Which end fed half wave transformer should you use for running high power?

[@stephentrier5569]

Nice work! Thank you for doing the work to get this data and for sharing it so clearly.

[@MM0OPXFieldRadio]

Check out my later video. I clear some things up and these are not as bad as thought.

[@TheSmokinApe]

Great work Colin, I have to admit I’m quite surprised by the data. Also, it just makes me laugh how people will only look at the SWR and then sing an antennas praises.

[@MM0OPXFieldRadio]

Thanks Ape 👍

[@super66craig]

Yep, you never hear anyone talk about losses, efficiency, and actual repeatable results from testing. Just opinions and like a butt we all have one and they all.....

[@super66craig]

Colin, you are a breathe of fresh air in a room full of steaming cow pies from the "experts" touting the next new killer antenna. Efficiency is everything and those of from military electronics had that pounded into our heads to improve every connection and never assume anything. We lived by the screen of those big analyzers to tune every drop of power from that gear so my desire in this new hobby is to get it right. With your help in this I'm well informed now. 73

[@MM0OPXFieldRadio]

I wouldn't call myself an expert, just an experimenter. I have a suspicion the 52 marerial tests had some measureny error. They still wouldnt be good across the whole of HF. The search for a QRO core goes on.

[@presidentspilot]

Thank you so very much, Colin, for doing all these tests! I am STILL amazed how you experimented, so much, as to come-up with a 56-1 ratio!! Was it by accident? WOW! That is a LOT of testing!! LOL!! Your MAJOR effort has saved me purchasing a LOT of equipment that I currently do not own, not to mention all the different cores, and all the winding configurations that you managed to wind!! WHOA!! Again my deepest thanks, Colin!!

[@MM0OPXFieldRadio]

Your very welcome.

[@stevehanlon7627]

great follow up, appreciate the testing and clear analysis of the results.

[@MM0OPXFieldRadio]

Thanks.

[@brianspilsbury8953]

What I think is that your videos are in the top few percent for easy to understand information for many of us non-techy ham operators on youtube

[@MM0OPXFieldRadio]

Thank you. I try to make them that way. That's the trouble I used to have also.

[@edahan]

Colin, these tests and incredibly useful. Your videos are just long enough and done well. Many appreciate the work you put in, and I have used your results as part of transformer builds. Thank you again.

[@MM0OPXFieldRadio]

Appreciate the comments Erick, I do my best to keep the waffle to minimum and get the information out there I think is relevant.

[@alanb76]

Great to see the data and the techniques to get it. Thanks.

[@MM0OPXFieldRadio]

Thanks 👍

[@hubercats]

Thank you for another excellent video. I especially like your comment to others concerning providing data to back up their claims. Many years ago, while attending a technical conference, the speaker mentioned an adage that stated “Data is an insult to good theory”, a saying that resonated strongly with me. - Take care - Jim

[@MM0OPXFieldRadio]

Options are nice, not a good as data though.

[@m0aze611]

Phew! Nice to know I’ve been doing it right. Great analysis Colin - I don’t tend to use a single wire for my /p operation I only use resonant lengths that take a few minutes to change. Being an oldie I believe that you get out what you put in when using resonant antennas. What I don’t know is at what temperature does the core decrease the efficiency. I’ve been using EFHW for years now for portable work but I do like the linear loaded ¼ wave vertical for home. But after 40 years it’s hard to say what has been the best performer as conditions constantly change. Best DX in the log was using a G5RV - over 30 years ago. Doesn’t actually prove a lot. Thanks for posting. Regards Mike

[@MM0OPXFieldRadio]

Certainly every antenna gets its 15 minutes of fame. As long as we are happy then thats as good as it gets.

[@Oldnoitall]

Fine Business KE4TDG

[@pcfreak1992]

Thank you very much for these videos! They're always incredibly insightful!

[@MM0OPXFieldRadio]

Your welcome.

[@N0BoW_99]

WOW!! Man I really enjoy these videos. I recently just purchased some cores and will be testing very soon!

[@MM0OPXFieldRadio]

Excellent, please share your results.

[@DaveW6OOD]

These are great results to show the true differences. I’ve always wondered about how efficient the different 9:1 winding methods are. Such as if there is any choking or efficiency difference between a 3/9 or 4/12 and the three times around method. I vote for that test. Thank you for the work.

[@MM0OPXFieldRadio]

The Smokin Ape has started to make these measurements and the results look very positive. Check it out if you haven't already.

[@mareksp9tkw908]

I compared 49:1 and 9:1 on FT140-43 cores and 9:1 was far more efficient than standard 21:3 and a bit better than KN5L method of winding using 28:4 turns. 9:1 was 3x5turns

[@mareksp9tkw908]

Hi Colin, thanks for sharing your results. I believe you can substantially improve accuracy of your results by changing a bit your measurement setup. I make my comments on the basis of image at 2:13min. Primary connections are much too long and this has effect on losses/efficiency especially, on high frequencies. It would be much better, if you made these connections very short

[@MM0OPXFieldRadio]

Thanks. I have seen this method but still to get my head round the maths. I need to try this for an existing. Yes I agree there is room for improvement, how much I don't know. It may be significant it may not be. Only one way to find out.

[@evil_lair_electronics]

I agree about the lead length, for both the transformer and the capacitor, and have seen a small reduction of loss by making this as short as I could. However it was only about 0.1dB and was only around 10 meters, the rest of the trace was virtually identical. I'm sure it would be worse with more frequency, but I don't know why anyone would want to use this type of antenna above 10 meters anyway. So yes any test jig is going to have some lead length and show this loss unless you really push it, making it far more time consuming when testing a ot of transformers, and the same is true when you put the transformer/capacitor in an enclosure and connect magnet wire over to an RF connector unless the enclosure is really a super tight fit for everything. This is one reason why I started making circuit boards and connecting the coax directly to them, the lead dress is reduced to the lowest practical level. I drew a board for a jig this week to do the single transformer tests which will have a 2.4k SMD resistor on it(VNA port 2 of course already provides 50 ohms of load). I'm interested to see if a single tests much different than a pair. Other problems we can't do much about, like permeability vs temperature, if you look at 43, it's about 750 at 20C, but double that by 95C. 52 is not much better, but 61 is WAY better.

[@super66craig]

​@@evil_lair_electronics I've suspected the 61 was the better choice based on curie points from mfgr data.

[@J123G]

Thanks for another extremely useful video. I know the title of this is related to higher power but for my use of 100w I will follow the results on your spreadsheet as it relates to the Fair-Rite - 2643251002 core. ie: 15:2 wide/crossover +100pf seems to have the best results of all for 40,20,15. Much gratitude for all your work.

[@MM0OPXFieldRadio]

I haven't seen anyone better this for a 100w core. Remember you can run 100w FT8 with the 1002 core.

[@LarrySavage-no4lk]

The issue is indeed transformer inefficiency. At 1500 watts and the typical efficiencies of these transformers you will generate approx. 300 watts of heat. Place these transformers in a plastic enclosure, without cooling, and they will surely fail if high duty cycle modes are used. Unless you can keep the heat generated to low levels or add cooling and limit duty cycles these transforms are not reliable. The military immerses this type of transformer in a coolant oil and use a metal enclosure to dissipate the heat. With the typical efficiencies of these EFHW transformers (80% range) It does not matter how many cores you use in the transformer. More cores will just increase the time it takes to heat up to the point of transformer circuit or enclosure damage.

[@MM0OPXFieldRadio]

I agree in principle. I wont however be tunning Digital or CW at these power levels. That's crazy. Im likely going to run 800w max SSB and it cruises at this.

[@jorgemolina4649]

I was going to use mix 52 but after seeing your video I will use mix 43, thanks again for the comparison.

[@MM0OPXFieldRadio]

Just depends how much power you want to run.

[@K9BCM]

Thanks for sharing Colin!

[@MM0OPXFieldRadio]

No problem 😊

[@larryprice6195]

Thankyou for your videos, they are very informative!!!!

[@MM0OPXFieldRadio]

Your welcome.

[@JayN4GO]

Thanks Colin. Always humble. I’d like to try the 52-3 core during a contest to see how long it takes before the cores heat and trip the amp at 2:1 swr.

[@MM0OPXFieldRadio]

If you do, let me know how you get on.

[@trig6712]

Excellent Colin Many thanks

[@bruceweight7319]

I really enjoy your videos. This end fed antenna thing is a really variable device. I tried a lot of the ideas on FB and really the resistor/SWR tests really didn't do so hot! No such thing as plug and play. Can't tell you how many times I see...oh you have to bunch the windings together and so on. But really the real test is to get an antenna hooked up and see how it plays. I did just get some of the thick cores and am going to see how they tune up. It would be nice to see how your vna showed the graph on your antenna on the field day test and or if you used a tuner to touch things up. I run 500 watts on there mystery cores same size roughly as the 240 cores. I think they might be 77 type or k type but not sure. They handle 450-500 ssb and 35-40 watts ft8 they are wound with 12 ga PTFE 200 degree c silver coated wire. I have been using it here at QTH for a number of years now 80m which I mainly use on 40 20 does ok and 17 but crap above that. But several known hams are using the 43 thick cores and I am looking for a good small portable 100 watt transformer and that short fat core looks fun so I got two.

[@KK6USYHamRadioAdventures]

Nice job Colin!

[@MM0OPXFieldRadio]

Thanks Chuck. Those 11mm poles arrived. Still to test the tension but look decent.

[@willian.direction6740]

Thanks Colin for doing the work on these cores. Since I swapped to an EFHW on SOTA I have been making DX QSO's almost every time on 15m and 20m. Maybe the up lift in propagation has helped but now I regularly get JA chasers on 15m and long path DX in my afternoon on 20m from summits.

[@MM0OPXFieldRadio]

Conditions certainly improved but always helps having the edge.

[@jerrym1183]

Murphy's Law dictates that when ever a new antenna/transformer is tried the "on the air" HF conditions will never be typical but will be "fantastic" or "less than favorable" hi hi. Is always best to try an antenna for a week or two to fully get a feel on how well it is working. The Black Magic of RF and Ferrite material is a fascinating and addictive subject, as I am discovering ... !

[@KS0JD]

I like the test. When are you going to post the spread sheet? I have a few I want to try. I want to end up with a 40M end fed that can handle 1.5Kw. Great video Colin! 73

[@MM0OPXFieldRadio]

Yeah I forgot. I added it this morning 👍

[@biringen]

Hi Colin i tried a theory of mine but not able to test with a vna… but when i winded a standard 2:14 49:1 unun 100pF on 240/61 core compared to a 240/43 on my rigexpert trough a 2k7 resistor tested both cores using same resistor and centerfrequency at 15Mhz, 28Mhz span…the swr curve for the 240/43 was very low and started climb fast above 14mhz but on the 240/61 core it went from high swr and after 14Mhz it falled rapidly down… so for my qrp station when im portable i use the 49:1 with 240/61 for frequency above 14Mhz as it seem more efficient on higher frequency than with 240/43 or the 240/52

[@MM0OPXFieldRadio]

Yeah just sweeping for an SWR curve doesnt tell you anything about the efficiency. Some people do however.

[@robertdeforge7809]

When winding the cores with the cross over how much wire is needed? What size magnet wire are you using? Any specific coating type? I have everything on order except for the wire. I cant wait to get this built. Thanks for all your great videos. Bobby KC2BBI

[@MM0OPXFieldRadio]

The wire was 1.16mm diameter. The wire is just high temp enamelled copper wire but its not critical. Its actually a pain to scrape off as it wont "tin" off with solder. I buy from Brocott here in the UK.

[@101blog]

Good standard of Video and investigation method . I was going to ask if you had done the experiments with a wide winding seeing that your cores were all tight wound then I saw that you have done both. I'm not supprised that the efficiency was much better in the wide winding as I remember they normally expect you to wind over 3/4 of the core circumference to get the full use of the core .. not sure where that comes from but seems to be held up here with your experiments. Not sure about any other effects such as mutal inductance been an issue here... good one

[@MM0OPXFieldRadio]

Thanks. I think inductance is playing a part but cant prove that.

[@philipwells2793]

Good bit of empirical science. I am surprised that your tight winding done on smaller cores don't work here.

[@MM0OPXFieldRadio]

I know. Perhaps the geometry is a cause of this effect.

[@kevinkc3onohelijeepworld953]

The best ferrite we found for common mode choke was a Amidom Kmix core and a stack of 4x2.40” cores gave us over 7500ohms of common mode blocking 😉not sure how well it would work as a transformer balun ?

[@MM0OPXFieldRadio]

Wonder what the mix is?

[@antennafarmer7380]

Some guys are saying now that stacking 2 cores is actually better. I'm still a believer that you should have some dedicated antennas for particular bands instead of a single omni.

[@MM0OPXFieldRadio]

I wouldn't be surprised. I tool agree, mono band is always best when possible.

[@KO4VNX]

Colin, huge fan of yours, of course, but I have two questions: 1) what gauge of wire are you wrapping these with? and 2) what gauge of wire would you use for an EFHW antenna (not the core, but the actual antenna) if you use one of these transformers, considering that I'm going to be pushing 100 watts (I already have a great EFHW for my QRP rigs)?

[@MM0OPXFieldRadio]

So I was doing my tests with 0.6mm wire and for your needs go with something around 1mm. I've used 1.12mm and 1.16mm.

[@KO4VNX]

@@MM0OPXFieldRadio okay, more questions, but see? that's what you do, you make us think and not just copy things! When I look at your worksheet for the 2643251002 for 100w (I think the 5th tab in your document) is this a single core? or a triple stack? It looks like a single, and if so, why do a triple stack when a single stack looks so good (again, just looking at your spreadsheet). If this is a single you are documenting in that tab, it seems really like a close race between the wide turns with the crossover and the close turns. Also, if I use this core and this winding, would it hurt my efficiency ratio if I did QRP on this one too? I mean, my goal is 100w, but was just wondering how QRP on that same core could have any negative consequences?

[@shaneeloina8474]

Hi great video, have you tried type 31 material in any of your tests?

[@MM0OPXFieldRadio]

No, I dont plan on either as I think there will little gain if any over 43 or 52.

[@stevewithnell911]

Binocular cores are more efficient than toroids, which is why we see them used as transformers in amplifers. I wonder if you have time to test that configuration - say a couple of the fair-rite 26343251002 epoxied together to make a binocular core. I'm not a fan of back to back testing, for absolute measurements, for relative comparisons its a better method. My findings were that the efficiency is highly influenced by the antenna loading which changes quite dramatically with frequency (I used core temp as the indicator). One test method I have not seen punted is to do a open, short and loaded S11 measurement. The reflected power measurement then shows what is being lost in the core - for open and short measurements, a perfect core would have 0dB reflected power - 100% of the power would be reflected back. It gives quite a rich view of performance.

[@MM0OPXFieldRadio]

I will look up that core and see availability. it would be good to try.

[@stevewithnell911]

@@MM0OPXFieldRadio I made up a binocular core from 4 FT140-43's it's really ugly, but the the efficiency with 4 turns comes out at 98%. The core I mentioned above was the one you were recommending in your video, it's not a new one from me.

[@donaldsmith3048]

I would like to see it without the cross over. Just spaced around the core. Most say it doesn't matter. But the cross over wire is Not near the core, and so may not be the same. It is done so the coil ends at the top of the box. I did one without the cross over and spaced the windings out. It works well but I don't have a way to test it. It was my first and only coil. With your information I will order 2 more 240-43 to wind a 3 core. But I still will not have a way to test if the way everyone windes them with the cross over is better or worse than just winding them around. I know winding with all the wire in a small spot will not work well. I have had some tell me that not using the cross over is better. I can't test! Just getting into the HF side of things. 73 W4DES

[@MM0OPXFieldRadio]

On the chunky 1002 core I tried both and there was no measurable difference. There may be with different cores though as the hole is larger.

[@JeffCowan]

Colin, are you aware of anyone stacking 2 cores of different material types? Just thinking out loud here but one better for lower bands and one for upper bands. Might end up with a reasonable compromise?

[@MM0OPXFieldRadio]

I hadn't but you are the second person to say this. Maybe some 61 material in the mix would be good.

[@FFdo.]

I think there is a commercial choke that implements just that wont mention names but there a long rounded balun which is sold that mixes different types. I think mixing it up can improve results and coverage across the HF range.

[@paulsengupta971]

I think someone's done a stack of two 43s with a 52 in the middle, or maybe I made that combination up in my head, I'm not sure what's real here! 🙂

[@Htfhjooffhgghwwwwdf]

Another very helpful and informative video. THANKS Daryl de VK3AWA

[@MM0OPXFieldRadio]

👍 👍 👍

[@R50_J0]

Good.

[@sgs1262]

Do you have a video showing how to do the auto transformer winding

[@MM0OPXFieldRadio]

Not specifically no. If you search "end fed half wave Autotransformer" you will get numerous diagrams. If I can get time I will make a video on this.

[@sgs1262]

Thanks, I'll check that out. Thanks for the videos they're very helpful@@MM0OPXFieldRadio

[@QuiltersHandyman]

You mainly address end fed half wave antennas in your videos. How about other types of antennas should they also use baluns and chokes? I pretty sure the answer is yes. I am putting up a verticle antenna, 10M thru 160M. The manufacturer ( it is a DX Engineering MBVE-5A) recommends using their 4:1 balun which is rather expensive so I would like to build my own. Can one have both a 4:1 balun and 1:1 choke on the same feedline? How does their UNUN cover 1.8 -30MHZ at 2000W? What is in that box? Max power i run ia 200W and a lot of QRP so power isn't the issue. I have watched many youtube videos and chokes and baluns but am still confused....somewhat normal for me. Can you help?

[@101blog]

BTW have you tried stacking those stubby cores that did so well does that extend out in to a double/tripple stacked core ... It would look odd I admit and harder to box but good if the values are better!

[@nickjparker1979]

hey 101blog, I've made just that. stacked stubbies x2. primary winding with 1.5mm2 enamelled and secondary winding with 1.25mm2 enamelled. in a 49:1 config. No testing like our man has done here as yet though.....

[@MM0OPXFieldRadio]

I havent stacked these but I have stacked the chunky QRP cores. What happens is the the low bands get marginally better but 20m and above falls off a cliff unfortunately. The same will happen here so based on experience no point in trying to stack these.

[@101blog]

@@MM0OPXFieldRadio OK I read ya, I've ordered 3 from Mouser so may well try stacking them for the sake of knowledge. Thanks again for producing a well collated and consistent body of test results !!

[@WayneHarris]

I'd love to scale up my EFHW antenna to be able to handle 1.5kw on digital. Can a 49:1 transformer be built to handle this much power?

[@MM0OPXFieldRadio]

Tough one, if its FT8 or RTTY big difference on duty cycle. sure its possible.

[@WayneHarris]

@@MM0OPXFieldRadio Yeah, I'm not sure either. Would have to be a triple stack of Ferrite rings and perhaps 10ga wire windings. At least.

[@vincenzofidanza2539]

Did you try to mix the mixes ex. type 52+ type 43?

[@MM0OPXFieldRadio]

No but I think if that did work perhaps 61 mix might be needed.

[@vincenzofidanza2539]

@@MM0OPXFieldRadio yes I agree

[@geirha75]

Given that I have ft-891 radio max 100W output.. can I go for a smaller core, 43 material ( FT140-43 )? Which one would you recommend?

[@MM0OPXFieldRadio]

You need the core ending in 1002. Have a look at my previous videos for one of the best 100w transformers.

[@75ohmHAM]

Would 3 primary and 21 secondary be a linear improvement with the 52mix? (Wide w 100pf)

[@MM0OPXFieldRadio]

I don't think so. I beleive it will improve lower bands marginally but at the expense of the higher bands.

[@bruceweight7319]

Do some workups on some FT-240 type k cores.

[@MM0OPXFieldRadio]

Whats the difference with a K core?

[@paulsengupta971]

Hi Colin, any chance you can do some testing with single 52 cores before you send them back? 🙂

[@MM0OPXFieldRadio]

I doubt it, I have very little wire left either unfortunately.

[@paul-c7541]

Colin very interesting, as my interest is QRP, I like every bit of power I can to radiate, I'm going to have a go at the QRP model with 21/3 turns thank you 73 Paul M0BSW

[@MM0OPXFieldRadio]

Best of luck 👍

[@shanerorko8076]

Yea but you can measure without a VNA using WSPR as a measurement tool. I hate this idea that contacts aren't a measurement tool and because you can make a contact with a fence wire all antennas are dummy loads unless you have 0.001db loss. It's not 100% correct to frame the argument like that, if I can get my fence wire to make more contacts on WSPR than any other form of antenna then it is somehow completing the task, just because you haven't considered the aspect at which it is doing this doesn't negate the fact it is.

[@MM0OPXFieldRadio]

Not quite sure how you think I feel differently? This video is regarding the transformers only. At no point in any of my videos have I ever said only focus on this, field testing goes without saying. WSPR measurments are great but only if you have a reference. In order to conduct whats better i.e A or B you need simultaneous WSPR RTX on at least 2 different antenna for the same band in the same proximity. Only 1 person I know does this and that HB9VQQ. Look him up if you dont know him.

[@75ohmHAM]

Do 3 cores of 43, 15:2 turns, cap and no cap

[@MM0OPXFieldRadio]

Thats just a 56:1, the winding style is no different. I'm done with testing these for now.

[@jamess1787]

Wen 56:1 video? 🙏😊

[@MM0OPXFieldRadio]

kzbin.info/www/bejne/jpaTqKmYhLaYpc0

[@jamess1787]

@@MM0OPXFieldRadio definitely need spoon-feeding. I figured out the winding pattern I think, took like 20 times watching it though and examining the core from every angle Cheers!

[@MM0OPXFieldRadio]

Not at all. We all learn somewhere.

[@mrx7956]

Great Video, as usual... very informative and professional in the EFHW community... thanks for the data. / 73 de DG1FBN

[@MM0OPXFieldRadio]

Thanks, your welcome 👍