issues and the
conclusions are harsh.
The amp is a disaster.? The radio is 80-10 the amp never made it to 20m.
The choice of devices and layout were big issues and the IRF510 was
blamed?for most of it and it was already a mess before we got to them.?
Plain english, the 2n3904 was a weak choice and poorly implemented
as asking it for more than 10DB at at 30mhz was impossible.? ?Layout
made for further issues with stability.
Output filters due to layout.? Again layout at HF still counts.
Lack of band pass filters before the amp.? I'll repeat what I've said?
garbage in nets garbage out, minimal filtering is wishful thinking.
IF gain issues.? Asking for 18db from 2n3904s is not engineering its wishful.
There are a few that will and many that will not.? IF one asked for 10db it
would not be an issue.? On that same boat , the high gain of the transmit
path leads to spurious signal creation.? To go from the modulator to the last
mixer no more than 16db of gain is needed across the 12 and 45 mhz if
for transmit.? We have on average 32 to 28db of gain.? Too much gain
causes problems those include poor carrier to signal and overload.
On top of that? The "12mhz" IF and filter layout made it susceptible to
the 11.997mhz carrier further reducing the carrier to modulation.?
Board layout is the major issue not circuit failure.
The above applies to transmit only but echos of that are in the
receiver as birdies and spurious responses.
The whole 12V thing was a result of the TDA2822.? My board I put in a
regulated voltage?for the TDA and end of issue. By doing that 13.8 was
the standard.? However that does not fix all the broken.? We or at least
I was not?tied to 12V.? If you think fixing that by running 20V or more?
is a solution its clearly a CB solution.
The bottom line is mod like mad for power?and your amping up a
lot of spurious trash and putting it on the band.
Allison
|
Re: Intermodulation Performance
On Tuesday, August 21, 2018, 3:35:46 AM PDT, Warren Allgyer <allgyer@...> wrote:
I am finished. This is simply not worth the effort. When it stops being fun, its time to quit \
Jim
|
Re: QRP SWR meter recommendation?
#ubitx
Jack,
Very interesting. Thanks.
Tom, wb6b
|
Re: Wow... 15 volts in and a bunch out..
So a question. I know some of the power boost suggestions like shorting out parts the were put there to provide stability in the amplifiers or overdriving the audio, do indeed increase power, but at the cost of vastly increasing unwanted emissions.
However, are we working too hard to optimize the RF chain, finding super transistors and such, just for the sake of being tied to 12 volts. I don't want to see my audio amp IC go up in flames or make the relays melt, but what if we selectively applied an increased voltage, say 18 volts to pull a possible voltage out of the air, and applied that to the finals, the infamous Q90, and Q92, 93, 96, 97, 911, 912, and possibility further back in the TX amplifier chain?
Would that result in a RF chain that has more useful gain to work with and allow more headroom for other power leveling efforts?
I have some new 150W boost converters coming from Amazon. They are around $4 each. It seems like for that low cost, freeing ourselves from the bonds of 12 volts may be worth it.
On a side note the boost converts have the worrisome name of "Gowoops". Sounds like if they live up to their name I should expect more smoke than volts to come out of them.
Tom, wb6b
|
Re: QRP SWR meter recommendation?
#ubitx
If you want SWR only I just got a nice cross needle MFJ-822.?? It covers 1.8-200 MHz, 30 or 300 Watts.? On the 30 watt range half scale is about 7 watts.-- Lee - N9LO? "I Void Warranties"
?
|
Re: QRP SWR meter recommendation?
#ubitx
Tom:
You can get a marginal speed improvement by moving the data definitions out of the loops. For example:
#define ELEMENTS(x)? (sizeof(x) / sizeof(x[0]))? // Gets rid of "magic numbers" in the code
float scaledReadin;for (int pin = 0; pin < ELEMENTS(_analogValue); pin++) {
?? if (_analogActive[pin]) {
???? scaledReading = analogScalingFactor * (float)analogRead(pin);
???? _analogValue[pin] += (scaledReading - _analogValue[pin]) * _analogFilterValue[pin];
}
}
The macro ELEMENTS() allows you to determine the number of elements in an array automatically a compile time, thus eliminating the need to hard-code array sizes into the body of the program. That way, you can simply change the definition of the array in one place in the code, recompile, and the new size is automatically changed everywhere in the code. The really nice thing about the macro is that it's completely typeless...that it, it will work with any aggregate data type.
Moving the definition of a variable before and outside the loop saves the compiler the instructions necessary to adjust the stack pointer on each pass through the loop in the "in-scope, out-of-scope" aspect of a data definition within the loop. Quite honestly, on 8 iterations of the loop it will not likely be noticeable. Still, in other programs where SRAM is pretty tight, it could mean the difference between running and a stack crash.
Jack, W8TEE
On Tuesday, August 21, 2018, 1:02:39 AM EDT, Tom, wb6b <wb6b@...> wrote:
Here are some code snippets from a project I did. It calculates SWR from the voltages read from analog pins A6 and A7. It does a number of other things, so here are the basic pieces of code for SWR extracted out into small snippets. This first code segments captures selected analog pin voltages and applies an exponential smoothing filter to them.
for (int pin = 0; pin < 8; pin++) {
if (_analogActive[pin]) {
float scaledReading = analogScalingFactor * (float)analogRead(pin);
_analogValue[pin] +=
(scaledReading - _analogValue[pin]) * _analogFilterValue[pin];
}
}
This next slice of code captures the max voltage and applies a slow decay to the max value.
// Capture the peak analog values with a slow decay time.
// The decay time is a starting point. May want to adjust it to taste.
float decayValue = 0.001;
for (int pin = 0; pin < 8; pin++) {
if (_analogActive[pin]) {
float analogValue = _analogValue[pin];
_analogValueMax[pin] = max(_analogValueMax[pin], analogValue);
_analogValueMax[pin] += (analogValue - _analogValueMax[pin]) * decayValue;
} else {
_analogValueMax[pin] = 0.0;
}
}
And here the SWR is calculated from the Max values.
// "SWR"
float vf = _analogValueMax[6];
float vr = _analogValueMax[7];
float vd = (vf - vr);
vd = max(vd, 0.002); // Limit max SWR reading before divide by zero
// Calculate SWR from directional coupler voltages.
result = String((vf + vr) / vd);
You can see the full code here: Tom, wb6b
|
Re: Nextion 3.5 with CEC 1.097
Thank you Nikkos.
Michel F1GTX
|
Re: uBitx diagnostic document from Facebook group
#ubitx
I must have missed something, where's this guide?
Roy WA0YMH
On Thu, May 17, 2018, 11:15 PM Tom, wb6b < wb6b@...> wrote: Thanks to?Ufi Auttorri for creating this guide. It was certainly a lot of work to do this.?
|
Re: NEW DISCUSSION GROUP SPECIFICALLY FOR HOMEBREW TEST EQUIPMENT
toggle quoted message
Show quoted text
On Tue, Aug 21, 2018 at 5:40 PM, Adrian Waiblinger <vk5zbr@...> wrote: Good idea, ?getting harder to get good second had test equipment
The new discussion group has been formed.?
?
Three persons have volunteered to be moderators (Gilles, Dennis, and Jim).? I will make them moderators
just as soon as they have subscribed to the new group.?
Concept for this group is that it will focus on Amateur Radio Test Equipment that you have designed and built.
Operation and testing with homebrew test equipment is also a suitable topic.?
The intent is to stimulate design and construction of near-lab-grade but inexpensive test equipment for use by
radio amateurs.? I am guessing that a lot of this will involve micro-controllers, but we shall see where the
discussions take us.?
Arv? K7HKL
_._
|
Re: NEW DISCUSSION GROUP SPECIFICALLY FOR HOMEBREW TEST EQUIPMENT
Good idea, ?getting harder to get good second had test equipment
toggle quoted message
Show quoted text
On 21 Aug 2018, at 10:31 am, Arv Evans < arvid.evans@...> wrote: The new discussion group has been formed.?
?
Three persons have volunteered to be moderators (Gilles, Dennis, and Jim).? I will make them moderators
just as soon as they have subscribed to the new group.?
Concept for this group is that it will focus on Amateur Radio Test Equipment that you have designed and built.
Operation and testing with homebrew test equipment is also a suitable topic.?
The intent is to stimulate design and construction of near-lab-grade but inexpensive test equipment for use by
radio amateurs.? I am guessing that a lot of this will involve micro-controllers, but we shall see where the
discussions take us.?
Arv? K7HKL
_._
|
Re: uBITX new build, quiet audio
#ubitx
Jeff,
I have the same impression and experience like You!
I prodduced a so called level diagramme, see may attachment.
Interestingly thsi result was verified on my UBITX: at -107 dBm
(=1?V) at the antenna input I measured about 300 mVpp? = 106 mVrms
at the louespeaker terminals; audio control potentiometer at full
gain. This level (into a 8 ohm impedance ludspeaker) gives about
1,4 mW of power.
Doing the samce lvel diagramme for the BITX40 gives 15 dB higher
gain: thre RF stages?? (but lower gain each stage), (two on the
UBITX) and a 6 dB higher audio gain (LM386). 10^(15/10) gives a
poer factor of 31.67, so the BITX40 would output 44.27 mW under
the same input conditions. ? ?
I also noticed a large amount of RF still at the loudspeaker
terminals, about 100 mVpp. Using a frequency counter I observed 12
MHz, so this RF are leaks of the BFO.
One remedy coud be to palce a 100 pF cap across R50 (AF preamp).
Henning Weddig
DK5LV
?
Am 21.08.2018 um 12:38 schrieb
jfein2000 via Groups.Io:
toggle quoted message
Show quoted text
I have a current version of the BITX which I just finished
assembling in the case.? It works after some
debugging....replacing the code in the Arduino due to a cw fault
per another discussion.? It is now working on ssb and cw
although power output should be higher on 80 and 40 meters. BUT,
I also feel the audio is too low.? I don't have a hum issue but
at full volume I can hear stations but even with an amplified
speaker it is just not up to snuff.? I saw Arv's suggestion for
wiring an amplified speaker to the volume control and will try
that to see if it helps. I am responding to this post as the cw
issue seems to be? general fault and with two of us with recent
bitx builds having audio issues I felt it would be good to get
this on the radar in case it is a trend building. ? Any other
suggestions to remedy would be appreciated. ?
cheers
Jeff?
|
Re: Intermodulation Performance
Hi,Warren, I will always appreciate the data that you generated!
Nicer radios, are certainly nicer radios! Some people like Honda FIT , some prefer a high end BMW!
For the purposes of coaxing local club members into learning HF and some radio, some building techniques, & electronics, this is perfect!
For HF ALE monitoring, it even beats the heck out of my ICOM 718¡¯s. For winning a contest? Nah!!
Thanks for all your input
Gordon
toggle quoted message
Show quoted text
On Aug 21, 2018, at 06:35, Warren Allgyer <allgyer@...> wrote:
response
|
Re: Nextion 3.5 with CEC 1.097
¦§i Michel,
I already know that the software has some issues, but it's the best option i've found thanks to Rich Neese.You can also find GUI here :?
SV9CVJ? Nikos Magafourakis
|
Re: uBITX new build, quiet audio
#ubitx
I have a current version of the BITX which I just finished assembling in the case.? It works after some debugging....replacing the code in the Arduino due to a cw fault per another discussion.? It is now working on ssb and cw although power output should be higher on 80 and 40 meters. BUT, I also feel the audio is too low.? I don't have a hum issue but at full volume I can hear stations but even with an amplified speaker it is just not up to snuff.? I saw Arv's suggestion for wiring an amplified speaker to the volume control and will try that to see if it helps. I am responding to this post as the cw issue seems to be? general fault and with two of us with recent bitx builds having audio issues I felt it would be good to get this on the radar in case it is a trend building. ? Any other suggestions to remedy would be appreciated. ?
cheers
Jeff?
|
Re: Intermodulation Performance
I am finished. This is simply not worth the effort. The difference between this radio and my FT-817 is about $350. The difference between the uBitx and my HobbyPC RS-HFIQ is $100. Both are worth the increment in order to assure:
1) FCC/ITU compliant transmissions
2) Excellent IMD and in-band spurious response
3) Adequate audio output
4) AGC
5) ALC
Not to mention FM, 6m, 2m, 70 cm for the Yaesu.
I completely understand the attraction of making up your own box, modifying the code, and other customizations. But there is simply too much to be added to the basic uBitx as delivered in order to make it a compliant, user-friendly, neighbor-friendly rig. I maintain it is not possible to make compliant transmissions from this board without the knowledge and equipment to modify it for compliance and a properly equipped lab to set it up and monitor it.
My board is going on the shelf marked "Nice try....... not worth the effort", along with a complete set of filter relays, failed output filter attempts, octal to decimal decoders to drive the filters, 45 MHz SAW filters, and a stack of test data. It was fun. It is done.
WA8TOD
|
Re: Nextion 3.5 with CEC 1.097
Thank you Nikos. I tried 270tft the screen is reversed. 90tft is fine except the S / Meter deviation is reversed.
friendships
F1GTX Michel
|
Re: Nextion 3.5 with CEC 1.097
I have ubitx_35_basic_90.tft and the card reader is down ,if card reader up load ..... 270.tft file . 
|
Re: QRP SWR meter recommendation?
#ubitx
Here is an example of one of the little swr indicators ?where you just dim the diode
I¡¯ve used that kind of a circuit as well as the two torrid system ?others recommended which also works fine but you¡¯ll need an external meter or voltmeter or something.?
toggle quoted message
Show quoted text
Hi Mike,
The tuner itself does not care how much power you use to 'set' it as long as it isn't *more* than it can handle. It is not impressed if your TX can pin the needle or not. If your uBitx will only drive the meter to half scale in it's "Forward" configuration
then set the "Cal" knob full right. Flip the switch to "Reverse" and adjust the tuner for minimum as per usual. If you can get it all the way to zero - well zero is still zero. And less is still less. So what if you can't read that it is 1.25 to 1 if it is
1.25 to 1. A zero in "Reverse" is 1:1 no matter how high the TX didn't drive the needle in "Forward".
I built an SWR bridge that could be pinned with less than two watts (maxed at about 30) and discovered that it doesn't make any difference as described above. What does make a difference is a tuner and SWR bridge built for QRP power levels can be made
much smaller and lighter:) Put the tuner and the bridge in the same box together for even more convenience out in the field. Like a scaled down version of your MFJ. If you build it into your uBitx you will forfeit using it with other radios but will be even
more convenient when taking the uBitx to the field.
Keep us posted with progress.
73,
Bill ?KU8H
On 08/20/2018 11:31 AM, Mike KK7ER wrote:
So far I have been setting the antenna tuner using my Kenwood and then
swapping in the uBITX. ?But now we want to take the uBITX on the road.
Problem is that the uBITX does not put out enough power for full
deflection on our MFJ versa tuner II so we cannot get an accurate SWR
reading to tune the antenna. ?Does anyone have recommendations for a QRP
SWR meter? ?Thanks!
73, Mike KK7ER
(and son KI7ZZS)
--
bark less - wag more
|
Lee
SV9CVJ Nikos