Topics

#80m Turning the 80 meter winding of T1 from hell to heaven #80m


Steven Dick
 

In a prior discussion of feasibility of a 160 meter QCX+, it dawned on me that for the 80 meter QCX+,  the large number of turns on T1 (68 turns on secondary 3) can be greatly reduced by using a type FT50-61 toroid instead of the T50-2 toroid.  On the FT50-61 toroid, this would require only 18 turns on secondary 3, and 2 turns for the PrI, S1, and S2 windings.  Type 61 material is a relatively low loss ferrite, while -2 material is a powered iron.  It's exactly the same size, and it's Q will be comparable to the -2 powdered iron core.  Here's how I arrived at this:

1. Use toroids.info to compute the inductance of secondary 3 for a T50-2 toroid with 68 turns.  This results in an inductance of 22.66uH.  Now compute the number of turns required for 22.66uH for an FT50-61 toroid.  It is 18.1 turns - very close to 18 turns.

2. Lets look at required turns ratio for proper impedance matching between secondary 3 and the other windings.  For 80M, 60M, 40M, 30M, 20M, 17M, turns ratio varies from 13.6 to 7.3. If we use 2 turns for the other 3 windings, turns ratio is 18/2 or 9, in the middle of the range. It is 13.6 for the current 80 meter design using the T50-2 toroid. 3 turns would be too low a ratio.
3.  Let's make sure the Q of the secondary 3 tuned circuit is adequate.  For the T50-2, I estimate that Q is around 200 or less, looking at some curves in the following micrometals document. Go on micrometals.com (required an account to be set up to get this document), then design and applications, literature, Micrometals Iron Power Core Q Curve catalog and download.  Page 15 of the document shows a T50-2 with 79 turns having a Q of 184 at 3.5 MHz.  That will be in the ballpark.  For the FT50-61, go to owenduffy.net/software and select the "Ferrite Permeability Interpolations" tool. Select Fair Rite 61 for the material type and hit "Submit". At 3.5MHz, it shows a u''/u' ratio of 1490 at 3.5 MHz.  That can be considered a measure of the Q of the material itself at 3.5 MHz.  That's a very high number and indicates to me Q is not an issue for this material for 80 meter operation. 

Type 61 ferrite toroids are commonly used in crystal radios where toroids are called for  in the design because of its relatively high Q inductors used in the AM broadcast band. It is also relatively low cost  - about twice the cost of a T50-2. Not too bad.

-Steve K1RF


@CurtisM
 

Steve

Interesting analysis,  and I appreciate your technical writing and presentation efforts.

Its worth an experiment. I do wonder about those 2 turn windings and whether they will be as uniform as the current design. I survived the 68 turns, even doing it twice on my 80m qcx+. I found 3 colors of wire for the 5 turn windings helped me greatly. If you try your recipe, report back and I will be glad to concede my thoughts.

I don't think many of us mind the designer's frugality,  it is quite off the charts good. This 80m qcx+ is an amazing nice build and operation, I didn't even mind the jumper wires for the PA TR cable. I wish Hans well in adding to amazing delivery of thousands more qcx rigs.

73 curt


Steven Dick
 

The 2 turn windings should not have any issue but of course the proof is in the building and trying.  A 2 turn winding on an FT50-61 (0.28uH) has about twice the inductance of 5 turns winding on a T50-2 (0.12uH) which is a good thing for these windings for some technical reasons. Here you are more concerned about impedance matching which is based on turns ratios.  Squeezing or spreading will vary their inductance but that's not important for these windings since their turns ratio is constant. 2 turn windings are typically used in up to several hundred watt level transformers for end-fed half wave antennas.  I also should have mentioned in my previous discussion that the Q that's important is the loaded Q.  The type 61 toroid will actually have a higher unloaded Q for the resonant circuit  than the T50-2 at 3.5 MHz but the loaded Q's for the two will be comparable.

-Steve K1RF

------ Original Message ------
From: "wb8yyy via groups.io" <wb8yyy@...>
Sent: 9/24/2020 10:50:49 PM
Subject: Re: [QRPLabs] #80m Turning the 80 meter winding of T1 from hell to heaven

Steve

Interesting analysis,  and I appreciate your technical writing and presentation efforts.

Its worth an experiment. I do wonder about those 2 turn windings and whether they will be as uniform as the current design. I survived the 68 turns, even doing it twice on my 80m qcx+. I found 3 colors of wire for the 5 turn windings helped me greatly. If you try your recipe, report back and I will be glad to concede my thoughts.

I don't think many of us mind the designer's frugality,  it is quite off the charts good. This 80m qcx+ is an amazing nice build and operation, I didn't even mind the jumper wires for the PA TR cable. I wish Hans well in adding to amazing delivery of thousands more qcx rigs.

73 curt

Virus-free. www.avast.com


Al Holt
 

I think as a single band device the ferrite material would work well. My experience with them is limited; I've read about chamfering the edges of the toroid to reduce wearing through the enamel coating on the wire. But, coating the toroid or not winding so tightly probably would work as well.

Good luck with the experiment!!

--Al


Al Holt
 

I came across a FT50-61 toroid in my parts box and now the idea of converting my 'old' model 1 QCX over to 80m  might be worth a try!

Now that I've got two working QCX+'s, the older model is up for grabs, so to speak. It will be going from 40m to 80m and I'm pretty sure I can dig up some t37-2's here somewhere for the LPF's.

Sounds like a plan.....

--Al


@CurtisM
 

80m is a wonderful cause - see you when you get your QCX there. 

if it were me, I would be so tempted to use 3 turns instead of 2 (yes that is essentially 1.5 turns and with the challenge of landing each winding in a pair of holes in the PCB - just thinking it might give a more uniform response across the 3 windings in the IQ audio chain - just a hunch.  enjoy your build. 

73 curt


Al Holt
 

Hello CurtisM and Steven Dick,

I've finished the modification and it's working! I stuck with Steven's original design with 18t on S3 and 2t on Pri, S1, S2. Got the L1-4 wound and installed along with the band specific caps. T1 is all about receive but in order to get it aligned the Tx side has to be setup as well.

The most noticeable thing is in peaking the band pass filter. I'm getting about a midrange 7 on the bar graph scale and adjustment is rather broad. I think you might expect that type of response with the value of the trimmer cap on this band. I haven't searched this forum to see if that's the norm. I did install the proper 'padding' caps that parallel the trimmer as needed for 80 meters. The receiver section seems pretty sensitive and I'm hearing signals and since we had some storms in my area earlier this evening, I'm hearing much lightning discharge static.

The transmitter side is putting out 4W (using a Oak Hills Research WM-2) into a 50 Ohm dummy load, but I haven't tried optimizing its LPF toroids as per Hans' recent video.  

I'll get some photos done soon and get them posted. I'd say, so far so good!

--Al


Al Holt
 

Here's a pretty good close up view of the area of interest.


I have the PA FET's and driver transistor in sockets. The display is mounted away from the board using short jumpers.

--Al


Steven Dick
 

Glad it is working for you.  I haven't built an 80M version so I can't comment on the relative broadness of the adjustment range.  But it sure looks purdy in the photo you posted using the FT50-61 toroid.

-Steve K1RF

------ Original Message ------
From: "Al Holt" <grovekid2@...>
Sent: 9/27/2020 9:47:22 PM
Subject: Re: [QRPLabs] #80m Turning the 80 meter winding of T1 from hell to heaven

Hello CurtisM and Steven Dick,

I've finished the modification and it's working! I stuck with Steven's original design with 18t on S3 and 2t on Pri, S1, S2. Got the L1-4 wound and installed along with the band specific caps. T1 is all about receive but in order to get it aligned the Tx side has to be setup as well.

The most noticeable thing is in peaking the band pass filter. I'm getting about a midrange 7 on the bar graph scale and adjustment is rather broad. I think you might expect that type of response with the value of the trimmer cap on this band. I haven't searched this forum to see if that's the norm. I did install the proper 'padding' caps that parallel the trimmer as needed for 80 meters. The receiver section seems pretty sensitive and I'm hearing signals and since we had some storms in my area earlier this evening, I'm hearing much lightning discharge static.

The transmitter side is putting out 4W (using a Oak Hills Research WM-2) into a 50 Ohm dummy load, but I haven't tried optimizing its LPF toroids as per Hans' recent video.  

I'll get some photos done soon and get them posted. I'd say, so far so good!

--Al

Virus-free. www.avast.com


Al Holt
 

Steven,

Thanks! It did come out looking pretty good. Those 2 turn windings flop around before getting soldered down.

That FT50-61 has a rough surface, more than the usual powered iron ones we use for LPF's.  But, it is chamfered on the edges. I avoided winding it too tight, trying to minimize abrasion of the enameled wire. I'm not sure if it makes a big difference.

The band is pretty noisy with the storms we've been having in the South. No QSO's with it yet, but it's receiving stations pretty well.

--Al
WD4AH


Steven Dick
 

Al - do you actually see a peak with the adjustment?  Where does the cap actually sit in its range? You just mentioned a "midrange of 7 on the bar graph scale" but unclear if the cap itself is at midrange. If its almost fully opened or closed, that may not be at the peak and it will look broader.  If the cap is not near its midrange, you can squeeze or expand the number of turns on the big secondary to get the resonant point to move around.  Squeezing or compressing will have a larger effect for the higher permeability type FT50-61 core than the standard T50-2 powdered iron core.  

-Steve K1RF

------ Original Message ------
From: "Al Holt" <grovekid2@...>
Sent: 9/27/2020 9:47:22 PM
Subject: Re: [QRPLabs] #80m Turning the 80 meter winding of T1 from hell to heaven

Hello CurtisM and Steven Dick,

I've finished the modification and it's working! I stuck with Steven's original design with 18t on S3 and 2t on Pri, S1, S2. Got the L1-4 wound and installed along with the band specific caps. T1 is all about receive but in order to get it aligned the Tx side has to be setup as well.

The most noticeable thing is in peaking the band pass filter. I'm getting about a midrange 7 on the bar graph scale and adjustment is rather broad. I think you might expect that type of response with the value of the trimmer cap on this band. I haven't searched this forum to see if that's the norm. I did install the proper 'padding' caps that parallel the trimmer as needed for 80 meters. The receiver section seems pretty sensitive and I'm hearing signals and since we had some storms in my area earlier this evening, I'm hearing much lightning discharge static.

The transmitter side is putting out 4W (using a Oak Hills Research WM-2) into a 50 Ohm dummy load, but I haven't tried optimizing its LPF toroids as per Hans' recent video.  

I'll get some photos done soon and get them posted. I'd say, so far so good!

--Al

Virus-free. www.avast.com


Al Holt
 

On Tue, Sep 29, 2020 at 11:38 AM, Steven Dick wrote:
Al - do you actually see a peak with the adjustment?
Steven, yes, I get a peak, but it's broad and between fully closed and open the graph scale doesn't change much. I can test it again later to see how much the scale does change and report back. 

Where does the cap actually sit in its range?
 
Here's a close up of where I have it set currently after going through the BPF peaking adjustment procedure. You can see it's about 1/2 open.


I like your suggestion of adjusting the turns to get the cap to fall in a better spot; that's a lot better than removing a turn!

--Al

 


Steven Dick
 

ok - I wasn't sure. Looks good

------ Original Message ------
From: "Al Holt" <grovekid2@...>
Sent: 9/29/2020 12:18:51 PM
Subject: Re: [QRPLabs] #80m Turning the 80 meter winding of T1 from hell to heaven

On Tue, Sep 29, 2020 at 11:38 AM, Steven Dick wrote:
Al - do you actually see a peak with the adjustment?
Steven, yes, I get a peak, but it's broad and between fully closed and open the graph scale doesn't change much. I can test it again later to see how much the scale does change and report back. 

Where does the cap actually sit in its range?
 
Here's a close up of where I have it set currently after going through the BPF peaking adjustment procedure. You can see it's about 1/2 open.


I like your suggestion of adjusting the turns to get the cap to fall in a better spot; that's a lot better than removing a turn!

--Al

 


Al Holt
 

While messing with the BPF adjustments, C1, the 20pf trimmer cap failed.  It shorted out, some of the plastic insulating film got jammed from my rotating it.

I made a temporary repair with a higher value trimmer from my junkbox and the unit's if functional again. But even with this replacement cap the difference between low and peak is only about one major division of the bar graph scale falling in the "7" range.

I just realized I do have some 'official' QRP Labs' 20pf trimmers from some unbuilt BPF kits used in their receiver. I'm not too keen on those plastic 'polycon' trimmers and may wait 'till I can order one of the ceramic type like the one used in the QCX+ before replacing my temporary fix.

--Al


@CurtisM
 

Al

Somehow my stock trimmer survived, but I digress ... if you are getting a peak I imagine things are okay. It is hard to translate measurements across rigs. If you get a good null as you tune up the 3 pots that will give you reasonable confidence. I have s9 signals needing my volume at minimum, so don't worry with how much gets through the trimmer action. Of course its okay to substitute parts you prefer. Looking forward to another qcx on 80m. Enjoy.

Curt


Steven Dick
 

Sorry about your failed cap. The very broad peak doesn't sound right to me.  Perhaps some folks with a standard 80 meter QCX or QCX+ could chime in on how the peak behaves on 80 when adjusting the trimmer and what the amplitude scaling factor looks like. Is it more like 09 rather than 07?  Thanks,

-Steve K1RF

------ Original Message ------
From: "Al Holt" <grovekid2@...>
Sent: 9/29/2020 8:49:53 PM
Subject: Re: [QRPLabs] #80m Turning the 80 meter winding of T1 from hell to heaven

While messing with the BPF adjustments, C1, the 20pf trimmer cap failed.  It shorted out, some of the plastic insulating film got jammed from my rotating it.

I made a temporary repair with a higher value trimmer from my junkbox and the unit's if functional again. But even with this replacement cap the difference between low and peak is only about one major division of the bar graph scale falling in the "7" range.

I just realized I do have some 'official' QRP Labs' 20pf trimmers from some unbuilt BPF kits used in their receiver. I'm not too keen on those plastic 'polycon' trimmers and may wait 'till I can order one of the ceramic type like the one used in the QCX+ before replacing my temporary fix.

--Al

Virus-free. www.avast.com


Al Holt
 

Yes, it would be good to hear the experience of those with a standard build as far as scaling factor. 

I decided to go ahead and install an 'official' 30pf varicon trimmer cap. This time around I have better experience with temperature sensitive parts. Tinning their leads before soldering them to the board makes a big difference.

--Al


Steven Dick
 

Hi folks. I had previously posted that for the 80 meter QCX+,  the large number of turns on T1 (68 turns on secondary 3) can be greatly reduced by using a type FT50-61 toroid instead of the T50-2 toroid.  On the FT50-61 toroid, this would require only 18 turns on secondary 3, and 2 turns for the PrI, S1, and S2 windings.  Type 61 material is a relatively low loss ferrite, while -2 material is a powered iron.  It's exactly the same size, and it's Q will be comparable to the -2 powdered iron core.

Page 177 of the QCX+ manual shows the measured characteristics of the standard 80 meter bandpass filter.  So it looks like the loaded Q (the in-circuit Q) is 3520 KHz/343 KHz = 10.2  That's a very low loaded Q.

Al Holt, WD4AH, had built an 80 meter T1 using an FT50-61 which works fine and he has since taken Q measurements on one using one of the techniques written up in "The Two Faces of Q" by Wes Hayward, W7ZOI, the measurement described on page 2 of the document:

http://w7zoi.net/twofaces.pdf

Al measured an unloaded Q of 149.7 with bandwidth of 22.8 kHz with a separately wound toroid using a Nano VNA.  The actual Q is a bit higher than that due to a small amount of loading by the test fixture. It is comparable to the unloaded Q of an FT50-2 with standard windings. See the in-circuit T1 wind and a separate core with a single winding for use in the test fixture below.  The unloaded Q is way way larger than the loaded Q and so has little effect on the loaded Q.  The FT50-61 is thus comparable in performance to the standard T1 at a cost for the ferrite FT50-61 toroid about twice the cost of the standard powdered iron toroid, a T50-2 but with an easier wind on 80 meters.

-Steve K1RF







Virus-free. www.avast.com


Al Holt
 

Steve,
Thanks for updating this post! And thanks for your guidance getting through the theory and measurements! 

Yes, the rig is sounding good! I had a QSO with W1DV up in North Carolina with it this morning. I'm in northern Florida and got a 479 RST report (probably due to my nervous fist). His signal was 599++ here.

I'll just pass along I converted my 'series 1' QCX from 40m to 80m using that FT50-61 toroid. In the process the LPF was changed along with the 'band specific' caps and the transmit inductor L4. Since Hans supplies all caps needed for all bands I had them on hand. 

I overlooked that Wes Hayward was author of that paper....definitely time for a closer review!

IMHO the ferrite for T1 is the way to go!

73,

--Al


Steven Dick
 

Thanks for the update Al.  Yes, I think the small additional cost makes it a lot easier to wind.

-Steve K1RF

------ Original Message ------
From: "Al Holt" <grovekid2@...>
Sent: 10/4/2020 6:18:03 PM
Subject: Re: [QRPLabs] #80m Turning the 80 meter winding of T1 from hell to heaven

Steve,
Thanks for updating this post! And thanks for your guidance getting through the theory and measurements! 

Yes, the rig is sounding good! I had a QSO with W1DV up in North Carolina with it this morning. I'm in northern Florida and got a 479 RST report (probably due to my nervous fist). His signal was 599++ here.

I'll just pass along I converted my 'series 1' QCX from 40m to 80m using that FT50-61 toroid. In the process the LPF was changed along with the 'band specific' caps and the transmit inductor L4. Since Hans supplies all caps needed for all bands I had them on hand. 

I overlooked that Wes Hayward was author of that paper....definitely time for a closer review!

IMHO the ferrite for T1 is the way to go!

73,

--Al

Virus-free. www.avast.com