Minimod finished - post build questions

[WaveyDipole]

I have been working on my Minimod project these last couple of days and had completed assembly and the enclosure today. On first power up the power indicator LED did not come on and I discovered that that the regulator that I had used from my junk-box was faulty and was putting pout only around 1.5v. After replacing it, the regulated supply was at 9.04v and the LED lit, although a bit dimly, so I replaced the 3.3k resistor with a 2.2k which was an improvement. Comparing the rev. 2.0.1 diagram with what looks like a later one that includes the power circuit, I discovered that there have breen changes to a couple of resistor values, including R9, the LED current limiter which is now shown as 1.5k, and also R11.

The next step was to check the antenna output and sure enough, there was a clean sine wave present with a frequency around the 1MHz mark. Using a sig gen, I added a 1KHz audio signal to the input and got the characteristic AM pattern. So far so good. I then had a bit of fun and games because Q2 failed for some reason, shorting from S to D. It didn't take a couple of minutes to replace and we were up and running again. I found a quiet spot on the radio dial and then tuned L1 to set the frequency and L2 for the best output signal amplitude. For the antenna, I had used about a meter of wire.

Although the oscilloscope shows a nice enough trace for all signals (the carrier, the 800hz audio signal and the modulated AM signal), I am unable to get a good clean signal on the radio. The radio is only 3ft away and barely picks up the signal from the Minimod and there is a lot of noise. However, it does appear to work.

I notice that the article in The Bulletin suggests lengthening the antenna to 2m which I will try tomorrow to see whether there is any improvement. I also have one of those HiFi loop antennas which I can also experiment with.

I couldn't get the frequency as low as 900kHz. Below 1MHz it drops rather rapidly and is impossible to tune accurately. From around 1MHz tuning is stable and I was able to go up to over 1.2MHz. I will need to check whether it can go as high as 1.6MHz, but the frequency does seem to start dropping back beyond a certain point.

Incidentally, I am curious as to why R11 was changed from 10k to 3.3k in the newest drawing?

[WaveyDipole]

Things look much better after a nights rest and so it was this morning. I discovered that the cracking was not being caused by the Minimod, but the Bush TR82C radio I was using for the test. The battery had become weak with the result that reception would start to get weak and crackly after a minute or two, something which was not immediately noticeable when tuned to a silent zone. Plugging it into the mains made a big difference.

The HiFi loop antenna was absolutely rubbish, however a 2m length of single core bell wire worked very well. I have added an antenna socket and fitted a banana jack to the length of wire.

I found that the Minimod will tune from around 1.039MHz to 1.711MHz, although the core was almost totally unscrewed at the top end and the emitted signal was weaker. It seems to be perfectly fine at around the 1.2MHz mark. I expect that the actual range obtained will very much depend on the actual value of the capacitor and coil inductance, which, of course, can vary somewhat.

Thank you G4JQT for the project and Electronpusher0 for making the PCBs available. I look forward, at some point, to being able to play some music on my valve radio.

[Ian - G4JQT]

The exact value of R9 is not critical, but it depends upon the LED. Older LED types need about 20mA to light sufficiently and that's why I didn't use one in the MiniMod mk1. Adjust R9 to give the required LED brightness.

You should certainly have enough output from the MiniMod from a up to three feet away to give a 'fully quieting' signal on the radio - I'm assuming the radio is working normally on other stations. (As described in the article the MiniMod is designed to be used close to the radio, not necessarily from a cross a room. That might be possible in an electrically very quite location with a sensitive radio, but is not typical.)

Some suggestions:

Due to the Q of the tuned circuits RF is strongest at around 950 to 1200 kHz. If C1 and C13 are both 270pF then the cores of L1 and L2 should be in approximately the same depth in the formers. If not, something is wrong. Not sure what you mean by "impossible to tune accurately".

Does the output clearly peak when you adjust the core of L2? I've not managed to do it, but it might be possible if L1 is at its lowest frequency and L2 tuned to its highest that it's tuned to the second harmonic. That's why C1 and C13 should be closely matched to ensure the tuning ranges of these two tuned circuits are similar.

As with many slug-tuned resonant circuits it's possible to get the same inductance at two points at frequency extremes. To avoid this just keep to the suggested frequency range.

A scope if connected directly to the antenna, will significantly detune the output tuned circuit and give false peaks. Adding more wire for the antenna is not likely to give a stronger signal - something is not right.

Are you using your own layout or one of the PCBs? Performance-wise it shouldn't matter, but it's harder to make an error with the PCB.

I changed R11 to 3.3k as it seemed to give very slightly better output, particularly when I changed from 2N3819 (which has quite a wide performance spec) to the BF256.

Hope this helps and let us know how you get on.

[Ian - G4JQT]

Oh, just seen your later post. Thank you.

Yes, the tuning range can vary for the reasons you state, and 1.7 MHz is near to the high end I have seen with the values suggested. (MW officially ends in Europe at 1602 kHz and many older sets don't go much beyond 1500 kHz.)

Regards,

Ian

[vinrads]

I am reading this with interest Ian, you say connecting the scope will de tune the coils is there a recommended way to use the scope to set the coil ,or would it be better to use a radio and monitor the AGC line . Mick.

[WaveyDipole]

Originally Posted by Ian - G4JQT

Due to the Q of the tuned circuits RF is strongest at around 950 to 1200 kHz. If C1 and C13 are both 270pF then the cores of L1 and L2 should be in approximately the same depth in the formers. If not, something is wrong. Not sure what you mean by "impossible to tune accurately".

Yes, C1 and C13 are indeed 270pF as specified and from the same batch. Also L1 and L2 do end up in approximately the same position after peaking. The 'not tuning accurately', was my misunderstanding of what was happening. Basically the frequency adjusts fairly gradually over several turns but once I got below the 1MHz mark, it would drop off through 900kHz, 600kHz and then zero very suddently over less than a turn. What was actually happening was that as I was adjusting L1, the increasing mis-alignment between L1 and L2 was causing the amplitude of the output to drop off beyond the point where the scope's built in counter could still measure properly. An adjustment of L2 to bring the two slugs into alignment to bring the amplitude back up (or of the scope Y amp control to increase the height of the signal on the display), allowed the counter to work correctly again. Having said that, the lowest frequency I can get is still around the 1.039MHz mark. From that point, whether I adjust up or down, it climbs again, which, as you point out below, makes perfect sense.

Originally Posted by Ian - G4JQT

Does the output clearly peak when you adjust the core of L2? I've not managed to do it, but it might be possible if L1 is at its lowest frequency and L2 tuned to its highest that it's tuned to the second harmonic. That's why C1 and C13 should be closely matched to ensure the tuning ranges of these two tuned circuits are similar.

Yes it does.

Originally Posted by Ian - G4JQT

As with many slug-tuned resonant circuits it's possible to get the same inductance at two points at frequency extremes. To avoid this just keep to the suggested frequency range.

I now realise that makes sense. The highest inductance will be at the point when the ferrite core is dead centre of the coil. It will decrease regardless of whether the ferrite bead is screwed down or up.

Originally Posted by Ian - G4JQT

A scope if connected directly to the antenna, will significantly detune the output tuned circuit and give false peaks. Adding more wire for the antenna is not likely to give a stronger signal - something is not right.

Thanks for pointing that out. That was indeed the case. The scope display allowed me to get close to alignment, but after disconnecting it, I had to tweak the tuning of L2 by ear to get the strongest and clearest signal. I could perhaps have used the VTVM.

Originally Posted by Ian - G4JQT

Are you using your own layout or one of the PCBs? Performance-wise it shouldn't matter, but it's harder to make an error with the PCB.

I purchased a PCB from this forum.

Originally Posted by Ian - G4JQT

I changed R11 to 3.3k as it seemed to give very slightly better output, particularly when I changed from 2N3819 (which has quite a wide performance spec) to the BF256.

That is interesting to note. I did wonder whether 2N3819 could be used. I am using BF256 and currently have a 10k resistor in that position. Since I had just removed a 3.3k from the LED circuit, I re-used that in the R11 position bringing it into line with the new diagram.

The project is now complete and working and I can easily pick up the signal on a radio 3 or 4 feet away. Haven't tried across the room yet...

[Ian - G4JQT]

Originally Posted by vinrads

I am reading this with interest Ian, you say connecting the scope will de tune the coils is there a recommended way to use the scope to set the coil ,or would it be better to use a radio and monitor the AGC line . Mick.

The scope will only detune if you connect directly to the antenna, but clipping around the insulation is fine. But all you really need to do is tune as described in the instructions - for the strongest, clearest signal on a nearby radio.

[Ian - G4JQT]

Originally Posted by WaveyDipole

The project is now complete and working and I can easily pick up the signal on a radio 3 or 4 feet away. Haven't tried across the room yet...

That's really great! Looks like you've done a nice job there and glad my reply helped.

[line sync]

I would like my mini mod to be mains operated and would like to fit a small 6v 0 6v transformer (12v) , bridge and smoothing cap inside the box next to the mini mod circuit board.
Is this likely to cause any interference problems ?

Robin

[Ian - G4JQT]

Originally Posted by line sync

I would like my mini mod to be mains operated and would like to fit a small 6v 0 6v transformer (12v) , bridge and smoothing cap inside the box next to the mini mod circuit board.
Is this likely to cause any interference problems ?

Robin

That should be OK, but it runs for ages of a battery! As described in the article, a mains power supply is more likely to generate modulation hum. To reduce this put say 10n caps across each diode in the bridge rectifier, across the low voltage winding and the electrolytic smoothing cap.

An X-type across the mains input is also a good idea.

Modulation hums is quite a common problem with all sorts of panty transmitters. Its cause is often hard to track down. Something I should write up for the Bulletin some time...

[line sync]

Thank you for the reply Ian.
As my mini mod will not get a lot of use the battery may get forgotten and i don't want to open the box to find it has leaked.
I will order some caps.

Robin

[keland_uk]

Its a shame I don't have access to the Winter edition of the BVWS Bulletin (2023?). Pictures of my effort, shoe horned into a an old Tobacco tin... R9 was increased to 4.7k to get my LED consumption down to around 1mA. Power consumption was around 18mA @ 9V, down to 12mA @ 5V. Tuned fine at 1.2MHz

[Ian - G4JQT]

Originally Posted by keland_uk

Its a shame I don't have access to the Winter edition of the BVWS Bulletin (2023?). Pictures of my effort, shoe horned into a an old Tobacco tin... R9 was increased to 4.7k to get my LED consumption down to around 1mA. Power consumption was around 18mA @ 9V, down to 12mA @ 5V. Tuned fine at 1.2MHz

Lovely job. Thanks for letting us know it went ok - assuming it did!