The National NC-303 is a real performer for its time. I would prefer the NC-303 to the HQ-170, in spite of my love for Hammarlund receivers. I would also rate it at better than the Hallicrafters SX101. I discuss the comparison of thes three classic receivers at length in my article on the HQ-170: /Projects/Receivers/HammarlundHQ170-180.html.

Unlike many of the cheaper National receivers, which delivered a lot of performance for the price, this one is heavily braced and has a very rigid thick front panel for good mechanical stability. It uses high stability ceramic coil forms for the oscillator, not cheap paper forms like those used in the HQ-170. The dial calibration accuracy and length of the dial itself is better than its contemporaries. It is dual conversion with the first IF at 2215 KC and the second at 80 KC. It has excellent sensitivity on all bands and superior selectivity. For AM it has an 8 KHz wide bandwidth position (presumably compatible with its VHF converters) which sounds beautiful when tuned to a Class E rig. (OK, it is not an R390 with a 16 KHz filter, but we are comparing similar cost brackets here. It is way better than the HQ-170). The CW can be operated in a narrow selectivity or a wider one. It can be operated without AGC in the CW position or with AGC in the SSB position in the narrow 400 cycle, or a wider bandwidth as desired. There is a product detector (that works) for linear detection of SSB. The AGC time constant is a little short by today's standards, but you can add a larger capacitor if you want. The S meter even works correctly! I measured it when I had the equipment set up on the bench to do an HQ-170. Each S unit is about 6 dB, and the scale above S9 is pretty close, until you get to full scale. As a bonus, the NC-303 low IF can be aligned much easier than an HQ-170, unless it has been the victim of major tampering.

Note that it is possible to have foreign broadcast images from around 24.5 MC on 10 meters. These are the only spurious responses I noted while testing this radio. Some have reported 9 to 10 MHz signals on 20 meters; I have a fix for that later in the article. The NC303 in the broad position works great on the top end of 10 meters to listen to the FM activity there. You tune off to the side and use slope detection in the broad position, just the way the old buzzards did. On the function switch, there is an ACC or accessory position to add a proper FM detector. Hmmm, maybe a synchronous detector? I feel a project coming on. The VHF converters teamed up with the NC303 made such a nice setup back when these bands were just beginning to gain popularity. If you were into 6 or 2 meters, this NC-303 would run rings around even the HQ-170A-VHF model. Granted, it cost more than the HQ-170. The SX-101 never offered VHF. These days, most are interested in these classic Boat Anchor receives as an HF only setup.

TIP: Note on page 7 the vernier next to the main tuning knob is for fine tuning of SSB signals and CW signals on the upper bands. This small knob is pushed in to engage the the main tuning knob for fine tuning. NORMALLY this knob is pulled OUT to allow smooth operation of the main tuning knob.

	To remove the chassis from the cabinet, here are some tips:

	Set the bandswitch to the 220 MC position to prevent damage to the frequency dial.

	These screws catch on the lip of the cabinet. You have to gently pry the lip while the radio is on its side to release them. They also will scratch the visible edge of the cabinet when the chassis is slid out, if you are not careful.

	This is the top view of the chassis after cleaning out some sawdust residue. Note the installation of the fancy crystal calibrator which includes a WWV receive function.

NC-303 CRYSTAL CALIBRATOR ADJUSTMENT

	This coil is used to set the WWV receive frequency to the RED MARK WWV on the 40 meter band. It is a 3 MC oscillator which converts 10 MC to about 7 MC.

	This coil peaks up the receive sensitivity on 10 MC receive after you have the other coils set. You should temporarily unplug the 6AK6 oscillator tube to get only WWV. Note that the antenna must temporarily be attached to the black binding post marked ANT to receive WWV.

	Once you get WWV all set up, this trimmer capacitor sets the 100 KC calibrator to zero beat with WWV.

S METER ZERO ADJUSTMENT

	This is the S Meter ZERO adjuster pot. The manual describes the procedure, but does not identify its location near the 5Y3.

BOTTOM CHASSIS VIEW

	This is a full view of the bottom of the chassis. Note that I have replaced the two square mica capacitors from either side of the line that had ruptured due to excessive line voltage. I used 2 KV ceramic capacitors.

	This is a detail view of the 80 KC IF section.

	This is a detail view of the chassis near the antenna connector and the RF amp compartment.

	This is a close up of the second converter and its crystal. The first IF is at 2215 KC. The crystal mixes with that frequency to get down to 80 KC.

	This is a detail view of the power supply wiring.

IF ALIGNMENT

The IF alignment is fairly straight forward. Just follow the instructions in the manual. No sweep generator needed. But you need an 80 KC signal generator. You can use most audio generators to get an unmodulated carrier. BE SURE TO USE A DC BLOCKING CAPACITOR TO PREVENT SHORTING THE RECEIVER BACK THROUGH THE SIGNAL GENERATOR! Also, be sure to use a frequency counter to get the audio generator exactly on frequency.

NOTE THAT THE MANUAL IS IN ERROR CALLING FOR ADJUSTMENT OF T8 FOR THE BFO. IT IS ACTUALLY T6 WHICH IS IDENTIFIED ON THE PICTORIAL IN THE MANUAL.

RF ALIGNMENT

This part of the process is more challenging.

First, you need a very accurate signal generator to work effectively on high performance receivers such as the NC-303. I like my solid state B&K E2000D. It has internal 1 MC and 100 KC crystal calibrators, which allow very precise adjustment of its main tuning dial. It is also very accurate in calibrated output level, which can be useful in knowing the exact sensitivity down to the microvolt level. With a high performance receiver like this, you will need an external step attenuator box to get the level down low enough for a true appreciation of what this receiver can do.

These capacitors for the oscillator and dial calibration should be all the adjustment you need to use to get the dial back in range. Note that they are high quality air variable capacitors. These are much better than the calibration capacitors in the HQ-170, explaining its stability. Be aware that these capacitors have sufficient range to get you way off frequency on the upper bands. Be sure you do not tune it up on an image frequency.

The associated coils are all high quality ceramic individual coils. The HQ-170 has cheaper dual coils which means you have to exercise extreme care to avoid adjusting the wrong coil during alignment. I suggest you try to avoid adjustment of the NC-303 coils unless tracking is way off. There is no slug to adjust. If you look directly into the coil, there is a half turn of wire. Using an insulated tool, carefully rotate the half turn coil in the center to get the adjustment right. Do not do a lot of this to avoid breaking the wire. YOU SHOULD BE ABLE TO GET ACCEPTABLE CALIBRATION WITHOUT THE ADJUSTMENT OF THESE COILS UNLESS SOMEONE HAS BEEN TINKERING.

This is the mixer alignment adjustment capacitor array. Use the manual to identify the correct capacitor to adjust for each band. The orange pointer in the photo indicates a TRAP to prevent 9 to 10 KHz foreign broadcast (out of band) signals from getting into the receiver on 20 meters. DO NOT ADJUST THIS CAPACITOR, UNLESS YOU HAVE THE EQUIPMENT AND KNOW WHAT YOU ARE DOING! If you are experiencing interference from 31 meter broadcast, a passive preselector will help because it adds another tuned circuit ahead of the radio. The too small tuner you bought as a Johnny Novice will also work, between the T/R switch and the receiver; it does not have to handle the transmitter power that way. (A T match circuit is a 2 pole high pass filter.) Or, to get fancy, use a MFJ 1048 or MFJ 1020 (save money, you do not need the T/R transceiver model) preselector or one of the older ones you got to soup up your S-38, as long as it is tuneable. An MFJ 956 will not work well enough here. This is a passive device (good enough in many cases, given our high ambient noise from lighting and computers and high tension AC lines). More importantly, the MFJ 956 has only one series tuned circuit. Download the manual and look at it; over $60 for that? Oh well. I use that technique in my HQ-170 to resonate the input coil with about 1000 pF to 1900 KHz. This will not work as well on 20 meters. I did not observe this 31 to 20 meter interference, but some have, and gave an otherwise fine radio a ding review. A simple preselector will do the job. Too bad the noise canceller MFJ 1026 noise canceller does not have a tuned circuit like the MFJ 1048/1020; you could have solved two problems at the same time. Anyway, I lined up the trap on the strong US shortwave broadcaster that usually causes the problem, just to be sure I did not see the problem, while I had the radio on the bench.

POSSIBLE MODIFICATIONS:

It is to the NC-303's credit that there are not a lot of modifications out there. It is hard to come up with a worthwhile improvement for a receiver this good.

The NC-303 has 11 meters (27 MC). This band could be changed to the new WARC band 24.890 to 24.990 with little effort. But this is for my AM station. Strong signals from the CB'rs and freebanders often are a sign that 10 meters is open. Get on and call CQ while the others are just tuning and listening. Of course, there are beacons on 10 meters that you can use. But a 5 watt CB is sort of a beacon too.

The so called "X" band of 30 to 35 MC for the converters could be changed with considerable effort to cover a range of 10 MC for the 30 meter band. It could also be used for 18 MC or 17 meters.

Of course suitable dials would have to be generated to replace the existing calibration, and the value of the receiver as a classic would be diminished. These are only offered as ideas.

Another approach is to create an UP converter for the new WARC bands that changes the WARC band to output on 30 to 35 MC. This would give a place for 60 meters (5 MC), 30 meters (10 MC), and 17 meters (18 MC). The 11 meter band (27 MC) would still be best as a conversion to 12 meters (24.9 MC). You could also use the up converter for your favorite shortwave broadcast bands.

I found the SSB AVC a little on the fast side. A bigger capacitor would help hold the signal steady.

HOW I QUIT WORRYING AND LEARNED TO LOVE THE 4H4C BALLAST TUBE REPLACEMENT:

Many modern ham operators are not familiar with the Ballast Tube. You will find them in high end receivers like the Collins R-390 and the Hallicrafters SX-88. You also find one in the NC-303. There are highly qualified engineers who swear that the performance of these (pricey and short lived) ballast tubes is superior to the (even more expensive) special solid state replacment units sold. A ballast tube at full ratings is specified at 1000 hours. If you reduce it to about 80%, that increases to 5000 hours. If you cycle the power on and off frequently, that also reduces life span. If you drop it or jar it, it will fail immediately like any light bulb. I am not going to debate the efficacy of the use of the original ballast tube or the specialized replacement here. You will have to do your own research. A very scholarly thesis comparing the ballast tube and its solid state replacement can be found in Electric Radio Magazine ER# 321, February 2016, which is part 2 of the series; contact Ray at Electric Radio to order your copy and ask for the whole series: https://www.ermag.com/

I tend to think things out in a scientific fashion, as I was schooled to do. The NC-303 has a B+ regulator, to prevent line fluctuations from causing oscillator frequency changes. I would believe that regulating the filament voltage would also be beneficial to a degree. National itself states the expected drift with certain line changes is on the order of 20 Hz on 10 meters. If you replace the 4H4C Ballast tube with a 6V6GT vacuum tube, the drift increases to 250 Hz or so on 10 meters. I would consider that quite acceptable on AM mode. Also, I primarily use this receiver on 40 meters and below. Given that the AC line voltage these days in America is much better than the 50s, I also assert that this is a non problem. However, I would also assert that using DC on the filaments by way of an appropriately rated flavor of the LM317 regulator chip for the NC-303 would work even better than the 4H4C Ballast tube. It only has to operate one tube filament, the HF oscillator. With the R-390 or SX-88, you will have to figure out the load current and select the proper regulator, possibly one in a TO-3 package, with an appropriate heat sink. I will stipulate to being an agnostic on the belief that the 4H4C Ballast tube is essential to the proper operation of the NC-303. OR ANY OTHER QUALITY RECEIVER. My bovine fecal matter detector has gone off on this issue. You make up your own mind after you do your research.

Here is one more idea for the NC-303 that solves two problems. On the higher bands, the NC-303 and especially the HQ-170 can possibly have a hum FM modulation that appears on SSB and CW signals that is caused by using AC on the filament of the oscillator tube. While you can maybe ameliorate the problem by careful selection of oscillator tubes, the best way to go is DC filaments. In the NC-303, there is another problem, the availability of the ballast tube 4H4C. This is conveniently on a socket. In fact, the service bulletin says if the 4H4C fails, just plug in a 6V6GT tube. That is exactly what I did, to save the unobtainium ballast tube from wear. I put it in a box and placed it near the unused octal accessory plug at the right rear of the chassis, in case I ever needed it. There is a real fetish about this old buzzard technology. It was used to address the bad AC line stability back in the day. We no longer have this problem. The 6V6GT works just fine. Consider this idea: You build your own regulator on an octal plug that includes a rectifier, filter, and 6 volt or so regulator made from an LM317 or LM340. This eliminates the hum and you never have to worry about where you are going to get a ballast tube again. BOTTOM LINE: Quit worrying about a replacement ballast tube (save your money for useful toys) and plug in a dead flat "pull" 6V6GT that you put in your box of tubes for target practice.

A more elegant solution that does not require designing a regulator module is to rewire the oscillator tube filament to the regular 6.3 VAC supply; then you do not need the 4H4C OR the 6V6GT replacement. If you are worried about the increased current drain on the filament supply, wire the dial lamps to the source for the oscillator filament and 4H4C, and change the bulbs to the right voltage for that winding; an automotive version of the bulb should be perfect. This will rebalance the filament and regulated filament supply loads to duplicate the original drain on both, by simply swapping the loads.

On the R-390, you will find discussions related to this problem which involve changing the tubes to a higher filament voltage rating and eliminating the Ballast tube. Remember that the military used the R-390 in very severe environments, and required such extreme levels of performance that the R-390 still ranks very high on the Sherwood receiver test web site.

Since this is a matter of faith, I have encouraged you to research the matter and decide what is appropriate to your value system on vintage gear and preservation of it; please do not threaten to burn me at the stake for heresy. Whether it is pristine and stock, or reasonable modern adaptation to increase the time these treasures can be enjoyed, it is still appreciation that we all can share for the revered ancient ones who created them.

REFERENCES

MANUAL ERRATA AND NOTES:

Note that T8 in the BFO alignment section is actually T6.

DO NOT ADJUST C86. It is a trap for spurious responses. There is no information on this adjustment in the manual. It is resonant on an image frequency twice the 2215 KC IF removed from the actual operating frequency. Use a grid dip oscillator to find resonance to do a coarse adjustment, if someone screwed with it. It is resonant in the 31 meter foreign broadcast band (9 to 10 MHz). I set mine for the US shortwave broadcaster with the powerhouse signal down south. Once you have it close, use the adjustment to null out the offending signal while receiving the image frequency (either from the actual station, or a signal generator set to the frequency of the interfering station.) While I never observed problems, it was on the bench, so I did it. I use only resonant dipoles; perhaps that helped. If you are using a link coupled tuner, that also will minimize the likelyhood of seeing it. The real solution to this problem is described elsewhere in this article, if you experience it.

THE BEST RECEIVER FOR AM AND CW: HQ-170/180, SX-101, OR NC-303?

The NC-303 has a single ended 6AQ5, but it does have reasonable audio. There is a page on the internet that did a full frequency response on the 303, and it is very good. The HQ170/180 can do pretty good, if you get rid of those circuit packs for interstage coupling and the kludgey "auto response" circuit which changes the frequency response depending on the volume setting. The 180 still only goes to 6 KC wide, and the 303 at broad is 8KC. The 303 has a real tone control (standard) that enhances phone, or can eliminate some noise or peak for CW. The 303 has huge bandspread and a nice mechanical bandspread enhancement (little knob that drives the big knob). The 303 has much better components in the front end for better drift and stability. The 303 noise limiter is separate optimized circuits for phone and CW/SSB; the HQ180 noise limiter is pathetic on any mode. The sensitivity of the 303 is better than the 180 on the higher bands. The alignment of the 303 is pretty straightforward with a standard signal generator. The 180 can only be aligned properly with a sweep generator. You can get it close, but optimum performance needs the sweep generator. Of course, the HQ180 is general coverage, so it covers the WARC bands and WWV on all frequencies. The NC-303 had dial calibration for the proprietary VHF (6 meter, 2 meter, 1.25 meter) converters; other brands of converters did not have the requisite 30 to 35 MC output. And the S meter is pretty accurate.

I am a hardcore Hammarlund devotee, but I LOVE the National NC-303.

All of the comments I made about the HQ180 apply to the ham band only HQ170. One additional flaw of the 170 is that the drift on 40 meters can be so bad that it leaves the upper end of the band inaccessible, intolerable if you operate AM phone. The only marginal fix is to misalign it so the cal knob is at the end of its adjustment range, and even that is not enough for some of the 170s that have crossed my workbench. The 180 was general coverage, so you could readjust the main tuning. The HQ-170 VHF, a later offering, had a pretty good nuvistor internal 2 meter converter, plus the abysmal performing 6 meter band standard. An external converter for above 30 MC is a much better choice, but more expensive. The NC-303 got that right.

A discussion of this type cannot exclude the SX101 SX101A. These are the three ham band only receivers that are the mid top end price range. The Collins receivers are not in the same price class, and have a different architecture. Collins used a crystal controlled first HF oscillator, which resulted in dramatically better stability on 20 meters and above, but it also cost a lot more. Even today, it commands a much higher price. None of the competitors to the Collins could be set up to transceive with a matching transmitter; that was coming in the next generation radios. The SX101 used a 12BY7 for the HF oscillator; I never figured out why a tube with such a high plate dissipation wound up in a spot where a smaller tube would do. Some versions left the oscillator filament on all the time (same as the Hammarlunds); the tube would fail in a year if you did not have an outlet strip with a switch to kill the AC power to the radio. The SX101 has 160 meters but no product detector; the SX101A has a product detector but no 160 meters. The HQ170/180 was triple conversion, about 3 MC, 455KC, 50 KC. The NC303 and SX101 are dual conversion, about 1650 KC and about 60 KC. Theoretically, that should result in fewer spurious signals. One SX101 I worked on had some weak birdies in unexpected bands; I believe that was due to a previous owner removing a sheet metal shield that covered the front end coils and discarding it. I wish I could have played a little more with it, but the owner wanted it back after repair. The HQ170 has no WWV receive for checking the calibrator. (The HQ180 is general coverage so it covers all WWV and WARC bands.) Neither does the NC-303 when outfitted with the standard calibrator; the deluxe one will receive WWV, but the WWV antenna must be attached to the top of the calibrator, klunky but it works. The SX101 includes WWV receive. The NC-303 and SX101A include a 30-35 MC range for VH converter accessories. The later HQ170 VHF had a nice internal 2 meter nuvistor converter and the standard 6 meter band, which did not work. HQ170A 6 meter reception was deaf as a post and the RF stage oscillated as you tuned the antenna trimmer or main tuning through various settings, due to the long coax lead to the front panel trimmer capacitor. The HQ170 6 meter was more stable but still deaf as a post. If you want to do 6 meters, none of these radios are for you anyway; solid state converters are the way to go. I assume you want a good HF radio for 160 through 20 to pair with a classic transmitter. HQ170 and 180 have a vernier tuning which makes SSB and CW easier to tune. NC-303 has a small knob next to the main tuning that can be engaged as a gear reduction for fine tuning. SX101 SX101A does not have any fine tuning provision. The quality of the RF stage and first converter coils is good in the NC-303 and SX101. Not so much in the small IF can cases used in the HQ170 and 180. The HQ 180 used a small crystal filter in the first IF at around 3 MC; the HQ170 and all the others did not. While this makes alignment a bit tricky (use a sweep generator), it makes the HQ180 a bit better at off channel QRM than the HQ170.

For comparative performance on 80, 40, and 20 meters of a number of classic receivers, see the website: http://www.w1vd.com/BAreceivertest.html

The NC-303, HQ170/180, and the Collins 75A4 are comparable at minimum discernable signal approximate -140 dBm. The 20 KC blocking of the NC303 is 110 dB, the Collins is 100 dB. The HQ170 is 108 sB on single conversion bands below 40 meters. The HQ170 blocking falls to only 81 dB on the triple conversion bands 40 and 20 meters (and up presumably), due to the gain distribution over the triple conversion architecture. W1VD also ranks the audio response on AM of various classic receivers. This site is a great resource if you are shopping for a classic receiver.

All the rigs of the time before crystal controlled first oscillators tried to deal with drift in various ways. The SX101 ham band only used a resistive heater with a thermostat. It also left the HF oscillator running with the receiver off. The HQ-170 and 180 had an accessory clock that allowed you to turn it on in advance of scheduled use, like a clock radio; not too bad. The "A" versions also installed a separate filament transformer for the first and second oscillators and the first 6BE6 converter. While this may have reduced warm up drift, it means that the tubes were always on; they lasted only a year. I advise everyone with one of the A versions to use an outlet strip and turn the juice off when not in use to prevent the loss of expensive tubes. The NC-303 did it the right way: high quality components in the front end, and it worked. Then came the Collins 75 series, with crystal controlled first oscillator; all the subsequent high quality receivers followed that format. The Heathkit SB-300 and its descendants plagiarized the superior technology. It was the end for Hammarlund and the companies that stuck with a tuneable first HF oscillator.

I hope you found this review of the NC-303 and its contemporaries helpful.

IMPORTANT HEADPHONE SAFETY NOTE

It is still a good idea to use an audio attenuator for headphones on this radio, since the direct audio speaker output is available on the front panel. This was designed with 2000 Ohm headsets in mind. I like Radio Shack light weight low-Z monaural full frequency response headsets for AM use, so I can monitor the outgoing transmissions. But I use a separate headphone breakout box for multiple headsets that I constructed for field day with individual volume controls (one for me, one for my logger). Why not just put a new headphone jack with switching contacts and a resistor to fix this? You could. The way I use my station, I share the excellent Hammarlund speaker with the radios on my operating position. I do not want to have to pull the radio out to connect it on the back, so I plug the speaker into the headphone jack of the radio I am using at the time. An attenuator in the headphone jack would prevent that desirable feature.

UPDATE 7/3/2017

The NC-303 has now become my favorite ham bands only receiver from that vintage, due to its superior performance and elegance of design, better than its contemporary competitors, the HQ-170 or SX-101.

One of the dilemmas presented to people who love to restore vintage gear is how to find unobtainium parts for your project. This update is about the chain that is used on the dial mechanism. It is shown on page 16 of the NC-303 manual in the dial cord stringing diagram. In the parts list, it appears not even to be listed. The fabric dial cords appear to be the only parts considered as likely to be replaced by a user.

It is important to determine what I would refer to as the "pitch" of the chain, since the band switch drives a sprocket which will only operate properly with the correct spacing. However beaded chain is specified by its "number of balls per inch". If you use the incorrect spacing, the sprocket will not drive the mechanism properly and it will eventually "jump" out of sync. The dial drum then will not indicate the correct band of operation. These are suppliers, with their sizing aids:

• https://www.mcmaster.com/#roller-chain/=18cmze8
• http://www.ballchain.com/chain_sizes.html

My first inclination was to check out McMaster-Carr, which is a gold mine of useful mechanical parts for engineers and maintenance people. I did a quick search and they seem to only have link style chain, not beaded chain. There is another company who sells beaded chain, and might be willing to send you a sample in the length you need upon polite inquiry. Keep in mind these companies are not set up to market small quantities, and may not be interested. You may be able to obtain samples from a dealer in small quantities. If you have an engineering company, an inquiry on company letterhead for a sample can work, but only once.

My guess is that you can go to a craft shop and find what you need, but probably not in the durability of the original. The other source that comes to mind is a plumbing supply shop. Beaded chain is commonly used for bathtub stoppers or toilet tank valves. You might get lucky and find just the right size there in a short length. Also, the older bare bulb electrical fixtures or ceiling fans often have beaded chain. You might be able to salvage one from there.

Also, window blinds employ this style of chain, and surprisingly Amazon sells it: https://www.amazon.com/Ball-Chain-Spool-Nickel-Plated/dp/B005GQ3OEC.

Failing at that, you might find an original part with one of the professional restorers like Chuck Felton: http://www.feltondesign.com/.

Otherwise, you are faced with searching for a "basket case" unit for salvage. This is not a bad idea either. For my "keeper" items, I often buy discards at swapmeets to dismantle. I was able to repair nearly ten SB-104s this way, ending up with only one carcass (which still had a few userul circuit boards left). If you do find a source, I would appreciate if you share it with me for posting. I will gladly give credit to your for your research, should you find a source.

Good luck with your restoration, and please let me know if you are able to find a part source.