Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 2 [3] 4 5 6 7 Next   Go Down
  Print  
Author Topic: Flex-5000A CW Impressions  (Read 30862 times)
K0OD
Member

Posts: 2533




Ignore
« Reply #30 on: February 21, 2011, 08:54:18 PM »

Brian, do you think it's possible that the keying quality might vary with band activity or proximity to very loud signals?

I noticed some odd squeaky background sounds in my headphones while keying yesterday on 80 meters. Just after 0000z those sounds stopped, along with dozens of S-9+ contest stations.

--
I just now did a full removal of all versions of Flex software on my machine,  some going back 18 months.  I used the Add/Remove Programs section in the Control Panel. Reinstalled 2.0.19, and built a fresh database for about the fifth time in three days. So far CW keying seems good on all bands except six meters which has always been a problem for me.

Brian, have you tested your keying on six?

 
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #31 on: February 21, 2011, 10:36:03 PM »

Brian, do you think it's possible that the keying quality might vary with band activity or proximity to very loud signals?

I don't see how. The receiver and transmitter are separate. I don't see how a receive signal can affect the TX.

Quote
I noticed some odd squeaky background sounds in my headphones while keying yesterday on 80 meters. Just after 0000z those sounds stopped, along with dozens of S-9+ contest stations.

The noises I CAN imagine. SDR always introduces a delay (latency). Literally the sound you hear now was on the air some number of milliseconds ago. Same with the transmitter. The signal you are generating now goes out the antenna some number of milliseconds later. If you are running QSK it is possible to get overlap between RX and TX and actually hear some of your own transmitted signal. PowerSDR tries to prevent this by blanking the receiver for some time after the TX unkeys so you don't hear the tail end of your own signal. But you want that to happen for as short a period as possible or it starts to eat into your QSK time. I am betting that by reducing your DSP buffers one notch you might eliminate that.

But I am guessing here. I haven't tried it. You might want to try different buffer settings while keying into a dummy load and see if that changes the characteristic of the sound or maybe eliminates it entirely. Try changing the driver buffers (buffer setting in Setup>Audio>Primary) too. That changes how long it takes samples to get from radio to computer. Smaller buffers are better for latency but if you get them too small the audio goes to pot. Try different values.

Quote
--
I just now did a full removal of all versions of Flex software on my machine,  some going back 18 months.  I used the Add/Remove Programs section in the Control Panel. Reinstalled 2.0.19, and built a fresh database for about the fifth time in three days. So far CW keying seems good on all bands except six meters which has always been a problem for me.

Brian, have you tested your keying on six?

No, I haven't tried 6m. I did my testing on 10m.

When I hear of TX problems that are band-related, I immediately think of RF-in-the-shack. Do you have a good dummy load? If so, try that. If the problem goes away, suspect RF in the shack. Make sure that your radio and computer are close together and there is a good, short ground wire between them.

I don't think this is a software problem. If it were, you would see the same problem on all bands. Since it only occurs on one band, I think this is good old RF.

73 de Brian, WB6RQN/J79BPL
Logged
K0OD
Member

Posts: 2533




Ignore
« Reply #32 on: February 22, 2011, 12:12:14 AM »

"The receiver and transmitter are separate."

But they share the same computer. There would be some millisecs of overlap during change-over. Wouldn't CPU usage increase with many loud signals being received, as compared with a dead band?

--
I've done many tests and fiddled endlessly with buffer and sample settings.  Yes, I ran tests Saturday with a dummy load connected directly to the Flex to assure that RF wasn't causing problems.

Anyway, things may be looking up after I deleted those many old Flex files a few hours ago. Brief tests on all bands sounded very good. Six meters CW sounds slightly "thumpy" in my headphones, but not bad. My CW is well formed with 2.0.19; I had been bothered mostly by minor clicks on change over to transmit. The more annoying squelch tails and dropped dits I experienced a few months ago (with 2.0.16) are gone.

Try six meters CW and report back. Others experienced keying problems there last summer.

Logged
KF6QEX
Member

Posts: 590




Ignore
« Reply #33 on: February 22, 2011, 12:59:36 AM »

Shameless temporary thread hijacking to clarify a couple of things:

Quote
19     eHam Forums / Software Defined Radio / RE: Public Apology From SM5BSZ - Flexradio NOT Scamming Anyone    on: February 20, 2011, 09:28:15 PM
 Shocked Wow!

I just learned that there was an SDR forum on eHam and came over here to look. Lots of people flaming Flex.

Quote
Position:    Member
Date Registered:    June 16, 2009, 02:12:38 PM

It's funny that:
Quote
Quote
He has been telling us for a year there was nothing wrong with cw on a 5K.

Can't help you there. He is not me. I am not him. Why don't you deal with me and stop worrying about him.


Ok..then lets do just that.
You have been a registered registered user for over a year and a half hear and year you claim you  "just learned".
 
You have access to early betas, but yet you claim no ringing of bells of your predecessor's call sign, however prominently it is (or used to be) featured on the flex web page. 

It wouldn't surprise me to find out you are a "plant", the new "friendlier" voice of flex on eham especially now that PSDR "just works".

Your motives may be sincere. But the sudden sprouting of your posts is most interesting.
Logged
N3EVL
Member

Posts: 40




Ignore
« Reply #34 on: February 22, 2011, 05:53:14 AM »

First Brian's a "Fanboy," now he's a "plant"?

You people have just broken your own silliness record by another 10dB.  Truly amazing.
Logged
W6UV
Member

Posts: 536




Ignore
« Reply #35 on: February 22, 2011, 07:05:21 AM »

Brian,

Was that you working as a volunteer in Flex's Pacificon booth last October? I seem to remember someone there who looked like your picture on QRZ.com.
Logged
K0OD
Member

Posts: 2533




Ignore
« Reply #36 on: February 22, 2011, 07:28:10 AM »

Quote
"You people have just broken your own silliness record by another 10dB.  Truly amazing."

Agreed, but I find it hard to believe Brian isn't familiar with W9OY. For the past two years, the Flex Radio site has had a prominent link to 'OY's Flex blog. I've always found Brian's posts to be informative especially since he's been active in frequency measuring tests (where Flex is the best radio on the planet under $100,000).

I'd love for him to stick around.


Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #37 on: February 22, 2011, 08:18:09 AM »

"The receiver and transmitter are separate."

But they share the same computer. There would be some millisecs of overlap during change-over. Wouldn't CPU usage increase with many loud signals being received, as compared with a dead band?

Not really. The code must go through the same loops to perform the filtering and detection. To the computer it doesn't matter if it is adding zero over and over or adding 1000 over and over. It is still going through the entire addition process.

Now understand this this is a gross oversimplification but hopefully you will get the idea.

--
Quote
I've done many tests and fiddled endlessly with buffer and sample settings.  Yes, I ran tests Saturday with a dummy load connected directly to the Flex to assure that RF wasn't causing problems.

Anyway, things may be looking up after I deleted those many old Flex files a few hours ago. Brief tests on all bands sounded very good. Six meters CW sounds slightly "thumpy" in my headphones, but not bad. My CW is well formed with 2.0.19; I had been bothered mostly by minor clicks on change over to transmit. The more annoying squelch tails and dropped dits I experienced a few months ago (with 2.0.16) are gone.

That does seem to be the case with 2.0.19.

Quote
Try six meters CW and report back. Others experienced keying problems there last summer.

I will try it. But at this point I don't think reports of CW problems from previous versions are going to be significant since the new code is SO different (I have been told). The problems that were there probably aren't there now. Now there is a whole new set of problems! Smiley (We hope not.)

Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #38 on: February 22, 2011, 08:29:39 AM »

Quote
I just learned that there was an SDR forum on eHam and came over here to look. Lots of people flaming Flex.

Quote
Position:    Member
Date Registered:    June 16, 2009, 02:12:38 PM

It's funny that:
Quote
He has been telling us for a year there was nothing wrong with cw on a 5K.

Can't help you there. He is not me. I am not him. Why don't you deal with me and stop worrying about him.


Ok..then lets do just that.
You have been a registered registered user for over a year and a half hear and year you claim you  "just learned".

Yup. I don't recall why I signed up for eHam back in 2009. I wanted to get access to something or to ask a question on one of the forums. But I certainly didn't make a habit of checking in here. What I "just learned" was that there was an SDR forum. So I looked. I probably would have gone right on by but I was pretty amazed at what I saw here. Of course, I couldn't just keep my mouth shut, right? Wink

 
Quote
You have access to early betas, but yet you claim no ringing of bells of your predecessor's call sign, however prominently it is (or used to be) featured on the flex web page.

Nope. Other than helping people who ask questions and trying to pull information together, there aren't too many people on the flex lists, beta or otherwise, that I directly communicate with on a regular basis. Neal Campbell is probably the guy I talk with most. He and I have become friends, mostly from working on the FlexRadioWiki together and figuring out how to tweak PCs to make them work better with PowerSDR. I lean toward Macs myself.

As I made clear, CW is not my preferred mode of operation. I like new things. I like seeing how much data I can send with the least amount of energy. That leaves out CW, RTTY, AM and SSB. So if he was oriented to CW, I didn't pay much attention to what he was doing.

As to what is on the Flex web site, I go there to get access to the download section and to the knowledge base. I couldn't even tell you what is on the front page.

Quote
It wouldn't surprise me to find out you are a "plant", the new "friendlier" voice of flex on eham especially now that PSDR "just works".

Your motives may be sincere. But the sudden sprouting of your posts is most interesting.

But you caught me! BWAHAHAHAHA, Actually I am Flex Secret Agent 000000111 (convert from binary to octal). After shooting JFK from the knoll I was dropped in here from a black helicopter flown by asian men wearing blue helmets. BWAHAHAHAHAHA.

(Sorry. It is hard to type when I am laughing.)

All kidding aside. Try this: deal with me directly based on what you know about me and what I say. "Sometimes a cigar is just a cigar." I truly do not know what W9OY said or did here. Believe it or not, I haven't looked at his web page.

I really do want to help Jerry if he is having a problem. I do this because I think that the Flex Radios are the most interesting amateur transceivers in production today. They have a lot of warts but it seems to me they also have the greatest potential. You can't make the current crop of knob radios any different than they already are. What you see is what you get. Not so with the Flex radios. That is why I like them. Let's see you do CW skimmer with your FT101. (Yes, you can add a SoftRock to a K3 but then, you basically have a Flex radio for your IF.)

73 de Brian, WB6RQN/J79BPL
« Last Edit: February 22, 2011, 09:21:46 AM by WB6RQN » Logged
K0OD
Member

Posts: 2533




Ignore
« Reply #39 on: February 22, 2011, 09:19:32 AM »

Quote
Now there is a whole new set of problems! Smiley (We hope not.)

Unquestionably that will be the case. I've done some reading of K3 groups and there's some real bashing by owners of that wonderful radio too. A major reason I opted for the Flex is its lab quality S-meter, a first for our hobby and a half-century overdue. I see lots of complaints about the K3's puny meter.

With Flex CW cleaned up so that high speed ops can push the limits, expect to see more discussion of relay chatter. Also I'd sure like to see less erratic performance from the SWR meter, which the reflector touched on yesterday. Both are minor problems and fixable down the road.

The real excitement will occur when Flex begins adding new features in future releases.   

 
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #40 on: February 22, 2011, 10:45:19 AM »

Brian,

Was that you working as a volunteer in Flex's Pacificon booth last October? I seem to remember someone there who looked like your picture on QRZ.com.

Yes. I worked in the Flex booth at Pacificon. Greg Jurrens said he needed help and I live in the area. I thought it would be fun to "eyeball QSO" some of the people I have met on the lists. Apparently I met you there too. I thought  recognized your picture too when I looked you up on QRZ.com.

I make no apology for liking Flex radios. As I said, I don't know of any other production radio that has the potential of the Flex radios. And I would emphasize that I was a volunteer in the booth. I do not work for Flex Radio Systems.

If I think Flex has done something wrong or needs to change something, I have no problem saying that. Any company that lies to itself or its customers about its products deserves to go out of business. The way a company gets better products is to learn, change, and improve. I'd like to see my Flex 5000 reach its potential. It certainly hasn't done that yet but 2.0.19 makes me really hopeful that it will.

And I can understand the Fanbois. Mostly they see the potential of the product and get all excited. They want everyone else to be excited too. And Fanbois tend to say things like, "Well if you don't see how good it is, you must be stupid." The problem is, with something like PowerSDR, there are a lot of warts and bugs in between what is and what could be. There is nothing wrong with saying, "Gee, CW sucks on this radio. It certainly doesn't meet my expectations," because it was the truth. (I think that 2.0.19 may have changed that. You decide.) OTOH, both sides need to back off and recognize that there is good truth too. From just a pure receiver prospective, you know, clean signals, good filtering, resistance to overload, etc., the Flex 5000 is just breathtakingly good. So I take the good, I get the bad, and I bug Flex to fix the bad.

Right now I am sitting in the shack at the school where I teach. Sitting on top of the Flex 3000 is the K2 that was donated to our school by the ARRL. I am really proud of my students who built the K2. (Wayne Burdick joked at Pacificon a couple years ago that we hold the record for the most people taking the longest time to build a K2. But when you consider that many of the builders were 5th graders you can understand how I could think that 20 good solder joints might be a good day's work.) We have used it for several years now. It has been the mainstay of our Field Day operation for three years. It is truly one of the all-time great radios. It is perhaps the best QRP CW rig ever made. (OK, the QRP K3 is probably better. Still, I *really* like the K2.)

But here is the thing. That K2 is never going to change. It is not going to get more features without bolting on some new hardware. It is what it is. And if all you want to do is QRP CW, it will do that job extremely well. But now contrast that with the Flex 3000 sitting underneath it. Even though the K2 *was* a better CW rig, the Flex 3000 already was a better digital mode and SSB rig. Now, through a new release of software, the Flex 3000 rivals the K2 as a CW rig. (Maybe the 3000 now rivals the K3. I don't know yet.)

And I know that the Flex has a bad rep for RF-in-the-shack problems. But it isn't alone in that. I had no END to RF-in-the-shack problems with the K2 running digital modes. RF killed our computers. The keyboard would stop working, or the mouse would stop working, the serial port would stop working, or the computer would crash. I had snap-on ferrite cores on all the cables between K2, Rigblaster, and computer. I had snap-on ferrite cores on mouse and keyboard cables. Those problems actually went away with the Flex 3000. Now the kids really like running PSK31. For them it is far and away their favorite mode. Maybe that is because it is a lot like texting on their cell phones, who knows. But at least I am getting them on the air and they really like the Flex 3000.

I can also do things in class with the Flex 3000 I can't with other radios. When I teach about amplitude modulation I can show them the upper sideband, lower sideband, and carrier on the panadaptor. I can filter out the unwanted carrier and sideband (razor sharp filters) so they can hear how much louder the sideband by itself is (because the carrier is reducing the gain of the receiver through the AGC). Then I can show them the difference in power output when I switch the transmitter to SSB. They can SEE all this on the screen of PowerSDR. I can't DO that with the K2.

So, I guess you *can* label me a fanboi. I am because I don't know any other production radio that will let me do what I want to do. But I am not a slave to any party or company line. What matters to me is that it does what it is supposed to do or that it is moving in that direction. But I do prefer to focus on how to get to the potential rather than focusing on how bad it is perceived to be right now. After all, it is just an engineering problem to be solved.

OK, I got carried away. Go ahead, beat me up now. Wink

73 de Brian, WB6RQN/J79BPL
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #41 on: February 22, 2011, 02:31:41 PM »

OK, I got carried away.

Indeed you have posted some of the  l  o  n  g  e  s  t   posts I have ever seen on eham. Lots of them too.

I hope you didn't have any problem reading them. Wink "If you can read this, thank a teacher," and all that. Wink

Regardless, I write pretty fast. My dad made me take typing in high school. It turns out to have been the most useful class I ever took.

Quote
You got way too much time on your hands Brian, I think I understand now why state governments are paring down the teacher force..   Wink

Quite the gadfly, aren't you Stan. Wink

It is fortunate that I teach at a private school. Normally Tuesday is one of my days off but I am in this week because our Science Fair is this week. I had some time this morning to answer questions. It only took me all of about 30 minutes to write those posts.

What would ever give you the idea that the public school system would allow a teacher to make ham radio part of the required curriculum? No, I'm glad I am not part of the public school system. I would go crazy in a minute.

Quote
I gotta an excuse, I am retired. No job, no Boss, no worries, no mortgage, no paycheck and plenty of time to play Flexradio...........

I used to be retired. Then I discovered I was merely unemployed. Grin

73 de Brian, WB6RQN/J79BPL
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #42 on: February 22, 2011, 03:54:57 PM »

Quote
Now there is a whole new set of problems! Smiley (We hope not.)

Unquestionably that will be the case. I've done some reading of K3 groups and there's some real bashing by owners of that wonderful radio too. A major reason I opted for the Flex is its lab quality S-meter, a first for our hobby and a half-century overdue. I see lots of complaints about the K3's puny meter.

Actually, there is an interesting story that ties Flex, that meter, and kids learning ham radio in school together.

Two years ago I had two students who wanted to do a joint science fair project. They wanted to build a back-pack HF station they could take hiking. Since I had had problems the year before with unequal distribution of work on joint projects I told them they had to split the project into two pieces. One boy, Josh, build the backpack, solar-panel mount, charging system, and mounted the school's K2. The other boy, Michael, undertook to analyze different wire antennas to figure out what would be best to carry into the field. He decided to rate the 1/2 wave dipole, full-wave delta loop, wire 1/4-wave ground-plane with elevated radials, and a non-resonant wire with counterpoise and tuner.

Michael compared the antennas for size, bulk, weight, ease of construction, and ease of erection. The problem came when he wanted to compare performance. He started out by building models at 2m and testing them in a home-made antenna range. That didn't work out too well so he was looking for something better. He and I were brainstorming and came up with the idea of using my new Flex 5000 with the two identical receivers and calibrated meters to compare signals. He chose to use PSK31 signals from all over since the signals are narrow and constant power.

Michael got his classmates to help him set up an antenna farm on our playground. I brought in my F5K and he set up to collect data. We spent most of a day tuning in PSK31 signals and comparing the signal strengths from the different antennas. He would take literally thousands of readings. For each station he would collect as many pairs of signal readings as he could and then averaged all the readings for a given station on a given antenna and compared that with one other antenna. (The dipole became the reference antenna.)

In the end we were quite amazed to find the results. By his calculation the delta loop had a 2.3dB advantage over the dipole. The dipole and ground plane were even up. He did notice the diversity effect where a fade on the dipole would be a peak on the ground plane and vice versa. The end-fed random wire came in dead last by almost 10dB.

In the end when he weighted his results, including the performance measurements, the 1/2-wave dipole came out on top.

How did they fare in the Science Fair? Michael took 1st place in our regional fair for Engineering. Joshua took 2nd place. At the California State Science Fair Michael took 2nd place for Electronics and Electromagnetics.

So, the Flex 5000 with its calibrated RX1 and RX2 ended up being used a a piece of test equipment and a boy learned a lot about antennas as a result.

73 de Brian, WB6RQN/J79BPL.

P.S. I have been trying to get Michael to write that up as an article for QST for two years. So far he has refused. <sigh>
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #43 on: February 22, 2011, 09:56:58 PM »

Great story Stan. Ham radio still finds its way into the occasional school. The ham station usually ends up happening because one teacher keeps it going. When that teacher goes, the club usually ends.

The difference here is that "Wireless Technology" in our school is an actual class. It is a *required* class for all middle-school students. All students work toward at least a technician-class license. That is what I was saying will never happen in a public school.

73 de Brian, WB6RQN/J79BPL


Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #44 on: February 23, 2011, 10:57:14 AM »


With Flex CW cleaned up so that high speed ops can push the limits, expect to see more discussion of relay chatter. Also I'd sure like to see less erratic performance from the SWR meter, which the reflector touched on yesterday. Both are minor problems and fixable down the road.

I have been thinking about both of these things. WRT the SWR reading, I think they use the same A:D converter for both forward and reflected power. That means they cannot sample them simultaneously. The end result is that the SWR reading is only valid if both forward and reflected power are constant and steady, i.e. when sending a steady carrier (like TUNE). It won't indicate properly when the signal is being amplitude modulated (SSB, AM, CW during keying, PSK31, etc.) or when the tuner is in the midst of tuning. This would require a hardware change, either add another A:D or add a sample-and-hold to both signals, to make the SWR reading work when the signal is changing.

But since I know that it is not accurate when the signal is changing, I would like the option to have SWR displayed anyway.

As for relay chatter, I was thinking about the characteristics of a relay vs. PIN diode switching for rapid turn-around, e.g. QSK. Relays are noisy but they won't degrade receiver performance in the presence of a really strong signal. A PIN diode can. I think that I would opt for a relay in that case and live with the noise. Of course, if Flex can find a quieter relay ...

73 de Brian, WB6RQN/J79BPL
Logged
Pages: Prev 1 2 [3] 4 5 6 7 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!