Balanced tuners
Locked
 
|
Strictly speaking, the term "balanced tuner" should not be used for ATUs
that have the balun on their output. They are inherently unbalanced. With highly reactive or mismatched loads, the balun could give rise to significant losses. A true balanced tuner has the balun at the input where, once the ATU is propery adjusted, will always be at 1:1 and hence no reactive elements. Everything else after the balun is floating wrt RF earth. I think a recent survey showed that these tuners typically had losses of 1% or less, and rarely more than 3% on topband. I have the Palstar BT1500A and have been very pleased with it. It's rated at 3kw, so with 400W UK legal limit, it's never stressed. Not a cheap ATU. Inside, good engineering with very heavy duty relays, silver plated twin roller coaster inductors. I have no regrets about trading in my AT1500CV (unbalanced) one for the BT1500A. I use it with a 140ft doublet at 50ft, and a vertical dipole fed with open wire feeder. One project I was thinking of was to make a switching box for several balanced antennas for it that also allowed the doublet feeds to be strapped together to make a top-loaded vertical being driven against my ground radial system for topband. It's a pity that no one that I know of does an auto balanced ATU, even for 100W. Now, if Elecraft did a kit for a 1kw balanced auto ATU, I'd be seriously interested. :-) Especially if it had switching for several antennas. David, M0DHO _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
David Honey wrote:
> A true balanced tuner has the balun at the input where, once the ATU is > propery adjusted, will always be at 1:1 and hence no reactive elements. This is not correct. As you suggest, placing the balun at the input of an unbalanced tuner is an attempt to solve the problem posed by operating a balun in the output of the tuner, where the impedance is uncontrolled. However, W7EL (I can't find the reference just now) has shown that this method is actually no better than the balun in the output! A true balanced tuner does not have a balun at all. For example, the Johnson matchboxes make the unbalanced to balanced transition by using an unbalanced link coupled to a symmetrical parallel-tuned tank circuit that is connected to the feedline. There are several true balanced tuners available: MFJ makes one, and I've seen others. Just Google 'balanced tuner'. -- 73, Vic, K2VCO Fresno CA http://www.qsl.net/k2vco _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
Vic K2VCO wrote:
This is not correct. As you suggest, placing the balun at the input of an unbalanced tuner is an attempt to solve the problem posed by operating a balun in the output of the tuner, where the impedance is uncontrolled. However, W7EL (I can't find the reference just now) has shown that this method is actually no better than the balun in the output! ----------------------------------------- That one keeps coming up, Vic. I've read W7EL's dissertation many times, and I've built tuners with baluns at the input that overcame the problems with one at the input. W7EL's argument seems to confirm mathematically that a balun looking into a highly-reactive load will perform equally poorly at the input or at the output of an ATU. That argument seems to ignore that, in actual practice, we adjust the ATU so the balun is looking into a 50 ohm load with very low reactance when placed at the input. Kevin Schmidt, W9CF, has a nice dissertation that includes Roy's findings and a reference to the original paper and where you can download it (http://tinyurl.com/ctkwb). Kevin agrees with Roy, but Kevin says that the reason it's "useless" to put the balun at the input is that it makes both "sides" of the ATU "hot". Of course it does! You need a purpose-built ATU. Others claim that not building a fully-balanced ATU with identical reactances (coils and capacitors) in each leg unbalances the antenna. I have not found that to be the case at all. After all, the idea is to use efficient coils and capacitors, so there is very little loss in them. The coils and capacitors become 'transparent' to the RF when the system is resonated. All that's needed is to make sure that stray capacitance to ground is kept to a minimum. That's true when routing the feed line as well. I've used unbalanced T and L matching networks at the output of a balun with excellent balance in the feeders. In my experience, there's nothing wrong with putting a balun at the input or the output of a tuner. There's a lot can go wrong if one subjects that balun to highly-reactive loads! Many of the popular tuners that have been sold over the past several decades were intended to be used with *resonant* antennas - antennas that presented a low SWR to the rig without a tuner. The popularity of "no tune" rigs that began in the 1970's created that demand. Sure, you didn't have to tune the output of your rig any longer but you had to have an antenna that presented, at worst, a 1.5:1 or 2:1 SWR to the rig. A lot of antennas didn't meet this criteria, showing a somewhat higher SWR to the rig. That's where the MFJ and other popular, medium-cost tuners came in. They could be put in the feedline and provide the required low SWR to the rig. Instead of tuning our final amplifier output circuits in our rigs like we had done since the 1920's, we now adjusted an antenna tuner to accomplish the same thing. Why not simply go back to tuned final amplifier outputs on rigs? The new Federal Communications Commission regulations here in the USA raised the standards for harmonic and other spurious signal suppression above what a simple tunable final amplifier output circuit could provide. More exotic multi-section filters were needed, like those in the Elecraft rigs. This is especially true above QRP levels. They aren't easily tunable. So, for years now Hams either connected an antenna directly to their rig or, if the SWR tended to range to high on some frequencies, used an ATU in the line to provide the low SWR the rig needed to operate properly. Those ATU's often included a balun at the output for hams feeding *resonant balanced antennas* such as a folded dipole. A folded dipole can be made up of 300 ohm TV 'twinlead' or similar line. It has a feedpoint impedance of 300 ohms. Using an ATU designed to look into something in the general vicinity of 50 ohms on the antenna side with such 300 ohm feeders was easy: use a 4:1 balun. The ATU would see 50 ohms, thereabouts. Other antenna configurations that presented a reasonably low SWR to the balun also worked well. Enter the WARC bands. Suddenly the old center fed wire using open wire feedlines that popular in the 1930's is back to accommodate all those bands! Forget feedline SWR, the ATU will fix it. Trying to use such an antennas with a modern ATU using a balun at the output led to a lot of grief and confusion. On some bands the system works. On other bands baluns simply convert most of the RF to heat. As others pointed out here, trying to measure ATU efficiency when feeding odd loads is a challenge. (The most straightforward way to do that is to put the ATU in an insulated box, measure the temperature rise while it's being operated and then calculate the amount of power converted to heat. Straightforward, but not especially easy to do). So now there's renewed interest in balanced tuners that can handle not only a wide range of impedances but can efficiently couple to balanced lines. A balun at the output can work FB on some bands, but it's tough to predict. It's even tougher to evaluate how well it's working since a few milliwatts radiated power can gain a 599 report halfway around the world under the right conditions. Personally, I use a standard link-coupled fully-balanced manually-tuned ATU to feed my doublet. It's not pushbutton quick. It sometimes takes me 10 or 15 seconds to retune on a new band. But it's very efficient and has a tremendous matching range across the HF bands. It's the same circuit shown in many old manuals and on Cebik's web pages http://tinyurl.com/7ez6r There's a picture of it above my K2 under my call listing on QRZ.COM. Besides, it looks "cool" to me (speaking of fashions <G>) and I can lay an NE51 under one end of the coil and watch it blink on and off as I send. (Tiny minds enjoy small amusements). Ron AC7AC _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Vic K2VCO
Vic: You are right. The URL for the "brief" version of W7EL's paper is
<http://eznec.com/misc/ibalbrf.txt> and a more detailed version is at <http://eznec.com/misc/ibalun.txt>. It's not clear to me why one would want to hang on to a balun in a balanced tuner. A simple L-C network on the input is very effective (or a movable link). Someone just posted that their best bet was a homebrewed transmatch, and I agree with this. A balanced homebrewed transmatch is a simple project. To be sure, it can be made very complex if one wants, but that is really unnecessary. I have used my Italian homebrew with great success for decades. I call it my Italian tuner because it looks like a plate of spaghetti, but boy does it work. best wishes, dave belsley, w1euy On May 13, 2005, at 11:13 AM, Vic Rosenthal wrote: > David Honey wrote: > >> A true balanced tuner has the balun at the input where, once the ATU >> is propery adjusted, will always be at 1:1 and hence no reactive >> elements. > > This is not correct. As you suggest, placing the balun at the input > of an unbalanced tuner is an attempt to solve the problem posed by > operating a balun in the output of the tuner, where the impedance is > uncontrolled. However, W7EL (I can't find the reference just now) has > shown that this method is actually no better than the balun in the > output! > > A true balanced tuner does not have a balun at all. For example, the > Johnson matchboxes make the unbalanced to balanced transition by using > an unbalanced link coupled to a symmetrical parallel-tuned tank > circuit that is connected to the feedline. There are several true > balanced tuners available: MFJ makes one, and I've seen others. Just > Google 'balanced tuner'. > > -- > 73, > Vic, K2VCO > Fresno CA > http://www.qsl.net/k2vco > _______________________________________________ > Elecraft mailing list > Post to: [hidden email] > You must be a subscriber to post to the list. > Subscriber Info (Addr. Change, sub, unsub etc.): > http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/subscribers.htm > Elecraft web page: http://www.elecraft.com > Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Ron D'Eau Claire-2
Ron D'Eau Claire wrote:
> Kevin Schmidt, W9CF, has a nice dissertation that includes Roy's findings > and a reference to the original paper and where you can download it > (http://tinyurl.com/ctkwb). Kevin agrees with Roy, but Kevin says that the > reason it's "useless" to put the balun at the input is that it makes both > "sides" of the ATU "hot". Of course it does! You need a purpose-built ATU. > <http://eznec.com/misc/ibalun.txt> (W7EL) > <http://tinyurl.com/ctkwb> (W9CF) That's not his point, Ron. He shows that the balun does not do a better job on the input than on the output, and then says 'why bother building an isolated tuner when it's no better than a conventional unbalanced tuner with a balun on the output?' He shows that a true balanced tuner is one effective solution to the problem. As you know, the degree of imbalance in a parallel wire line can be expressed as the ratio of differential current (flowing in opposite directions on both conductors) to common-mode current (flowing in the same direction on both conductors against ground). In a coax line one doesn’t talk about imbalance, but the common-mode current is what (undesirably) flows on the outside of the coax shield. The transmission line displays a differential impedance – that’s what we usually refer to as the impedance ‘seen’ by the transmitter or tuner – and a common-mode impedance which is based on its undesired function as an antenna for common-mode currents. What W9CF shows is that an unbalanced tuner, in transforming the differential impedance to 50 ohms, inserts an unbalanced reactance which actually increases the common-mode impedance. Therefore, a balun at the input to the tuner is stressed exactly as much as one at the output. This assumes that the effectiveness of the balun doesn’t depend on the differential impedance of the line, and this is more or less true of true current baluns such as bead baluns or transmission-line baluns. Voltage type baluns, such as the 4-1 baluns often found on the output of cheap tuners do not have this characteristic and so are even worse when working into extreme impedances. On the other hand, a true balanced tuner like the Matchbox will reduce common-mode current significantly in itself. > In my experience, there's nothing wrong with putting a balun at the input or > the output of a tuner. There's a lot can go wrong if one subjects that balun > to highly-reactive loads! W9CF shows that for reducing common mode currents, the best choice is a balanced tuner with a balun on the input. -- 73, Vic, K2VCO Fresno CA http://www.qsl.net/k2vco _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by David Honey
OH9NB shared with us:
>Another very clever (with memories and all) can be seen >in www.hamware.de and is available in 200W and 1000W versions. >That was originally made for a starshape multi-V-beams. EUD1.6k, or even EUD450 for 200 watt kit version. Nice, but ouch. Would take saving a lot of rice money to be able to afford that. ;^) Never thought much of auto tuners, though they do open up some interesting antenna possibilities if remoted. Instead of a SteppIR dipole, I would love to take an old 40m beam element (without the linear loading) & feed that with a proper tuner - essentially Cebik's 44-foot doublet, hopefully without any of the potential difficulties in feeding it. Automatic or just remote control of the knobs, I don't care as long as it works good. It's easy to beef up an L/C/relay tuner for higher power, but I always come back to putting motors on a Johnson kW matchbox & don't get any further than that. It's good to see tuned feeders being rediscovered; it's unfortunate that it's so dear to treat them right nowadays & that the market has skewed so far away from balanced in general (not to mention transmitting components themselves - I hope my stash of ARC-5s is still safe back in Dad's basement ;^). 73, VR2BrettGraham _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
«
Return to Elecraft
|
1 view|%1 views
| Free forum by Nabble | Edit this page |