Project: S66D/E 9018-based AM Radio Kit

If you’ve been wondering about all the random radio kits I’ve posted about recently and had the thought “Gee, that guy must’ve just looked for any cheap radio kits on eBay and bought them all!” You’d be absolutely right. In fact, there’s a few more on the way. I tried not to spend more than about AU$15, as it’s amazing how much you can get for your dollar nowadays. I tried not to choose kits that used the same ICs even if they had different designs, just to get a chance to look at a variety of kits.

This kit concerns itself with an AM radio made of completely discrete components. There is no “cheating” as with the other kits which rely mainly on a TRF “radio IC”. This one is based around (mostly) 9018, 9014 and 8050 transistors along with a number of transformers. Even the audio output stage is made without any IC amplifiers. Consider this a “discrete” version of something like the Funway Pocket Transistor Radio kit, complete with a case and a speaker for AU$8.27 delivered.

The Kit

Just like a few other kits, this one starts with everything in a single plastic bag. The colour on the listing had a more attractive “leafy” green case, and instead, the one I have is a more … yellow-lime colour with slightly different printing on the case. All the goodies are stuffed inside the case, using it as “protection” in transit.

When I used the word “protection”, I meant it quite literally, as there were bits of the back cover that had broken off in transit because of the parts rattling around inside. There is a 16-ohm speaker with a curiously small voice coil and a plastic diaphragm, a plastic grille that push-fits into the cover, a dial cover with self-adhesive tape on the rear, and a bag of components.

The bag of components unpacks to a reasonable number of components, with quite a few transistors and resistors, and a smaller number of capacitors and transformers, along with the other parts.

You also get a single page of instructions, with the PCB wrapped up inside. The instructions refers to the unit as an S66D, although the casing has S66E written on it. Because all of the information is in Chinese, I can’t really understand it, but I suppose if you are proficient in Chinese, you might be able to learn something about it. From what I can tell, it claims to be an AM receiver with a frequency range between 515khz to 1605khz.

The schematic is clearer, because it transcends language to some degree. The transistor types of each are labelled above the diagram (and I believe it says VT5/VT6 is 9013H *or* 8050). There are a few points where the connection is broken (A, B, C, D) where it seems that some indicative currents are provided as a check that the circuit is correctly built. After measuring, the cuts should be bridged over with a dab of solder.

From what I can see, this design uses a ferrite rod antenna with an antenna and tuning coil. The first stage with VT1 forms a tuned circuit with the variable capacitor tuning knob that helps select the frequency of interest, with subsequent two stages (VT2 and VT3) amplifying and filtering the signal through an L-C arrangement in the transformer (I’m assuming). VT4 forms the first of the audio stages, with the volume knob selecting the input to VT4 which is used to drive the high-impedance side of an audio transformer (Dick Smith kit style – you don’t see this very often nowadays) which then is used to drive VT5 and VT6 as the final stage of audio amplification. You also have an LED for power indication and a power switch as part of the volume control knob.

There is a nice duotone print to help constructors with identifying the antenna coil turns and connections (a, b, c, d – note the lower case, the test points are split and marked upper case). The transistors used are all the same pin order, which is nice. It shows you how to mount the LED, audio jack, capacitors, and read the colour codes in Chinese. This comes in handy, as the transformer colours are marked in Chinese – T3 is white, T2 is red, and T4 is black (according to the print).

The PCB had some scuffing at the top as well, and is a single sided paper type PCB with silkscreening on the top. The silkscreening isn’t particularly good when it comes to transistors with the three legs oriented in the triangular orientation, because it can mislead constructors into thinking that it might go a certain way – instead constructors need to pay attention to the e, b, c markings and get the pins in the right way.

The underside of the board shows it has lacquered copper pads, which have a solder mask applied and relatively good pad sizes. It should make for a comfortable soldering experience.

Construction Experience

As with all kits, you’ll need to bring your own soldering iron, sidecutters and solder. In the case of this kit, it’s also good to have a hot glue gun with some glue to assemble the ferrite rod and speaker into the casing, small Phillips screwdriver to assemble the tuning knob, volume knob, variable capacitor and PCB screws, small flat-blade screwdriver to adjust the transformers, another working AM radio to check the tuning (recommended), two AA batteries, a digital multimeter (DMM) to do continuity checks and optionally some desoldering braid in case you make a mistake.

Because this kit is a little challenging in terms of the physical fit of all the components, I decided to go slowly and do things “by process of elimination”. As a result, I constructed it in the most boneheaded way possible, and I don’t advise you do the same. It was, however, still entertaining to me to see how hard I could make my own life. First thing’s first – the obvious hardware parts went in – the speaker grille and the front dial indicator.

Then, I checked the PCB orientation and fit – noting the retaining screw hole. Importantly, this kit has the tuning knob towards the front side of the radio and the volume knob on the rear. As a result, I had to pay attention to how the variable capacitor was to be mounted. I also checked the fit of the speaker, which was extremely loose in the designated area. Superglue would not be advisable, as it would rock around somewhat. At this point, I then realized the batteries were to be mounted just behind the speaker – so the “shorting bar” that connects the remote terminals can be installed by pushing them into place.

It was at this point, I realized that the tuning capacitor was to be installed in such a way that the tuning lug was to face through the PCB, and secured in place with two screws. Wedged in-between was a plastic piece, which has to face towards the rear, to hold the ferrite rod antenna. Now confident about this, I threaded the legs through and screwed it in.

I then immediately soldered it into place and clipped the leads off, so that the tuning wheel can be fitted. I quickly checked the mechanical fit of the tuning wheel – it was correct in its position in the case, but their choice of tuning capacitor and the design of the tuning wheel means that the tuning indicator was always 90 degrees out! That’s not very nice … I suppose a spot with a paint marker might also be a good idea.

With that crisis averted, I checked the fit of the volume knob and power switch, which was to face backwards. No problems there, although the switch was rough in operation. Low quality components … how disappointing. After I had sorted the colour of the tuning transformers, I soldered them in as well. I handled the electrolytic caps as well, and then went to mount the audio transformer. To check the correct orientation of the transformer, you need to find the winding that maps the top left side pin to the middle right pin. Once this is identified with a DMM, then you can match it with the silkscreen drawing, as pin 1 on the transformer is not identified.

If my former electronics teacher saw me constructing the kit in this order, he might think I’ve gone mad. This is a silly way to do it because you’ve now restricted your access to the board for fitting the other parts. However, I don’t mind – it helps me quickly eliminate the obvious parts first and get things installed the right way.

The solderability of the board is excellent – the lacquer burns away nicely and the pads are well sized for soldering. Solder flows quite well. It was at this point I realized that I have hit my first missing part – the 3.5mm headphone jack. How disappointing.

Now that’s left is to get in the resistors, capacitors and transistors. The transistors need special attention, as they go via the e, b, c markings on the board – the triangles are misleading! The LED can be installed with its long legs ready to protrude through the casing, and the ferrite rod can be slipped in with the coil over the top. Note that the aerial winding connects to a-b and the tuning winding to c-d. Because the windings are formed on a piece of cardboard in the open, b-c will be found at the point two wires emerge together, and a will be the further winding from this point, and d will be the closer winding to this point. If you want to confirm, I suggest using a soldering iron to burn off the lacquer at the end of the windings and measure the resistance ratio. Once the coil is in place, the leads can be trimmed and then tinned. The pads can be tinned, and then the wires tack-soldered on.

It was then, I realized that I was missing even more parts, namely R1 and R11. R1 was supposed to be 200k ohms, but I received a 300k ohm resistor instead. R11 was supposed to be 330 ohms, but I received an extra 1.8k ohm resistor instead.

I contacted the seller to see what could be done, but it seems he’s also enjoying a long weekend at the moment. As a result, I raided my junkbox to find some replacements. The closest to 200k was 220k, so I went with that, and the 330 ohm resistor was replaced with a 470 ohm (after all, it’s just for the LED, so it means it just won’t be quite as bright). Because the footprints seem to be made for compact 1/4 watt or 1/8th watt resistors, the larger bodied resistors went in rather inelegantly.

I then decided to solder wires to the speaker, then solder them onto the BL terminals. I soldered wires to the battery contacts as well, and then to their corresponding GB terminals.

At this point, the kit is almost ready to tune up. In the case you wish to tune with batteries, you can just put the contacts into place, fit the batteries and start tuning. But because I had a bench-top power supply to use, I just clipped the supply onto the battery terminals and began tuning that way.

Tuning and Testing

As I couldn’t understand the instructions, I was on my own. I started by grabbing my Agilent U1241B handheld DMM to read out the current through the test points. My results deviated quite badly from the print – this could be partly due to substituted resistor value (especially R1) but I couldn’t get any change even if I turned the transformers and the currents weren’t “lethally high”, so I accepted my fate and solder-bridged the pads to go onto the next stage.

• A = 0.65mA (should be 0.3mA)
• B = 1.54mA (should be 0.5mA)
• C = 3.28mA (should be 2mA)
• D = 9.55mA (should be 1.5mA)

I then decided to start at the low end of the band and try to catch a station I knew would be around there. I twisted all of the transformers to their middle setting (approximately), and then rocked the tuning wheel until I heard something. I started with the first stage (red) to try and get the tuning alignment in first along with a slight tweak to the screws on the rear of the variable capacitor to see if I could get a little more range. Once it was audible, I tweaked the following stages (white, then black/yellow), to try and get the best audio, while tuning up the band to check for the range of the receiver. Once that was completed, I considered my task done, as there were no other real adjustments to be made.

Once tuning is done, then the board can be secured into the case with the tapping screw, and the speaker and ferrite rods can be glued into place with some hot melt glue.

Then, the rear cover can be snapped on, and the unit is complete.

Unfortunately, as it was missing the headphone jack and I couldn’t be bothered to solder on some bare wires, I didn’t record an audio sample. However, the result was not particularly special or impressive – it’s just a bog standard AM radio. It tunes the whole range just fine, although depending on how you’ve tuned the filters, it can distort badly and or oscillate and create “electronic fart” noises, or squeal. It’s a bit of an “instrument” in a way. The included speaker and case don’t really make for an ideal listening experience either, with the audio quite hollow and lifeless, and battery replacement somewhat unusual in needing to take the whole radio apart by prying off the rear cover. The plastic itself was somewhat weak and flimsy as well, but I suppose that’s expected for the low price of the kit.

There was a hole in the bottom of the front cover, with no obvious reason. In fact, it’d probably be one of the worst places to put a DC power jack if you wanted to … a design choice I’d only expect from Apple …

Also, the hole on the side for where the headphone jack would be … if it came with one.

Conclusion

When I purchased this kit, I expected it to be a little more difficult than the others, primarily owing to the case design and fully discrete nature of the circuit. The soldering part was actually quite easy, because the board had properly sized pads, a silkscreen and solder resist. The silkscreening was not perfect, and there were a few points where I needed to do some lateral thinking to understand what to do, but after taking things slowly and patiently, it was possible to construct the kit as intended, align it and get it operating.

However, it seems that the quality of the case and the speaker is somewhat poor, it was partially damaged in transit and had an extra hole in one of the case panels and also suffers from the poor quality control. This issue seems to plague cheap Chinese kits, namely missing or incorrect components and could be quite disappointing to beginners and first-time constructors. In my case, one resistor was missing, one had an incorrect value, and a headphone jack was missing. I’ve contacted the seller to see what they can do about it, but I suspect they won’t be too sympathetic.