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: [1] 2 Next   Go Down
  Print  
Author Topic: Better TX IMD  (Read 9617 times)
ZENKI
Member

Posts: 906




Ignore
« on: October 27, 2011, 05:55:36 PM »

I wish Flexradio would design radios with better TX IMD. I just read the December review of the Flex 1500. The worst case TX IMD was  20 below PEP, -14db below one of two tones.   I wanted to buy the Flex1500 to drive a small 4cx350FJ tube which has -40db 3rd order IMD, however with such horrible IMD the only place that its fit for use is on the CB band to cause a lot of splatter and interference. Flexradio needs to place lets focus on the ears of the radio and place more focus of the mouth. All you gonna be hearing soon is nasty splatter and flex is contributing to the malaise!

Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #1 on: October 28, 2011, 03:49:48 AM »

I wish Flexradio would design radios with better TX IMD. I just read the December review of the Flex 1500. The worst case TX IMD was  20 below PEP, -14db below one of two tones.   I wanted to buy the Flex1500 to drive a small 4cx350FJ tube which has -40db 3rd order IMD, however with such horrible IMD the only place that its fit for use is on the CB band to cause a lot of splatter and interference. Flexradio needs to place lets focus on the ears of the radio and place more focus of the mouth. All you gonna be hearing soon is nasty splatter and flex is contributing to the malaise!



I agree.  When I owned my Flex-5000a the TX IMD was pretty bad and I know a guy who owns a Flex-3000 which is worse than the 5000.  Nice to hear the 1500 is even worse than the other two.  You'd think they would be getting better than worse.  All the focus on the receive side is useless if you fill the bands with TX crud.  Instead of using the TNF, just turn the radio off and find a new hobby at that point!

Gene
Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #2 on: October 28, 2011, 05:58:19 AM »

All the focus on the receive side is useless if you fill the bands with TX crud. 
Gene

Flexradio has never been too interested in TX xmit quality. One only has to look at the 2.4khz transmit spur debacle which took 3 years to fix. They only fixed it because of my constant public complaining.

Stan K9IUQ


I remember a few posts on the Flex radio list by the president of Flex down playing and trying to minimize the spur issue.

Whatever happened to the claim that the Flex-5000a's PA could be run in class-A mode?  Supposedly they had the provision to change the PA bias.  I remember they played that feature up when the 5000a first came out.  Then, like the promised Deep Impact, the class-A feature also never materialized.  When I inquired I was told that the cooling of the F5K's PA cannot handle class-A operation and that was not taken into consideration during the design of the F5K.   Another poor design decision by Flex made it not possible.  Now there is no reference to class-A operation in the F5K literature at all.

Gene
Logged
K0OD
Member

Posts: 2521




Ignore
« Reply #3 on: October 28, 2011, 07:33:38 AM »

Zenki said:   
Quote
I just read the December review of the Flex 1500.

Gotta link to this December review?
Logged
G4PNX
Member

Posts: 16




Ignore
« Reply #4 on: October 28, 2011, 08:51:38 AM »

Stan wrote:

One thing the Flexradio experience taught me: Promises are easy and cheap to make. Implementation is a completely different story.

But not necessarily without potential consequences. Check out Section 5 of the US Federal Trade Commission Act.

David - G4PNX
Logged
K0OD
Member

Posts: 2521




Ignore
« Reply #5 on: October 28, 2011, 09:50:29 AM »

OK, here's a link for League members. Click on "December" for the PDF review:
http://www.arrl.org/qst-product-review-and-short-takes-columns

Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #6 on: October 28, 2011, 12:02:39 PM »

Ah yes, the usual posts from the usual posters.

FWIW, I decided to look at the IMD output of my 1500. One of the nice things about PowerSDR is that it has a built-in two-tone generator. I used my Telewave wattmeter between the output of the 1500 and my Bird dummy load. I used the -40dB attenuation tap on the Telewave to feed the signal into a stepped attenuator and then into my Flex 5000 for analysis. I did my tests at 10m because 20MHz is the lower frequency cut-off for my Telewave.

As I expected, IMD is strongly affected by output power level. With the drive level set at 50% the output of the 2-tone signal showed up at 25% power. The Telewave showed power output to be 1.0W. I adjusted the output of the stepped attenuator so that each tone was at -67dBm. At this output the each of the 3rd order IMD products showed up at -105dBm, or about 37dB down. Increasing drive to 100% caused the power output to display as 50% but the actual power output rose to 2.5W. The 5000 showed the signal level per tone to be -63dBm. At this level the power in each of the 3rd order IMD products increased to -84dBm, only 21dB down.

If I turn the power down in small increments, IMD drops rapidly. Turning the power output down by 10dB (250mW out) and the each of the 3rd order IMD products is now down 45dB. If I turn the power down 3 more dB the 3rd order IMD products disappear into the noise at -130dBm. That is 54dB. (BTW, I changed the attenuator to bring all the levels up by 10dB so I could clearly see the 3rd order products again and that 54dB number is correct.)

But that also shows how much better things are when operating class-A. I am sure that the devices in the 1500's PA are not cut off at any point in the cycle at those levels. So I don't see anything all that unusual about the 1500's IMD compared to other ham transceivers. OTOH, like Zenki, I would like to see the overall crop of transceivers be a lot better than they are. But this does show that, if you want a clean signal, turn the power down.

But this thread on eHam shows one other thing: signal to noise ratio on the eHam list is also strongly affected by bias levels. If you want the truth it pays to do some homework on your own. ;-)

73 de Brian, WB6RQN/J79BPL
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #7 on: October 28, 2011, 12:04:08 PM »

BTW, that QST review didn't seem all that negative to me. I thought it was pretty straight forward.

73 de Brian, WB6RQN/J79BPL
Logged
W4HIJ
Member

Posts: 367




Ignore
« Reply #8 on: October 28, 2011, 01:34:46 PM »

I wish Flexradio would design radios with better TX IMD. I just read the December review of the Flex 1500. The worst case TX IMD was  20 below PEP, -14db below one of two tones.   I wanted to buy the Flex1500 to drive a small 4cx350FJ tube which has -40db 3rd order IMD, however with such horrible IMD the only place that its fit for use is on the CB band to cause a lot of splatter and interference. Flexradio needs to place lets focus on the ears of the radio and place more focus of the mouth. All you gonna be hearing soon is nasty splatter and flex is contributing to the malaise!


Not trying to argue the point here Zenki but wasn't it your contention in a discussion with me not that long ago that TX IMD performance on ALL ham transceivers was bad, in your opinion?
Logged
W4HIJ
Member

Posts: 367




Ignore
« Reply #9 on: October 28, 2011, 01:58:22 PM »

BTW, that QST review didn't seem all that negative to me. I thought it was pretty straight forward.

73 de Brian, WB6RQN/J79BPL

Well Brian, you know that if you go LOOKING for negatives you'll certainly find them.  I did notice some negative issues pointed out but I also noticed a bunch of positive attributes mentioned as well. Of course, you won't find any of the positives listed here by "the usual suspects". Myself, I have never owned a radio yet that didn't have both positive and negative attributes about it.  Every radio I've ever owned, including my Flex 5K and 1500 has had things I've disliked about them.  Usually though, on the rigs I  decide to keep for long periods of time, the positives far outweigh the negatives. Such is the case with my 1500.
73,
Michael, W4HIJ
« Last Edit: October 28, 2011, 02:02:21 PM by W4HIJ » Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #10 on: October 28, 2011, 02:29:53 PM »

BTW, that QST review didn't seem all that negative to me.
73 de Brian,

If you want the truth it pays to do some homework on your own. ;-)

Given a choice between getting the truth from the Flexradio Ambassador or QST magazine I will take my chances with QST.  Cheesy Cheesy

That is fine. They give you one data point. I gave several, complete with actual measurements. Full-power IMD measurements are always higher. But it does not surprise me that you missed the point of my message.

Quote
Ambassador Brian says Not Negative ? Lets see here:
From the QST Reviewer
CW Performance and latency so bad the reviewer resorted to an external non-flex keyer -Sound Familar?? Cheesy Cheesy

No, that is not what he said. What he said was this:

"CW performance using the internal keyer remains a problem because there is a significant delay (see Figure 1) between the fingers and what is sent over the air and as a sidetone. At any speed over 20 WPM, the delay made sending difficult for me even after optimizing the USB interface control settings. This is due to USB interface issues not under the control of PowerSDR as noted in the section below on PC Dependencies. I was able to use an external keyer with the FLEX-1500 sidetone turned off."

This does not say that CW performance is bad, only that the latency in the sidetone at speeds over 20WPM is annoying so he used an external keyer. So, I think you are stretching a bit Stan.

Quote
CW Break-in causes clicks in audio and sudden display changes.

CW Transmit SPURS These spurs are easily seen and noted in the Spectral Display and review text. Sound Familar??  Cheesy Cheesy

Yes, there appears to be something at +/- 1kHz. One side is -41dB and the other is -50dB. Are they keying sidebands or real spurs? That is not clear.

Quote
The Flexradio 1500 that QST purchased stopped working and had to be sent back to Flexradio for repairs.

Here is the exact quote:

"During initial testing in the Fall of 2010, our FLEX-1500 transmitter stopped working on 6 meters and we returned it to the factory for service. After repair (a bandpass filter had failed), testing revealed several issues including the CW keying latency discussed above, sidetone glitches, mediocre worst case transmit IMD on 40 meters and poor unwanted sideband rejection on transmit."

(Yup, sounds bad. But if you keep reading in the next paragraph ...)

"Subsequent versions of PowerSDR improved operation, and FlexRadio provided an alignment generator to adjust unwanted sideband suppression. The alignment procedure took about 5 minutes and increased the unwanted sideband suppression from 44 dB to 53 dB and the carrier suppression from 63 dB to 65 dB. All radios shipped in 2011 have had this alignment performed at the factory, and FlexRadio will provide an alignment generator on loan to those who have a FLEX-1500 shipped in 2010. Transmit IMD on 40 meters was still not great, but typical performance on the other bands is good."

Huh. The 1500 broke and Flex fixed it. I have certainly had that happen with products from other manufacturers. Also, Flex provides a calibration generator to anyone who wants it to correct the poor carrier and unwanted sideband suppression on older radios. Current production radios don't have this problem. Also, even though IMD performance on 40m was a problem, he went on to add, "...but typical performance on the other bands is good." Huh. He used the word 'good' to describe the performance on other bands. So, when you read the whole thing, it doesn't sound nearly as bad to me as you make it out to be.

Quote
Mediocre worst case transmit IMD on 40 meters and poor unwanted sideband rejection on transmit.

Which, if you read the actual quote above, was solved.

Quote
Contest Operators will find the PSDR interface too busy and disjointed.
Yep, not negative says the Flexer.. Wink

That suggests a personal preference. Don't get me wrong, I think that the UI in PowerSDR is not very good. And I agree with the "too busy and disjointed" comment. But when it comes to UIs, one man's meat is another man's poison.

Quote
Now Please before all you Flex Lovers get ready to denounce me and call me names, remember I did NOT write this review. QST did. Don't disrespect the messenger just because you dislike QST's message...

Oh, I don't have any problem with the QST review at all. If you read the whole thing it seems, over all, pretty balanced, and not unduly negative. Sure you can cherry-pick the negative stuff and blow it out of proportion, something you do at almost every turn, but that doesn't make the review negative. In fact, I think it is spot on. In other words, I think the QST review is quite good.

Here is more of the review quoted:

"Whether the FLEX-1500 system will be right for you depends to some degree on your radio weltanschauung or view of the world. Is the system a platform for experimentation or a communications tool? The system’s basic performance is good — it’s well instrumented and very flexible. For use as a spectrum analysis tool, it makes a very nice piece of lab equipment! As a techie, I loved the multiple displays, meters and controls."

Gosh, that sounded downright positive. "The system's basic performance is good -- it's well instrumented and very flexible," sounds like downright praise. Continuing:

"They were so cool, it was almost more fun to tune around the band than to make contacts — and there’s the rub.

A competitive operator will find the user interface too busy and the many controls simply too disjointed to operate without having to think about them, disrupting concentration. For example, using the CLICK-TUNE feature moves the spectrum across the display to center the filter in a fixed location. It’s fast but all of the waterfall information has to be rebuilt. Does the operator move across the spectrum or vice versa? These and similar questions never had to be answered before SDR became a reality."

In that I am 100% in agreement. It drives me crazy that click tuning moves the spectrum and ensures that the "memory" held in the waterfall is lost. I have been bugging Flex about that for a couple of years. So, yeah, there are bad things in the flex radios and PowerSDR. OTOH, there are good things too. How about we look at what they wrote under Bottom Line:

"The FLEX-1500 provides excellent performance at an entry level price, assuming you have a satisfactory PC to host it. The Flexers, an active community of SDR enthusiasts, will become your best friends.

If you want a portable or mobile radio or don’t like to use computer interface devices, I’d recommend staying with a traditional standalone rig. However, if you want to experiment and jump into SDR the FLEX-1500 makes for an easy route."

Huh. The first words out of the reviewer's mouth for the Bottom Line were, "The Flex-1500 provides excellent performance at an entry level price."

Let's see, "... excellent performance..." Huh. I guess we derive different meanings because I normally associate the word 'excellent' with 'really good'.

Quote
LMAO
Stan K9IUQ

Stan, I think we would all love to see you ass-less. Alas, that is unlikely to ever happen.

73 de Brian, WB6RQN/J79BPL
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #11 on: October 28, 2011, 04:56:33 PM »

Well Stan, you did say, "Laughing My Ass Off" (LMAO). I was just suggesting it would be difficult if not impossible for you to lose your ass. :-)

As for quoting copyrighted material, I did that, not you. It is my problem, not yours. Why don't you let me worry about it, OK? BTW, I recommend you check out something called "fair use". The quotes were a subset of the work and QST was clearly attributed.

And I wouldn't expect you to include the supporting material that would make things perfectly clear. Including complete and balanced source material for your claims is not historically your way of doing things.

But I do think it served the purpose to make things more clear.

73 de Brian, WB6RQN/J79BPL
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #12 on: October 28, 2011, 05:35:59 PM »

Thank you for your concern for my welfare in this matter, Stan. I really appreciate it. But I think it will be OK. If the ARRL, eHam, or Flex are concerned, I am certain they won't come after you.

73 de Brian, WB6RQN/J79BPL
Logged
VA6LM
Member

Posts: 1




Ignore
« Reply #13 on: November 05, 2011, 09:20:45 PM »

Stan K9IUQ,
I heard that you want to sell your Flex 5000A.
I would like to buy one... seriously pm me and
lets talk. I really like the notch video by the way.
Logged
W6UV
Member

Posts: 536




Ignore
« Reply #14 on: November 20, 2011, 09:31:38 PM »

The UI needs a little work related to the addition of the tracking NFs.

If you have several filters defined grouped together (like I do to get rid of groups of carriers generated by some unknown equipment in my neighborhood), then click-tuning is effectively useless in that area as clicking to tune activates the TNF edit function rather than tuning the receiver to that spot.
Logged
Pages: [1] 2 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!