More Chinese AM radio kits

[Jolly 7]

Here is my completed radio. It is not perfect because there is some squeal/ motorboating around 900 kHz which I can't seem to get rid of, but then my signal generator is fixed at 455 kHz and the radio's IF is 465 kHz, so that might be the cause.
Overall I'm still quite happy with it and this set will probably outperform many portable AM radios on the market due to the low noise floor and very good sensitivity.

[Argus25]

Reading through this whole thread is a reminder that when building any superhet radio, be it a kit or a self designed one, when it is completed, all parts of the system need to be tested & checked.

Firstly the supply current to the audio & driver stages and the whole radio. At least one check with a 1kHz sine wave and audio into a dummy load, to check that the audio stages are basically distortion free and have around the correct output level.

Then a sweep of the IF (ideally with a signal generator and scope) to ensure the IF is stable, has about the correct bandwidth & gain.(if not the bypass and grounding arrangements and AGC system checked and modified if required) Then a check of the local oscillator level and frequency range. Then an overall radio sensitivity check.

There are a lot of variables which could sabotage the operation and performance of the radio kit, so these checks, on a newly built unit, are definitely required. If they are not done you end up with all the reports of bad performance or failure to operate, many a case it will just be one unsuitable part or component or something simple. Without the systematic tests, the faults will remain unsolved and elusive. A lot of the time you can "get lucky" and everything will appear to be ok first pop, but as shown by the remarks on this thread, it is a bit hit & miss.

[poppydog]

Mines all sorted as well, for some reason the 9018g that was used as the detector diode wouldn’t detect unless it was prompted with my signal generator. I tried a silicon diode in place of it but that wouldn’t work so I tried a BC377 and that worked fine. One of the electrolytic caps tested awful so I stuck one of my own in. Took quite a while to get all the stations to come in and to get them in their right places on the dial but its done and its picking up everything very well now including Absolute (fading in and out)whose transmitters were turned off near me a while back now. For less than a fiver delivered its been a good learning curve. 5 mins with a meter or a tester, testing components are in spec is recommended prior to installation of components.

[regenfreak]

I built mine a few months ago and it was dead. I think the output transformer was defective. So I tossed it away.

Then I bought another more expensive and bigger superhet kit (about £16)from a Ham dealer in Thailand. It worked right away although there were a number of errors in the thai/English instruction and over current problem.

I ordered two weeks ago the same £5 Chinese superhet kit for the second time from a different dealer. This time it arrived quick and it has much clear and better instruction. I intend to use the oscillator and IF cans for IC NE612AN mixer/oscillator experiment.

I have also bought the Chinese 2P3 kit for £20. It is a very nice kit with very good instruction. Well-worth the money.

[GeoffK]

I decided to have a go at making one of these radios, bought from a Singapore site as it was under £5 including postage and arrived quite quickly. It went together reasonably well, the red plastic aerial ferrite holder needed one screw hole countersinking as the screw wouldn’t go into the tuner even after the edge was bevelled as it was hitting a couple of the solder joints and wouldn’t lay flat. I tried a diode instead of the 9018G transistor as the detector, a BA219 as it was the only one I had, a 1N916 should also work here. Because the agc is taken from the anode of the diode via the last IF secondary the cathode connects to the volume control end, unlike a conventional pocket radio with npn transistors where the cathode connects to the last IF transformer.

The radio has only been temporary put together along with a temporary speaker to try it, and it worked straight away. Only the oscillator trimmer, the one nearest the edge of the case on the tuner, needed some adjustment to get the stations in roughly the right place, this will be improved once the dial is fitted. It still needs some final putting together but I am tempted to leave the speaker as it sounds very good, it was quite a revelation as I wasn’t really expecting it to work all that well if at all. The quiescent current is around 20mA so would like to get this down a bit, possibly a larger value of base bias resistor or the output transistors emitter resistor added.

The 9018G transistors give plenty of gain in the IF stage, the IF transformers are spot on. The only components I used from my own parts was a 100uF capacitor instead of the one of the two supplied, and the 0.01uF mixer input coupling capacitor as, like others here, it was the only part missing. Orange sleeving was used on the resistor leads but they are so thin the sleeving wouldn’t go around the curved part. The gaps in the printed board for current measuring had wire from the offcuts soldered across rather than just solder to bridge them.

[poppydog]

I had trouble with the ferrite/tuning cap mount, mine was cured with the stanley knife, also the vol wheel kept coming loose but a blob of nail varnish seems to have stopped that. Whilst I'm fairly happy with mine I find that small pocket radios are terrible for listening to music on.

regards poppydog

[GeoffK]

Whilst I'm fairly happy with mine I find that small pocket radios are terrible for listening to music on.

For the size of the speaker I quite like listening to music. Sometimes it is the lack of full freqency that makes music easier to listen to. Speech can also be clearer with a small speaker.

[GeoffK]

Finished the radio, a piece of white gauze was dyed black with printing ink, and after drying put in front of speaker to make a grill instead of leaving the speaker cone exposed. The wires are all to long so these might be shorted. The aerial ferrite is held in place with a piece of sticky tape, the speaker needs this or glue.

[Jolly 7]

The speaker cone protection is a very good idea. I might do this myself with some very light black fabric I have lying around. We sometimes forget that dust, humidity and smoke can eventually ruin our prized electronic possessions.

[bc312]

Having followed this thread with interest, I decided to take up the challenge and order a HX108-2 kit. Once the kit arrived my first action was to test every component and record the various measurements.
The doubtful components were the capacitors as others have found – the electrolytics all had ESR’s which were a little high, and the disc ceramics were out of spec, even applying the worst case tolerance of +80% -20%. I sourced suitable replacements from my stock.
The supplied transistors were 4x 9018G and 3x9013H, so no 9018H for V3, V4 and V5, again something which others seemed to have experienced. However the gain figure for the transistors was within that specified, so I decided to use them in the kit, using the highest gain 9018G for V1, and close matched 9013H's for the output pair V6 and V7. The diodes are all 1N4148 as supplied, and tested ok.
The oscillator and IF transformers all showed continuity and reasonable inductance measurements, and the Chinese letters on the schematic translated as B2 – Red, B3 – Yellow, B4 – White, and B5 – Black.
All resistors supplied were within the ±5% tolerance.
The audio transformers were identified as Driver – Green, and Output – Yellow, and primary and secondary windings checked for continuity – both measured ok.
The kit went together well and once completed was powered and tested with a DMM across the five bridging points. All measurements agreed with the values shown on the schematic. The bridging points were then soldered with a wire link, and everything assembled into the case.
Switch-on resulted in a speaker hiss and tuning brought in a few stations. The frequency was a little high against the tuning scale, but adjustment of the ferrite rod antenna/coupling coil position, trimmer capacitors and oscillator transformer corrected that. The IF transformers were peaked for maximum audio output. I was surprised by the level of audio output - given a 3 volt supply voltage, and the sound quality was very acceptable from the tiny loudspeaker. Altogether very pleasing for a £5 investment.

[Tully Bascombe]

For the last year or so I've been steadily working my way through these cheap Chinese kits. At the moment my collection consists of:

FM Alarm Clock Radio with LCD display. Works OK except very tricky to tune without distortion due to stiff tuning knob. Frequency indication is almost correct. Clock keeps good time.

ZX921. If it wasn't for the whistling, howling and motorboating this might actually be a good sensitive radio. Once tuned in, volume level and sound quality is very acceptable.

HX108-2. Have built two of these. With the first one I messed up and got an output transistor the wrong way round which explained the awful distortion. After correcting my mistake the sound quality was no better so maybe the transistor was fried. The second one had a damaged speaker so I swapped it with the other one. Works quite well but sound quality still not great.

S66E. Have built three but only paid for one (long story). All work surprisingly well and sound is better than expected from the plastic speaker. Headphone socket is a bonus although the hole in the case doesn't line up with the PCB. As an experiment I built one of them with closer tolerance resistors and caps. Result: no noticeable difference.

210SP 'Paeansonic'. Good FM performance but AM very weak. Stiff tuning due to the horrible dial pointer method. Locks onto FM stations well, clean, clear sound from the tiny plastic speaker. Awkward to tweak as the batteries come out when you take the back off.

ZX2051. As with the 210SP, good FM performance but weak AM. LED tuning indicator actually works. The real speaker gives it an advantage.

ZX620 AM/FM in an ugly yellow case. The actual radio I received from Banggood was the AM/SW version ZX623 and got a full refund. SW is one band 15M, not very useful. AM performance is excellent as is the sound quality from the 'large' real speaker. I can quite happily live with this one, just a shame that's there's no FM. It appears that the ZX620 is no longer available.

Next on my list is the multiband EDT-2902 which could possible be a bit of a challenge.

[Neil Purling]

I have seen the ZX62X advertised that mentioned SW, thus being ZX623. Yet the image with the listing still showed the ZX620 FM version. I thought there was a typo and the listing was for FM/MW.
After all it does not seem like the SW band would be that useful.

[G8UWM-MildMartin]

One EDT-2902 now ordered. Thanks for the tip-off!
I do love these kits; I get my money's worth in the entertainment of building them, regardless of how they turn out.

[Tully Bascombe]

Originally Posted by Neil Purling

I have seen the ZX62X advertised that mentioned SW, thus being ZX623. Yet the image with the listing still showed the ZX620 FM version. I thought there was a typo and the listing was for FM/MW.
After all it does not seem like the SW band would be that useful.

I ordered two ZX620 AM/FM, one from Banggood and one from eBay. In both cases I got the ZX623 AM/SW even though the listings clearly showed the ZX620. Banggood refunded me in full and with the eBay seller I asked for the correct radio. Instead I was also offered a full refund.
There seems to be confusion amongst the sellers of these radios. Only Aliexpress has got it right as they are listing the ZX623 and the 620 as unavailable.

Originally Posted by G8UWM-MildMartin

I do love these kits; I get my money's worth in the entertainment of building them, regardless of how they turn out.

Totally agree, I find the whole process very therapeutic and a working set is a bonus.

[Tully Bascombe]

I just couldn't live with the ZX623's ugly and cheap looking yellow case so I decided to paint it black. Looks a lot better.

Assembly gave me a couple of problems. There's no indication which way round the tuning capacitor goes and I got it wrong. All the stations were reversed if that's the right word. Anyway, all the stations at the high end of the dial were at the low end and vice versa. I devil to desolder and make good.

The bracket for the ferrite rod has to be mounted under the tuning cap on the component side but I found the leads were then too short to solder. In the end I cut the bracket down and glued it to the PCB.

[Tully Bascombe]

Another kit to add to the collection, the RW08-7B, one I haven't seen before.

The circuit looks like the HX108-2 but the addition of a headphone socket and power LED.

The case is similar to the ZX2051 with the batteries side by side rather than either side of the speaker.

[colourking]

Originally Posted by Jolly 7

Here is my completed radio. It is not perfect because there is some squeal/ motorboating around 900 kHz which I can't seem to get rid of, but then my signal generator is fixed at 455 kHz and the radio's IF is 465 kHz, so that might be the cause.

The squeal is probably due to 2nd harmonic of IF (i.e. 455x2 = 910kHz) being picked up by the ferrite rod (which is close to IF components).

realign to shift the IF up a bit ( or back up to 465 ? ) would at least clear it of BBC on 909kHz which I assume is being affected!

[Argus25]

Originally Posted by Tully Bascombe

Another kit to add to the collection, the RW08-7B, one I haven't seen before.

The circuit looks like the HX108-2 but the addition of a headphone socket and power LED.

The case is similar to the ZX2051 with the batteries side by side rather than either side of the speaker.

Unfortunately the circuit is really badly designed by a person with a limited knowledge of transistor design and who hasn't bothered to read Horowitz & Hill.
You will notice stages V2 and V5 with grounded emitters and base bias resistor to the power rail. Their circuit function and the collector currents are very dependent on the hfe of individual transistors that varies wildly among specimens. So the radio would work ok with transistors with just the right hfe, but if its too low or high then radio will not work or malfunction.