The AM Forum
March 29, 2024, 08:41:23 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: 1 ... 10 [11] 12 13   Go Down
  Print  
Author Topic: The 6M-160M HPSDR Transceiver Project Begins  (Read 222716 times)
0 Members and 2 Guests are viewing this topic.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #250 on: February 12, 2011, 11:50:38 PM »


"Tom,
Do you know power SDR has a built in two tone generator? I plan to do some testing Sunday afternoon with the HP141 on the KAZ antenna."



Sure. Been using it for a while.  Download the Flex software manual and dig in. There are wonderous things this system will do that are not user obvious.  I still want to try that zero beat function Al talked about, assuming it works on HPSDR.

How about trying an IMD test with the Erb using the generator and Penny?

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #251 on: February 13, 2011, 09:21:33 AM »

Tom,
I was able to zero beat the TX by putting a 30 hz offset into the TX. I have not tried the zero beat function. This fixed the SSB offset you guys were telling me about. I need to build the reference board one of these days.
Yes, I plan to check IMD today with the 24 volt supply on the final with different bias settings.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #252 on: February 13, 2011, 11:25:41 AM »

Interesting on the ssb offset. I thought it was my rig or imagination, but I also heard the tiny offset between TX and RX. I wonder what that's all about in siftware?

Anyway, GL with the IMD test today. Too bad you don't have one of those 48V supplies for the finals.  Come on over and borrow mine if you need.

I'm still in the middle of building a new power supply for the TMC linear. I decided to build a new separate 1600V plate supply and the regulated LV supplies. Will take some pics of it later and post.  The original supply I built had so many changes it started to look JS. Just had to start with a new chassis.... Grin

T

Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #253 on: February 13, 2011, 12:33:28 PM »

Tom, I just finished some IMD  tests with Erb. At 30 watts out I was able to hit 40 dB IMD3 but it took a bit of doing. First I had to drop my sample rate back to 96K to reduce the unwanted sideband. Then I had to set up an EQ profile for maximum attenuation. When I looked at the two tone signal it looked very over modulated on P SDR display. The audio profile with max attenuation allowed me to reduce test signal levels to look cleaner on the display.
When I run the output up to 100 watts the IMD drops to about 32 dB with heavy bias. I have one more 48 volt supply but want to run around 40 volts and just need to finish the supply I started.
I think Erb can be improved and will get into it later when the new supply is up and running. I know the first spot to distorte is the gates of the final so maybe the driver output transformer needs help.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #254 on: February 13, 2011, 02:00:54 PM »

Tom,
I looked at penny stand alone and found IMD3 better than -60 dBc IP3 at 75 meters with reduced audio levels. I received a file from Joe K5SO to reduce audio gain at the DSP. I just sent you a copy. IMD3 was around -50 dBC IMD3 before I changed the two tone level.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #255 on: February 13, 2011, 02:58:09 PM »

Yep, -60db is outstanding running Penny barefoot. I can barely detect any side crud, even when it is +40 over on the RX.   With Penny driving the TMC, I am able to get the TMC (with extra NFB) within about 1-2 S-units of that when running at 40w.

I saw the same 2-tone overload too. It was just a matter of backing it way down.  I already sent you the same .bat file last week for line input to essentially reduce it.  The 00XX08, etc code that Phil just gave you on line is in the .bat file.  That's what that discussion was about last week.

Gee, I definately will have to bring my Erb over for comparison to yours. If you are seeing -40db at low power or -30 at higher power, then something must be amiss with mine. I even used -45 vcc on the driver and final. Though if the TMC works out, it will be cleaner than the Erb and run a coupla hundred watts out to boot  - or be a +50db 3rd 40w driver when I'm done wid it.

*** A little known secret is that the input to the PA board to most riceboxes is a VERY clean spot to tap off 10 mW or so. These low levels stages are usually very clean but get mucked up by the succeeding power amplifiers. My FT-1000D at this 10mW level was about equivalent to Penny at -60db 3rd.  Heck I shud drive the TMC with that and screw the  HPSDR... just kidding.. Grin

T

Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #256 on: February 13, 2011, 04:46:28 PM »

Very common information that a double balanced mixer and crystal filter will generate some very clean sSB. Usually stages up to about .1 watts are class A so there are no harmonics generated driving the fianl. Usually the driver and final run class AB where the distortion is created.
Your erb is exactly the same as mine except I have 4 bias pots and you have 2. I bet your two tone generator is also too hot. I checked Joe's code and Phil's code and they say the same thing for level. Phil's code just selects the line in. I suspect the internal test panel still runs the same.
I tried 08  15 , 08   14   changed to both lines 08   14 and as I suspected no gain reduction.
At 30 watts out I saw about -32 dB and at 100 watts about -25 dB until I set up an EQ file with the levels way back. Then it was about -40 dB and -30 dB.
At 192K samples I had a good spur in the unwanted sideband. As you saw on my signal unwanted sideband comes up when you hit the audio too hard.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #257 on: February 13, 2011, 05:36:11 PM »

The average newbie ham would not know about the clean low level tap. It's not printed in radio docs.  It's an EZ way to get a super clean exciter signal. That's why I mentioned it - not for sophisticated guys like you... Grin.

I was able to get a clean 2-tone by adjusting software, EQ level backed off, etc.  Also used SigJenny program from computer. Clean tone, same result on Erb.

Still hacking away on the TMC PS.

T

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #258 on: February 13, 2011, 06:07:59 PM »

Yup, I just found a two tone generator on this machine. Maybe better to try an external source. I'm thinking Erb needs some help at the driver output/ final input. Maybe those small cores need to be stacked. I need to study the schematic some more.
I bet you could do the TMC +/- LV supply like we did the audio driver. This is for the plus and minus 150 volt sources. TMC did a bit of JS to create the lower voltages.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #259 on: February 13, 2011, 06:42:13 PM »

Ya know, when I was testing Erb stage by stage on the scope, I did notice that the driver output was the first stage to look funky as the drive increased.  Look there first.

However, the IMD results were poor even when lightly driven at 30w out. All I wanted was a clean 30w at that point.

Will probably hook up the TMC tonight and try on the bench wid the new supplies.  I'm hoping it's even cleaner at 1600V/ 40w pep out than with 600v before..

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #260 on: February 13, 2011, 06:56:18 PM »

I guess it will depend on how low the tank Q goes when you raise the operating Z. Interesting how plus and minus 150 volts was used on the two driver stages. good luck!
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #261 on: February 14, 2011, 11:07:37 AM »

Looks like they fixed the spike unkey problem I reported. EXCELLENT! So it WAS in software and just simple T/R sequencing as suggested. The solution was basically a variable sequencer created in software. It will be nice to see a soft T/R break now. I won't be able to test it til my smoked RX board gets back... sigh.:

T



"Doug, W5WC, has found the problem in PowerSDR that was causing the unkeying spike.
What was happening was that RF was still being sent for some 10's of mS after the PTT had released.
Doug has added a variable delay so that you can select how many mS after PTT is released, and the RF stops, that you hold the PTT on.
Thanks for bringing this to our attention - hot switching a PA is never a good idea!"

Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #262 on: February 14, 2011, 12:01:33 PM »

Tom,
please send me a link to the file to my home email.
I'm not seeing the transient since I reduced the buffer size. I didn't listen for it with another receiver though.
I ran some numbers on the driver output transformer and find the inductance needs to be about 2 to 3 times higher at the drains of the driver. Also the balance transformer would also need more inductance. So maybe stacking cores will help the drain waveform to be cleaner at the final gates. I also saw the problem on the scope when I was playing with transformers but changing to type 43 didn't fix it. Need to study some more to find the best solution. The 4 output transformer cores can be easily stacked but the balance transformer is fairly tight layout. I'm thinking it may need to be a torroid to get the inductance higher and still fit. 
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #263 on: February 14, 2011, 12:29:24 PM »

Frank,

There is no available file fix yet. W5WC is supposed to announce the fix (for his own software version, I presume) later today. Then the various writers will have to incorporate the fix into their own software. So it depends what you are using and when it is fixed.

Good on the Erb. Since mine is sitting with blown finals, I will wait til you optimize yours and go from there. I may not even use Erb once the TMC is working, I dunno yet.

Gads, I'm running outa room to fit in this stuff. I was almost tempted to tear down my perfectly running 4X1 rig to make room, but snapped outa the stupor before I did any damage... :-)   The class E rig takes up some prime space too.  Some of the SDR guys have just a computer and an SDR box for the whole station. (and linear)

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #264 on: February 14, 2011, 12:35:08 PM »

There is a whole new world outside the shack double doors...
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #265 on: February 15, 2011, 12:37:08 PM »

Yep, Mo Room. That's what we all need.

This TMC linear amp is killing me. The 6CL6 predriver kept popping until I found I had +40 volts positive bias on the grid. I went thru four tubes b4 realizing it. There's a weird + - 150 supply configuration AND they borrowed some bias from the finals - that I didn't realize and built the PS incorrect.

Then I blew out a 90V SS regulator.  But little by little I'll get the bugs out and it will be a nice amp. Gotta stick with these problems sometimes even though they kick your ass around.  One day I'll come into the shack, flip the main HPSDR switch and everything will come up working FB... Grin

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #266 on: February 15, 2011, 04:45:55 PM »

I guess you didn't read my 2-13 post at 7 59
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #267 on: February 15, 2011, 04:53:16 PM »

Yep, I was aware it was +- 150V cuz I built the supply for it.  But the AH designers sneaked in extra bias from the finals that ran back to the predriver. This set the grid reference for the 6CL6 grid circuit. I shuda traced out the circuit instead of relying on the description that called it "PA bias" only.  Cost me four 6CL6's plus new PS mods to find out.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #268 on: February 15, 2011, 05:47:37 PM »

So you didn't look at figure 4.2
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #269 on: February 15, 2011, 06:14:07 PM »

Hindsight is 20-20 Frances... Grin

Of course I didn't read the manual - does anyone before they get in trouble?

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #270 on: February 15, 2011, 07:04:15 PM »

I think the Designers did a good job
Logged
Mike/W8BAC
Contributing
Member
*
Offline Offline

Posts: 1042



WWW
« Reply #271 on: February 15, 2011, 07:21:15 PM »

I have a request and a question. The request is can I see a simple translation list of the names of the boards and peripheral equipment for what they do regarding the HPSDR?

My question is will the single board transceiver that is in the works need the same back plane/power supply as the earlier component HPSDR TX system?

I'm still in SDR-1000 mode here. I am upgrading to an FA-66 sound card soon so the technology has raced ahead a bit. If I could have my wish I would want a single board HPSDR and a clean 100 watt PA in one package. Any hope of that? I know, that's two questions. Charge me extra.

Mike
Logged
Mike/W8BAC
Contributing
Member
*
Offline Offline

Posts: 1042



WWW
« Reply #272 on: February 15, 2011, 07:38:32 PM »

It might be helpful, for future reference, to insert a post near the beginning of this post with a reference table for acronyms. I know, it's imposable to insert a post but in this case, and for future long running posts like this, it might be something a moderator could grant to give a thread more continuity. By the way, What's an ERB?  Grin

Mike
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #273 on: February 15, 2011, 07:44:06 PM »

Hi Mike,

Look back near the beginning of this thread for links for info on what you axed about - in greater detail.  But I'll do my best here...

The new Hermes transceiver board will be using the Atlas backplane. The Atlas is more or less a universal set of sockets for power and data distribution.

The Hermes is still a 500mW output board, but the PennyWhistle is the 100w external linear. PennyWhistle is not as clean as the Penny exciter, only about -30db third IMD or so, so some of us are seeking out cleaner external amplifiers.

Penelope (Penny) is the exciter, Mercury is the receiver, Atlas is the backplane, Metis is the brand new communications board, though I still use Ozy, the current comm board used by most. Janus is an older sound card, not needed anymore.   Pennylane, the 100w linear amp.

Go thru the thread for more info.    The boards I mentioned above will probably be easy to find used and fully working, once the Hermes is avaialble.

Erb is the company name (ErbTec) of a 10mW > 150w MRI SS amp both Frank and I picked up on eBay a year ago.

Hope this helps and it wud be fun if you joined in, Mike.


Frank, yes, the TMC designers did a good job - just wish I found a factory-made  PS to go with it.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
Mike/W8BAC
Contributing
Member
*
Offline Offline

Posts: 1042



WWW
« Reply #274 on: February 15, 2011, 07:58:49 PM »

Will I need the communications boards (either Metis or Ozy) with this new Hermes transceiver? When I get home in a few weeks I'll look into the boards your using and the next generation in more detail. Thanks for the downshift.

Mike
Logged
Pages: 1 ... 10 [11] 12 13   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.101 seconds with 18 queries.