JRC "Japan Radio Co." NRD-345 HF Receiver

The discontinued "Japan Radio Co." NRD-345
(photo : N9EWO)

(Discontinued receiver)

We don't normally cover HF receivers that have been long discontinued, but decided to cover it here lightly anyway as I had a chance to use another sample for awhile.

The JRC NRD-345, Now a Rare Bird on the Used Market. AOR Japan ??

The now rare discontinued " Japan Radio Co. NRD-345 " had a short life on the market. With the data I have, it was in the JRC active line-up (made in Japan of course) between late April 1997 to April 2000. Actual REAL production dates I remember were much shorter than this (more like early-mid 1999 ?).

J.N. from Japan reported to me some time back that the NRD-345 was NOT manufactured by JRC. He says that it was made for them by AOR Co. Japan. However I do not have any solid information on this and we may never know the real truth on this one ? I do not think this rumor has any real truth to it.

General Stuff including 2 VFO's. Meter Lamp In A Socket.

Dual Up-conversion. The NRD-345 uses a 1st IF of 44.850 Mhz. The normal JRC 1st IF is 70.455 Mhz. If you take a GOOD look at AOR's old AR3030 receiver, the general software buttons/general operation are quite simular I will admit. Even the IF frequencies are identical.

Real HF coverage is from 10 Khz to 30 Mhz. 1st mixer uses 4 FET's (just as it is with the JRC NRD-535 and NRD-545). One IC DDS (Direct Digital Synthesizer), that allows down to 5 hz tuning steps. Tuning knob is very smooth and uses a rubber track around the knob, it the right size along with being properly weighted and uses a excellent optical encoder. The general feel of the knob doesn't get any better than this , even if it lacks any "feel" adjustment. The only real downer here to me is that it lacks 5 Khz steps either with the tuning knob or the up-down slewing buttons. So one has to make do with 1 Khz steps for general tuning around the SWBC bands.

2 VFO's and a real tilt bail with protective rubber tubes on this and real rubber feet on the rear as well. This is the only JRC RECEIVER to ever offer 2 VFO's or the tilt bail. I never knew how useful having 2 VFO's really was. Why this was never offered on other JRC receivers is a row of question marks ??

A solid steel cabinet as with all other JRC receivers. This is far better over the AOR AR7030's soft metal cabinet to me. We have an internal speaker, but as is the case with most tabletop sets it's pretty much worthless for any real listening (more on this later). Excellent front end filtering, image rejection, sensitivity/selectivity. All buttons have a very good feel and do not use any rubber or rubberish buttons, and support a solid feel. All are hard plastic and use tac switches. The downside is as just like on the NRD-545 most buttons are painted, and are prone to show wear with any real use. Here is a switch : The rear mounted external speaker and line output jacks are using 3.5 mm (1/8 inch) "Phone Jacks" , not the RCA phono type as with ALL other JRC receivers.

Rear view of the JRC NRD-345. The FAX jack is actually the standard "Line Output".
The RECORD jack is heavily attenuated for use with a MIC input on a tape recorder only.
Either one of these outputs has a distortion trait to my ears with SSB modes (see text).
(Photo : N9EWO)

You can enter frequencies on it's keyboard as Khz or Mhz, or even punching in the direct "meter band" numbers. It's so easy to use and no second function with the buttons (a few minor exceptions one being with the tuning steps, but this was done right). 100 memory channels which works OK for me, but many would find this to be a real drawback. Still uses a DB25 serial connector for use with a personal computer (RS232), and yes it requires a null model adapter or cable just like all other JRC sets that can be connected this way.

This is one extremely easy to use HF receiver, the ergonomics are first rate.

The LCD is properly lit with LED's with perfect contrast to boot. The meter is lit using a 6.3 volt 150 ma lamp. A bayonet SOCKET is used and not just direct wired in. This is so rare these days to see a socket used with any panel lamps ! It burns on the dim side too, so it should not require replacement too often.

Included NBB-429 USA Power Supply Is A Good One, Improved The Transformer Buzz.

The Japanese made external NBB-429 117vac power supply's transformer at first made a pretty bad buzzing sound within the room (output is very clean). Also with the first sample tested back in 1997, when the volume was turned up (loaded down), the panel lamp and LCD backlight dimmed with the audio peaks. However, I did not experience this bug with the sample I used for this report (it was from later production).

Important Tip : Be sure and keep the s-meter light bulb contacts and socket clean, more so with the solder "tip" on the bulb. If this is not done the LCD backlighting and S-meter lamp can still indeed dim with audio peaks !!

It turned out that the buzz was caused with the internal power transformer not being properly seated in the 2 grooves in the bottom of the plastic case (it pops in quite hard). Once this was done, this buzz issue was down to a improved level even under normal load. However it was never as quiet as one might hope for.

Better yet, unlike the Yaesu FRG-100 (and others), this is a "regulated power supply" floor wart and is rated at a 800ma output. It uses a 12 volt Sankyo SI-3122V linear 2 amp IC regulator on a large heatsink, which is all good news (the 3122 is in a TO-3P package, see picture below). 4 bypass disc ceramic capacitors are provided across the bridge rectifier block as well to tame any switching noise from these diodes. 2 actual screws hold it together (not just glued together). It too is made in Japan.

A very stiff 3 wire cord is used and the ground "green wire" from the power cable is connected to the ground on the power supply PC board , see the picture below. So a word of warning here if you have any "audio" ground loop issues as this could make it worse. This power supply runs a bit warm if not near hot when used for awhile. The receiver however runs very cool, even being on for hours.

The included NBB-429 power supply is of a linear regulated type (at least the USA version I used was).
It's power transformer buzzed a bit under load but this was improved with the 2nd tested sample (see text above).
(Photo : N9EWO)

A crude looking schematic of the NBB-429 power supply I drew up.
This is one you will not find in the NRD-345's service manual.
(photo: N9EWO)

Real mechanical S-Meter, 2 markings over the production.

During it's production run JRC had changed the printing on the numbers on the S-meter on mid-later units. The printing on the later meters were on the lines that were broken between the numbers (the bar below the numbers), where the early units had continuous lines (bar). Who knows if any other changes have been made inside ?? Yes, the early version appears to be the same markings (if not the same meter) as used in the old AOR AR3030.

2 Line Output Jacks, not 2 in the same however. Either fixed "audio output" has distortion to my ears on SSB modes.

On the rear of the NRD-345 we find two 1/8 inch (3.5mm) phone jacks for a "line" audio output. This type of jack being used here is strange for JRC as on all other consumer models this has always been RCA-phone jacks. But it gets a bit more weird (also see picture above of rear panel).

These jacks are marked as FAX (line out) and the other as RECORD. The FAX (line out) jack is at a perfect level for input to a consumer recording device or connection to another amplifier or even amplified speakers. The RECORD jack is at a very low level and is good for direct connection to a MICROPHONE jack on a tape recorder. This RECORD jack sees a 22K resistor in series and after than a 680 ohm in parallel (attenuation) vs. the FAX one. They do indeed share the same buffered fixed audio output. Additionally there is no way to adjust the output of either of these (no internal level control at all).

Sadly, we discovered an annoying amount of audio distortion from either the lower level RECORD or the higher level FAX (line out) jacks on SSB modes. For most users this may not be an issue (who cares right ?). But as I make serious recordings from a receiver, it was a no go here I'm afraid for me.

Audio Quality is OK, So-so Sync.

Using the narrow AM mode "Wide" IF filter, audio quality suffers as it lacks any high end audio frequencies. One can tune off a Khz or 2 and improve this a bit without adding too much distortion (it does not to my ears anyway). SSB modes par much better and overall I feel is better over the AR7030 for any quality listening to SSB stations. Manual ECSS works EXCELLENT as long as the station is on top of the 5 Hz tuning steps, additionally there is no hiss at all. After connecting to a stereo receiver amplifier , I noticed a bit more bass response that you don't normally hear with a communications receiver (however see the SSB "line out" distortion issue above). If this is a plus is going to be personal thing but is the most bass I have even heard to come out of a JRC receiver.

Just as we have experienced back in 1997 with a sample we had a chance to play with briefly, the "double sideband" only Sync Detection is limited. It can actually add a bit more distortion depending on the signal. But on the other side of the coin, once in awhile it can help to reduce fading distortion (again depending on the signal).

External Speaker a Requirement, Rated 8 Ohms (not the usual JRC 4), No Muting Function for Use With a Transmitter, Missing Squelch Control.

As it is with most tabletop sets a GOOD external speaker is a must. This is even more important with the NRD-345. The internal speaker is downright awlful. I will say choose one that does not give added bass response as that will make the "woofy" sound even worse. With all other JRC sets I have touched over the years the speaker rating is at a 4 ohm load. Not so with the 345, it's rating is at 8 ohms.

Lots of punch to the audio, so one does not have to worry about a whimpy output here. If one fully turns down the volume control, I was able to hear some audio left anyway more so in a totally quiet room. Again, I must stress the the distortion issue with the Record and FAX outputs as covered above, if that is important to you.

The volume control not is microprocessor controlled like with the NRD-545. It's adjusted via the good old fashioned analog method.

There is no "Mute" function or squelch, but I did not find either of these missing to be a drawback.

Feedback "Beep" Too Loud For My Taste, Owners Manual Has Typo For "Frequency Reference Adjustment" On Page 25.

As it was with the NRD-535 and yes even with the NRD-545 the keyboard feedback "Beep" was much to loud out of the box for my tastes. But as with the other JRC sets, there is a way to adjust this volume or to just shut the beep off. However, I would like to have some keyboard feedback just not this loud. In the case of the NRD-345 for adjustment of this this level is done with the internal trimmer resistor RV7. This is located on the "bottom" CAE-368 board on the rear edge in between the FAX jack and the DB-25 computer port. It is very well marked on the board (you can't miss it).

For any slight frequency display errors, the reference oscillator trimmer capacitor can be tweaked. I will say after a 3 HOUR warmup with the cover(s) on. This internal CV3 trimmer capacitor is found on the "top" CMA-648 board, where the control is also located along the rear edge under a metal cover.

Page 25 of the owners manual indicates this trimmer is marked as CV0, well this is a downright typo. No such thing as a CV0 even in the receiver at least on this 2nd test sample. I would use WWV at 15 Mhz to do this, USB mode with the widest bandwidth filer. As it goes with these type of adjustments, the higher you go in frequency the more touchy this is. So if it's off at 10 Mhz, it will really be off at 15 Mhz as so on as you go up. CV3 is also well marked once the metal cover is removed (it just pops off).

I must pass along the usual WARNING here. This is a very touchy thing to get right and may take many attempts. Pass on this one if you are not good with these sort of adjustments. Also being this is NOT a high stability type of crystal being used here (and never was an option for such), you may never get it to stay quite "dead" on. It might be 1 or 2 Hz off anyway you slice it depending on room temp. But this is not a drawback at this price point to me and actually has excellent stability.

20 db MW Attenuator (but can this can be removed ?)

The NRD-345 uses a "extra" attenuator in the MW band to help deal with dynamic range. This is not to be be confused with the regular ATT function (not the same beast).

Looking at page 12 of the NRD-345's "Service Manual", it indicates an additional "20 db ATT" with the low pass filter on the MW band between 540 Khz to 1.8 Mhz. If you want to see the change with it bypassed (without doing any modifications at all), tune between 539 Khz and 540 Khz. Boom, you can see how much signal is actually lost.

As many already know this additional 20 db attenuator is not defeated with a button press either. One who is handy with surface mount electronics should be able to remove the 3 SMD parts on the top CMA-648 board ,and install the required 2 zero ohm (000) SMD jumpers of the right size. However, I have NOT done this (and don't plan to) and if you mess up or this information is bunk, it's YOUR problem. See the picture below. The finished modification photo below was NOT taken by me (this picture source unknown).

Needs additional wider voice bandwidth filter, adding a "Defacto" one to the "AUX" spot using the CFQ-8673 option board ?

The NRD-345 desperately needs an additional wider filter for AM mode signals to help with audio quality when one could use it. There is a option for another filter using the CFQ-8673 plug in option board. However this is made to work with JRC (NTK) huge crystal filters that are were all on the narrow side anyway, so again useless.

Optional service manual does not give details of the CFQ-8673 "Aux filter board option" components or values of those components (not even a simple schematic). On this board there is a series variable coil , a SMD series capacitor and one more SMD capacitor connected in parrell , all repeated on the other side of the filter board.

Looking back at the schematics from the NRD-525 and 535, JRC provided a "Defacto" super wide filter if one did not install the optional filter(s). This was actually no filter at all and was bypassed using a jumper for a nice wide filter setting.

So I was thinking why not do this with a the open AUX spot on the NRD-345 (using the CFQ-8673 Aux board to make it easy) ? However, unlike the older NRD-525 and 535 models, the 345 uses a 6 Khz "IF Tail Filter" at the end of the IF chain (but measures more like 7 to 8 Khz). Why even bother with a filter for a wider bandwidth anyway with this tail filter in the chain.

Following the idea from the NRD-525/535, I went ahead and did the modifications to the CFQ-8673 plug in board and it worked 100 % perfect. For the details on how I did this see the picture below. It requires the "techie person" to remove all of the SMD capacitors and coils on the CFQ-8673 board and then adding 3 jumper wires and 2 SMD 000 jumpers. Mind you this only adds a couple of Khz of bandwidth (perhaps up to 7 to 8 Khz ?), but with the case of the NRD-345, this really does help with the audio in the AM mode. It can be pulled out in a flash of course and no modifications to the actual receiver at all.

Before (top 2) and After (bottom 2) Pictures Retro-Fit of the CFQ-8673 "Aux Filter Board Option"
See text above for more information. Sorry, I will be unable to do any retro work.
(retro-fit and photo's : N9EWO)

A Nice Receiver Overall.

I feel this is generally a good receiver overall. It's a more useful set for use with SSB uses (Manual ECSS,Hams,Utes,etc.). Adding the wider "Defacto" bandwidth using the CFQ-8673 "Aux Board" does help the audio in the AM mode. But one needs to be aware of the distortion issue with the Record and FAX outputs (SSB modes) if you do any serious recording.

Of course one will have to hunt on the used market now as any new samples of the NRD-345 have been long gone for years.

Dave N9EWO
© 2008 N9EWO
Ver. 3.5
(Discontinued receiver)

Internal pictures of the JRC NRD-345.
The meter light uses a 6.3 volt 150 ma bulb in a standard bayonet socket. The LCD backlighting uses LED's.
Strangely that these two are wired in series with each other, so if the meter light burns out, the LED backlighting goes out as well.
Important Tip : Be sure and keep the s-meter light bulb contacts clean, more so with the solder "tip" on the bulb. If this is not done the LCD backlighting and S-meter lamp can dim with audio peaks !!
(Photo's : N9EWO)

JRC NRD-345 Full Color Brochure (in pdf format)
(via Paul Lannuier's web page)

Back to Home Page