Receivers for LF

My comments are added in italics.

Receiver performance on LF

My personal advice is to go for the TS-850. It has a very clean DDS, which cannot be said for the TS-930, and its sensitivity in the LF band is second to none, at least among the commercial Ham transceivers.

Alberto I2PHD


I would recommend getting a good filter for the 455 IF, it really makes the receiver so much better. The skirt selectivity of the 8MHz filters is not as good as the 455 ones so you will notice an improvement there, in addition the fact that you will have two CW filters means that you will be able to use the bandwidth controls properly to roll off either the HF or LF side of the passband. The best idea is to have two filters of the same bandwidth but a 400 or 500Hz filter at 455 will still be very useful in conjunction with your 270 at 8.8. Try International radio for a good range of filters at slightly less money than the Kenwood ones.

Their web-site is at

Dave G3YXM.


Concur Dave's advice on the Inrad filters. Their 125Hz @ 455kHz filter, although it might seem ridiculously narrow by 'normal' standards is perfect for LF use. In fact it's super for conventional CW, PSK31 and such at UHF (topband and above) too.

Steve, W3EEE


Talking of TS-850s, does anyone know why my TS-850 changes gain at 500.000 KHz? Seems like an attenuator is switched in above 500 KHz. Is there a mod to defeat it? The TS-850 is manufactured with an attenuator for the MF AM broadcast band (included within that band pass filter).



The TS-850 is manufactured with an attenuator for the MF AM broadcast band (included within that band pass filter). The attenuator is made from surface mount resistors, on the underside of the PCB, so it is a fairly tricky job to rework. If you do a web search for TS-850 mods, you will find a text and figures as to how to remove the attenuator.

73, Bob ZL2CA


Yes, that is typical for Kenwood, probably they fear their rigs get saturated in the MW band.... when modifying my TS-950SDX to improve its sensitivity in the LF band, as described in a past message, I eliminated that T-network attenuator. In my case it was rather easy, as Kenwood had kindly provided two test pins (???), just before and after the attenuator. So all it was needed was just a jumper. It looks like they had foreseen the need to bypass it. I have noticed no ill effects after its bypassing, probably in Japan they have much stronger MW transmitters than here :-)

73 Alberto I2PHD


José de Sá asked

What about the TS-870 or the Icom IC-746 ?

According to the Kenwood specs, the TS-850 is rated at 0.2 uV for 10 dB (S+N)/N at 100 kHz.

The TS-870 has a rating of only 1uV in the same conditions.

Don't have numbers for the IC-746, but a friend OM that owns one tells me that it is deaf as a doorknob on the LF band.

73 Alberto I2PHD


I once had the chance to compare an old IC735, an IC706 and a TS850 running side-by-side connected to a tuned LF-antenna (150m long wire).

Results: - TS850 with 500Hz and 270 Hz filter: Could hear the grass grow and OH1TN call CQ on 136kHz. A fine rig, as already expressed by others (and in the LF handbook). Considering the price I once paid for it, and for what it can be found nowadays on the second-hand market, I would almost buy a second one as a spare unit!

IC735: quite deaf, some IM products audible which were not there on the TS850. But still a nice rig on 160m and HF.

IC706 ( the 'old' one, no MK2, with 270 Hz filter): Very deaf on 136kHz, no matter if the preamp is on or off. Turning the VFO knob OR THE VOLUME CONTROL results in a hissing sound, which stops after the motion. Looks like the rig picks up some noise from its own serial interface between the front panel and the microcontroller ! Hopefully the IC706 MK2 ( & -G) are better at 136kHz. At 77kHz, I cannot even hear the DCF time signal transmitter near Frankfurt.

Some years ago, I compared the TS570 and the TS850 on HF for CW. There were severe blocking effects in the TS570 in a heavy crowded CW segment, because there is not enough selectivity in the early stages in the IF amplifier chain. The S-meter went up to S-9, but no signal could be heard because the DSP decided it was outside the passband. And, with modestly narrow CW filter settings the audio from the TS570 sounded like a block flute, which is not very pleasant. Maybe there was something wrong with the DSP. We never did further investigations because the owner of the TS570 immediately traded it for a TS850...

Wolf DL4YHF. _


A couple of months ago the "which receiver?" thing came up over here, too. The attached below is a message I sent one of the guys (he was toying with getting a TS-440S, with which he is reasonably pleased); I'd just got a TS-140S as an LF-specific radio (really cheap) and measured what I had to hand. The following evening I measured an IC-735 (a radio I really like - probably optimum radio-per-dollar) which is excellent 160 - 10, but laughably deaf - even with the BCB attenuator removed - at 136. AGC threshold of about 60uV, if I remember. A friend's 746 was similar.

If anyone's interested, I'll get hold of a 756 (which has the same front-end as the PRO, too) and 'measure' that.

One issue that hasn't come up in the discussion so far is that of stability and the means of improving same. Old radios which otherwise rock (such as the IC-751a) have too many (i.e. more than one) crystal oscillators that affect stability; mine has been 'retired' for that reason. By way of contrast, even the bottom-of-the-line TS-140S has a single reference, which cooks up nicely.

Attached message

Well, I got to play with the TS-140 this evening. Thought you might like to see these numbers:

.                 IC-781     IC-715         TS-140S

Freq           Required signal for 'S9'

3.6MHz     56uV         45uV             50uV

200kHz     56uV         80uV             160uV

        AGC threshold

3.6MHz     2.5uV         2.7uV         1.7uV

400kHz     0.9uV         1.1uV         1.1uV

200kHz     1.1uV         1uV             1.8uV

135kHz     1uV             1uV             5uV

100kHz     1.25uV         1.3uV         15uV

The oscillator I was using leaks too much to do a proper Minimum Discernible Signal measurement, so I thought that agc threshold would tell a reasonable story.

The '140 holds it's own with the 'big boys' very well; the very rapid drop-off at the low-end is a mystery - it's too rapid for just general giving-up-the-ghost; it's almost as though it's the edge of a notch there or something. I'll track it down eventually.

Yes, the '781 is best, but it should be!

The '140 chucks in an almighty attenuator between 500 and 1600kHz (for good reason) and then very nicely removes it below 500kHz.. The Bad News is that the (not especially strong) front-end makes a pig's ear of all the now un-attenuated broadcasters - intermod everywhere. A low-pass filter or some other selectivity is essential to use it on LF. Good news. I threw in a sloppy thing-in-an- Altoids-tin filter and it breathed a sigh of relief.

Then, just for good measure, I opened it up and chucked in the IRC 125Hz filter out of the 751a (such as the one you had). . .


The Yaesu FT-767GX does not appear to have LP filtering for medium wave band, tunes well below the 100 K mark and is very stable. Does anyone else have experience using this rig for LF receiving?

Mike VK2ASR _


The fact that a RX has a 'lazy' S-meter doesn't mean that it is a poor RX. A correct S-meter is nice to give exact reports, but what really counts is the sensitivity and good IM behaviour of a receiver.

Rik, ON7YD

Lacking direct experience of that rig, I used the available data to try to formulate a thumb-estimate. Ideally, a good S-meter should occasionally flicker on the noise produced by a 50 ohm resistor at room temperature placed across the antenna input. The AGC action should however enter into play only after a given threshold. But this would be to ask too much to a receiver. After all, they aren't measurement instrumentation... I think the only calibration the manufactures do is to set the S-meter indication at S9 with 50uV input at 14 MHz. Any other reading is totally unreliable.

Alberto I2PHD


Thank you all for your comments and advice. It made it much easier to make a decision. I got a TS-850SAT with 270Hz filter on the 8.33MHz if. So maybe I have to invest in a second filter for 455KHz. Its great to see how a single question creates not only useful answers, but also generates some discussion about the topic.

Ulrich, PA7EY


Iīm close to buy a WG SPM-33A selective level meter (50Hz to 2MHz). Itīs an SSB receiver after all, with IF BW of 25, 1700 and 3100KHz, balanced input (75, 150 and 600 ohm) and a unbalanced 75 ohm one ... sensibility from -120dBm to +20dBm. It has internal ni-mh batteries, so I think itīs good to portable operations as well.

Did anybody uses a selective level meter as a LF receiver ? And about this specific WG meter? The manufacturer says that the demodulator is "SSB with built in speaker" but seems that Iīll recover audio from "DC" to IF BW - no offset :o(.

Does anybody know this "rig"?



I use the selective level meter MV62 very successfully for LF reception, I also have a Siemens unit (see There is also an article on the internet about the general use of level meters for reception (although not about the unit you mentioned ...), for details see

Geri, DK8KW (W1KW)


Hereīs some information on the WG SPM-33A Selective level meter I finally received and testing now. May be usefull for someone that find a SPM gear around say SPM 32, 33 or 34 - the hand held meters.

The SSB demod seems marginal, I mean, need to adjust speaker volume for each signal received. Ok, thereīs 120dB range there but seems that some sort of internal generated noise and harmonics show up on certain volume points. I disconected the internal speaker and installed a mini-P2 jack so there is provision to wire in headphones, speaker or PC line.

Using DL4YHF software (thanks sir) I created some "receiver mode" template to translate the 2KHz zero beat tone to a more confortable one (800Hz to my ears). This means use software to receive 2KHz.

The WG, when in demod mode, donīt show the received level: or demod or measure. It alternate between USB and LSB by pressing the demod key. The IFs BW are 25Hz, 1.7KHz and 3.1KHz, centered on 2KHz zero beat tone.

Can run with two common 9V batteries (LR622 etc) or the rechargeable version. 6 hours on common batteries, 8 hours on NiMh. Or external PS from 13 to 15 Vdc. Good for mobile/field day ... the balanced input directly atached to a cooper loop with 4.8m perimeter and a 220nF capacitor across loop terminals tuned to 163KHz, enough to detect the French BC on 162KHz with itīs time signal. The same signal I got using the 33m high vertical run using the 75ohm input.

Thereīs 2 antenna inputs: a BNC one (75ohm or infinite - 6Kohm) and a balanced one (75, 150, 600 or infinite)

Internally seems a simple equipment despite SMDs. I hope I can got some service information to recover IF somewhere and use an external demodulator, may be with an DSP56002EVM or sound card software or change the beat tone, or recover audio before the volume control while in the measure mode, etc.

On the air tests: 18.2KHz VTX3 Morse any time of the day using 80m dipole inverted vee + 33m feedline as a capacitive hat vertical antenna. And at night several LW BCs detected, + WWVB on 60KHz, MSKs on 37.5, 40.4 and those between 16 and 25KHz. The 82Hz test, with the 25Hz IF was not good since 60/120Hz are too strong for a home instalation. May be with some active filter help behind antenna I can do it.

Anyway, stable and self calibrating. I think this surplus 70 dollars rig is very good

Marcus, PY3CRX/PY2PLL,



IC706 on LF

Regarding the IC-706, I was also quite taken aback by the "chunder" that could be heard with no signal at 136k, this is probably the PLL reference. It is something I do not hear on the 7030 (even at the 10nV level) despite DDS LOs not being famed for their quietness.

Alan G3NYK

DDS LOs are extremally quiet - there is no added phase noise due to the synthesizer process and the phase noise you do get is that of the clock divided by the frequency synthesis ratio. There can be spurious frequency products at well defined levels, but they can usually be dealt with by filtering and proper frequency planning. You are probably thinking of PLL synthesizer phase noise

Andy G4JNT


IC706 ( the 'old' one, no MK2, with 270 Hz filter)

Very deaf on 136kHz, no matter if the preamp is on or off. Turning the VFO knob OR THE VOLUME CONTROL results in a hissing sound, which stops after the motion. Looks like the rig picks up some noise from its own serial interface between the front panel and the microcontroller! Hopefully the IC706 MK2 (&-G) are better at 136kHz.

At 77kHz, I cannot even hear the DCF time signal transmitter near Frankfurt...

Wolf DL4YHF.


Wolf seems to be unlucky with his IC-706. I have a Mk1, which is a little deaf on 136kHz, but adding a one-transistor pre-amp makes it a very good receiver. I experience no interface noise when using the outside antenna, though a few years ago when I did a lot of 73kHz portable work, I noticed that the front panel display radiated LF for a few centimetres. The synthesiser on the 706 is excellent and during last winter's transatlantic tests it stayed within 0.1Hz for hours at a time, and once it was calibrated with Argo and BBC 198kHz it could be re-set reliably day after day.

Mike, G3XDV


Re 706s, I have found my nice new IC 706 GII to be a dead duck for 73 KHz receive. I agree with Mike that it is usable on 136 with a bit of assistance, but not on the lower band.

Not hearing much on one band and *!*!*! all in the other, I decided to do a proper test. If anyone is interested, the CW 10dB S+N/N in the narrow filter (270Hz) checked using a TF2008 sig.gen. (admittedly not recently calibrated but within a dB or so against a known source) at the two frequencies was -

136 KHz 10uV p.d.

73 KHz 83uV p.d.

Now unless I happen to have a particularly dud sample, it would seem that there is a pretty good HPF in there.



Lots of info around about the 706 and different versions. I have a late MK1 model and it works FB and I use it as a generator source to drive my 73/136 PA. I use it on 7.3/13.6 MHz low power and divide by 100 for the appropriate LF bands. It is very stable on transmit and works well also on receive 136kHz but I have a large antenna. On a small antenna it is not sensitive enough.

I compared it with other rigs like the TS50 for example on a short antenna in the early days and it was not very good. When I put the big antenna up then it was more than adequate with the pre amp on. The TS50 was too sensitive on the big antenna and needed external attenuation. I have other receivers and all are very good but it depends a lot on the antenna in use as regards sensitivity. What would suit one persons needs would not suit another. Using the 706 with my present antenna, the RF gain has to be backed off slightly to obtain max signal over noise but has more than adequate sensitivity. The antenna will determine what receiver suits your needs best.

Mal, G3KEV

It is interesting to see the differences in opinon regarding the using the IC-706 for LF. The IC-706 is not exactly regarded as a top-of-the-range receiver, especially for this application. However, there is no doubt that G3KEV is obtaining very good receiving results with his big antenna and IC-706 combination. Even more interesting is the use of a similar receiving setup used by VA3LK on his distance record QSO with G3AQC. For this QSO G3AQC was running 350mW erp and VA3LK approximatley 3 to 5W erp.

G3AQC. I have just been looking at the pics you sent me. They are excellent especially the later ones, since my power was relatively low. I wonder what receiving antenna you were using ?

VA3LK, 100M semi vertical wire with bandpass filter and band reject filters as well - same antenna as for TX.

G3AQC. The traces seem so clear, no QRN! was it the TX antenna or something like a loop or E probe ?

VA3LK; Clear, yes brute force filters and national political connections to get the hydro carrier current crap turned off!

G3AQC. I gather you were using an ICOM 706 and ARGO software, was there anything else?

VA3LK. No just brutal politics and very good filter stuff I borrowed unofficially. The filter stuff so we dont say much about that of course. The politics were interesting, I had to really get rough to get the carrier current stuff turned off - this was only possible as I was able to prove that they were not using the stuff, it was just sitting there running and making a mess on 136 and area.

G3AQC; I am still amazed at how well you received my signal which must have been 10db down on yours. I know you don't want to say too much about the filters etc, but is there any more information you can let me have, for example did you use any pre-amplification in front of the 706 ?

VA3LK; Yes certainly, a Burhans as well as the gain in the filter system. I started the fall of 2000 to experiment with determining what was really on the frequency versus what the receiver put on the frequency - there is a difference. I borrowed some "good stuff" to really put band pass as well as frequency reject, it was amazing what went away as I got better at it. Big thing was to get rid of Loran and second was to get rid of some powerful signals not far from 136, when that was done and the carrier current was turned off it was amazing what was left, mostly amateur signals or others that were really on the frequency.




New Receiver Technology

I am now putting the finishing stages to a custom LF receiver, designed to be a useful accurate measurement tool as well as for operating. The design is optimised for narrow band and ultra narrow-band working, but with an audio bandwidth suitable still for CW. An A/D converter is included as an integral part, with formatted data sent to a PC via the serial port making soundcards, with their poor frequency stability and difficulties in driver software, unnecessary. ALL frequencies and A/D timing are derived from one master oscillator which may be an onboard TCXO (< 1ppm) or a highly stable external 5 / 10MHz reference.

The basic receiver line up is as follows :

A 7th order LP filter cutting off at 150kHz to give greater then 100dB rejection of the image frequency in the MF broadcast band. LPF cutoff frequency could be higher, but we have very high power LF broadcasting in Europe and I didn't want to tempt the possibility of overloading if close to a transmitter. MAR-3 modamp RF amplifier driven at constant current supply for temperature stability and interference rejection at low frequencies SRA-8 Double balanced mixer, upconverting to 455kHz IF Low phase noise LO derived from a DDS module over-clocked at 10MHz, with extra filtering to eliminate spurii Automatic External / Internal reference switching, with frequency doubling of a 5MHz signal if needed 300Hz bandwidth mechanical filter at 455kHz IF gain provided by an Analog Devices AD603 prescision voltage controlled gain stage, from a manual gain control pot.

This chip is laser trimmed to provide 10 - 50dB gain with 0-1V and enables relative gain to be determined very accurately, to a fraction of a dB by just measuring the applied voltage with a DVM. AGC is not an option in a measurement Rx ! A BFO at 456kHz derived from the 10MHz reference mixes the IF to 1kHz centre frequency. Audio Bandpass filter at 1kHz reduces the bandwidth to 100Hz 12 Bit A/D converter sampling the 1kHz audio at 4kHz, generating 14 bit Complex I / Q pairs at 1kHz sampling rate, centred on 1kHz. ie a 1kHz tone appears in the data as if it were at zero frequency with up to +/- 50Hz sidebands

Sensitivity of the receiver is sufficient to be used directly with a 0.8m loop (at least for urban noise levels) without having too much gain that the IF has to be wound back too far when connected to a full size antenna.

The receiver hardware is working perfectly, but at the moment tuning is only under computer control of the DDS and I still need to add a manual tuning mode. This will be 10 turn potentiometer which, via an 8 bit A/D, can set 0 - 255 frequency increments from a base value. Channel (increment) readout can use the same DVM module already in use for gain setting, by measuring the voltage directly. Both Base frequency and Increment will be programmable from a PC and stored in non-volatile memory in the PIC controlling the DDS, ready for instant turn on. Doing it this way means the simple addition of the A/D chip to my existing DDS module with new PIC software.

By setting an increment of, say, 10Hz, the whole 137kHz allocation can be covered without reprogramming, as well as most of 73kHz for aural CW listening. Alternatively the DDS LO could be replaced with the other designs around that make use of rotary encoders and LCD displays.

When finished a write up will follow which will hopefully be published in a magazine.

This is not a cheap low cost receiver just for routine operating, but has been designed to be almost a laboratory grade piece of equipment useful for precise frequency and amplitude measurements.

Andy, G4JNT