Many thanks to SWLing Post contributor, David Day (N1DAY), who has completed a thorough review of the MLA-30 loop antenna. In this review, he compares the MLA to the Wellbrook ALA1530-LF and a 30 foot square ground loop antenna.
In short, he finds that the MLA-30 performs fairly well on the AM broadcast/mediumwave band, but can’t compete with the Wellbrook otherwise–especially in terms of noise floor. David also noted that “the MLA-30 had a very serious issue with IMD and in some cases, stations that were clearly heard with both the Wellbrook and the ground loop were totally obliterated by IMD when switching to the MLA-30.”.
Click here to read David’s full review on his blog, Ham Signal.
It’s not about reducing loop losses – it’s all about minimising the loop inductance.
Otherwise a typical 1m diameter loop will become self-resonant below 30MHz and / or the excessive loop inductance will result in the reactive part of the loop impedance dominating at the HF end which degrades the S/N ratio.
Changing from a thin wire loop to a ‘fat’ tube or metal strip can result in >10dB difference in S/N ratio at the receiver almost regardless of loop amplifier input impedance.
Loop amplifiers with a very low input impedance favour the LF and MF bands.
Using an amplifier with a higher value of input impedance, say a few 10’s of ohms, can improve the S/N on the HF bands, but also results in greater ingress of unwanted common mode noise and reduced null depth.
1. Clean up the power supply. Change the bias T if needed and use a regulated linear power supply; not a 5v USB-style switching wall wart. o.0
2. Replace the coax feedline. RG-58 or RG-6 should suffice for an upgrade, even for moderate length runs if need be. BTW, I tend to doubt the feed point is closely or maybe even vaguely matched to 50-ohms anyway.
3. Get a RF choke on the feedline near the antenna feed point. Ferrite(s) recommended, but even just wound coax would be a good idea at minimum. Also, whatever the choice, it needs to pass DC for the amp if feeding via the bias T.
4. Tweak the voltages and/or gain pot setting. For those attempting it, I suspect wide open on the pot is not a great idea, at least at higher voltages. YMMV.
5. Perhaps consider bypassing the bias T setup by feeding the amp directly. One user boiled the PCB to remove the potting, so one probably could work out where to scrap a couple of points to directly feed the amp board.
1) Yes, that’s almost certainly the first & biggest problem cab off the rank.
2a) Meh… 10M of cable isn’t going to make any real noticeable difference; < 1dB between RG-58/6 and RG-17x @ HF.Any random bit of RG-17x is also much more likely to be copper than RG-56/58/6/8 (these days mostly CCS/CCA, & lossier at lower freqs due to higher skin effect R). A bigger concern for me would be that SMA connectors aren't meant for multiple connect/disconnect operations.
2b) Define 'feedpoint' 😉 – the loop itself certainly won't be, but the amp output (if it's what it appears to be) will be pretty close to 50 ohms. But you're right; not that it's critical for a receive antenna…
3) I'm … not a fan of using chokes on the feedline (e.g. winding the coax through a core, etc) – I think the few hundred ohms of impedance it adds to common-mode noise is of marginal utility (especially in balanced/non-earthed antennas like loops), and at best masks the problem rather than fixing it. I prefer to side-step the problem altogether by other means (e.g isolating the coax & using a floating supply), but each to their own. In any case, you really only want/need one nearest the earth connection point e.g. the bias-tee.
4) Depends on the particular IC used. From what little can be seen in M0LMK's teardown, it *looks* like it's a LMH6550/clone op-amp/line driver. Regardless of what it is exactly, most actually perform best noise & IMD-wise just above their min supply voltage & with not-too-extreme gain. The general circuit configuration visible suggests the amp has fixed Rf/Rg, & the trimpot controls voltage – so 5~6v at the amp chip is likely to be optimum except in the presence of very large signals.
5) Always a good idea. Alternately, use a decent bias-tee – the simple designs favoured by the RTL-SDR crowd & copied by China are sub-optimal at even high HF frequencies.
Matt / M0LMK has a short post on his website where he gets under the epoxy coating. He doesn’t analyse it too much himself, but it looks to me like a fairly lazy design – loop, clamp diodes, what appears to be low-pass filters (or maybe band-reject – FM? – or possibly a poor attempt at impedance matching) in each leg, and straight into what, from the rest of the layout, looks to be a cheap cable-modem or video line-driver amp IC.
That approach can actually work OK for a tuned loop (as the fairly popular Techlib “hula hoop” & similar show) – but there’s a number of problems with that approach for broadband loops, mostly due to the poor impedance match to the loop itself and amp low signal performance. From experience, I’d expect it to be OK at lower frequencies and even have reasonably loop-like nulls, but for the null & noise to degrade rapidly up to 10-15MHz and flatten out from there. It’ll also be inherently noisy from the amp alone, and have poor-to-average dynamic range/overload & IMD.
Which, funnily enough, is pretty much what N1DAY found. Clean power may at least help with the noise somewhat, though I wouldn’t expect it to be great in any case.
(FWIW, loop conductor diameter doesn’t make much difference, provided it’s large enough that resistive/skin losses aren’t too high in the frequency range of interest. Loop conductor *width* (i.e. “front” to “back”), though … that makes a big difference to impedance, & therefore ease of matching across total loop bandwidth. All things being equal, a loop of 1/2″ tube will have ~3/4 the inductance of a loop on 14Ga wire. But a loop of 3/8″ x 1/8″ flat bar – about the same surface area, bulk, and resistance of thick-wall 1/2″ tube – will have about 2/3…)
For this type of amplified loop the wire diameter is irrelevant. These loops are typically around 150-200 ohms impedance at the feed point and input to the preamp and there is no need for a huge conductor like on a magnetic transmitting loop which is very low impedance.
W6LVP sells a loop experimenter kit based on the same preamp for his fixed amplified loops and I’ve used both thin wire and RG-11 for loops with this kit and there is no difference in performance.
“Not a fair comparison, the loop conductor is too thin and not intended for the final setup. It should be replaced with at least 1/2? diameter”
Mike , can you explain in details what you mean exactly for “this type of amplified loop” ?
Some loop designs are very low impedance, in the fractions of an ohm and large conductors are needed to avoid losses. Other designs like this MLA-30, W6LVP, etc, have a higher impedance at the feedpoint before the preamp and and there will be virtually no change in performance going from a thin wire loop to one made of say 2″ copper pipe.
My experience with very low impedance loops is more for transmitting where the main loop is tuned with a capacitor and usually excited by another small loop. For these, loop conductors are usually large tubing or a wide flat strap to minimize resistance and lower inductance. Amplified receiving loops like the MLA-30 are simply a loop of wire, which has a a fairly constant impedance over the VLF/HF range around 150 to 200 ohms, which is insensitive to changes in conductor diameter.
How do I do that? 1/2 …3/4 copper tube to replace the super thin SS wire?
Not a fair comparison, the loop conductor is too thin and not intended for the final setup. It should be replaced with at least 1/2″ diameter.
First replace the included power inserter with a generic one. Then feed the antenna with 12V from a linear power suply or battery pack. That should improve things a lot.
After that, try lowering the voltage for additional improvements. I have a Mini-Whip that is listed as 12V. IMD was a problem. Reducing the voltage in steps I reached 5V, and all is fine now. Even the seller now says that the antenna can be fed with 5 to 12V.
Would be interesting to see how it would perform with a better power source and voltage. I have a Chinese broadband preamp which the manufacturer says to run on 12 volts but I find it works much better with about 9 volts. Gain is lower and I believe IMD as well.