Dutch RF - 6w DRFS06

[mauri]

Nice that you fixed it

A 5mm aluminium plate below the tin can is enough to keep the rd06 cool.
So you dont have to buy a heatsink only a piece aluminium in you local hardware store.
My advice, first finish the transmitter in the tin can and than start tuning the coils.
It will make a big differance in output power.

Ok thanks guys. On the edges of the band I'm getting about 9w. In the middle I'm getting 3w. I can only test for a few seconds due to the lack of heatsink.

Next step cooling. I'll see if I can grab one cheap on ebay and start testing. Then I'll grab some feedthroughs and come back with an update and some more pics.

Cheers guys.

Lazy.

[LAZY]

Ok guys, It's boxed and working, heatsink paste on the final. The feedthrough capacitors are installed (pain in the ass job that, drilled the holes too big and had to fill with solder) and a phono plug for the audio. Not sure if the feedthrough caps need to be oriented, but I installed them all one way round and they seem to work ok. I managed to get a cheapy heatsink off ebay that was almost the perfect size for a couple quid, and it runs cool.

With regards to power output, in the middle of the band it's doing 12-13 w at the lower end 7w and at the very top 6w. Should I try for more than this at the ends of the band? I'm only using 12v, but have a switchmode on order so could supply 13v and test again?

The audio sounds good, nice quiet carrier no humming or buzzing. I've managed to track down an enclosure, which I need to go and collect and will encase it in one neat package.

With the LCD, do I need to clu a slot for the cables, or install more feedthrough caps? Or can I just squeezy the lid on carfully and be done with it?





Thank you for everyones help, I couldn't have got this far with out everyones comments.

What should I have a go at next??

Lazy.

[pjeva]

Glad that it works! It's a 6w device so don't push it over. You have more power in the middle of the band and it is normal, but doubled power is bit too much. I think it's harmonic content that shows more power. That could be determined on spectrum analyser. Low pass filter maybe needs a bit of spreading coils, maybe you could try first and last one...

[teckniqs]

Can you take another photo of it facing down on to the board?

[Maximus]

I wouldn't push it. Those RD05HVF1 mosfets pop easily.

[mauri]

DSC06055 (960 x 540).jpg

DSC06055 (960 x 540).jpg

For the display cable just make a cutout in the tin enclosure (see manual)
I dont think the high power is a harmonics problem.
Maybey your powermeter is a little optimistic...
Do you use a good 50 Ohm dummyload to test the transmitter ?

I normaly tune this transmitter for maximum power at the high freqs (108mhz).
The output power will be more equal than allover the fm band.
8 or 9 Watt is very normal for this exiter at 12.5 Volt in the middle from the band.
BTW i never blew a RD06 yet even when it runs above 10 Watt.
It's a very strong device (even survive changing the positive and negative lead )

[LAZY]

Turns out I don't know how to use my power meter. I had it set to 75 ohms not 50. Now the readings are more to be what's expected, with 12v in I get 7w in the middle of the band and 5w at the ends.

I do have a little problem with the audio though. It may be related to me fitting the 10K pot for volume where I should have fitted the 5K one? When the audio is low it crackles, it sounds fine when louder. Could this also have something to do with the coils not being canned as in Mauri's picture by the output, or that i've fitted a phono socket and not used a piece of coax to pass through the can again like Mauri's picture.

Here's what i've done so far.



Lazy.

[teckniqs]

When the audio is low it crackles but when it's louder it doesn't? Never heard of that before.

...It doesn't matter about the phono socket being connected via coax/shielded audio cable as the socket is nicely earthed by the enclosure.

[Shedbuilt]

Turns out I don't know how to use my power meter. I had it set to 75 ohms not 50. Now the readings are more to be what's expected, with 12v in I get 7w in the middle of the band and 5w at the ends.

This is quite strange. It should read higher when set to 50 Ohms, than when set to 75 Ohms. I'm fairly sure the Zetagi is taking a direct sample of the RF voltage on the stripline transmission line (not via the pickup lines it uses for Forward and Reflected power in SWR mode). As an example, 10V into a 50 Ohm load = 2 Watts. Into a 75 Ohm load, it would be 1 and 1/3 Watts.

I do have a little problem with the audio though. It may be related to me fitting the 10K pot for volume where I should have fitted the 5K one? When the audio is low it crackles, it sounds fine when louder. Could this also have something to do with the coils not being canned as in Mauri's picture by the output, or that i've fitted a phono socket and not used a piece of coax to pass through the can again like Mauri's picture.

From the diagram, I can see that the audio input is DC coupled to the deviation pot, and the other end of the pot is DC coupled to a grounded resistor (R22). If the crackling is the kind of crackling you get from a noisy pot, it could be that the output of your source device is also DC coupled, with the result that the DC offset is passing through the pot and R22. It might be worth experimenting with putting a cap in series with the audio in (eg between the centre of the phono socket, and the audio in pin on the PCB). The audio in - at least at the highest deviation settings, will present a low impedance, due to the low value of R22, so the capacitor value should be relatively large. It should really be in the region of 100uF, but you could probably go down to 47uF without noticing any difference. Lower if you're not using anywhere near the "full" setting on the deviation pot. In theory, it should really be a non-polarised type (or two electrolytics, double the value, connected back to back; eg +ve to +ve, in series, with DC bias via a high value resistor at the centre point), but a normal electrolytic will be fine for experimentation. It's unlikely the DC offset will be high enough to matter much which way round the electrolytic is connected, but probably best to check your source first anyway. If the DC offset is measurable, use the polarity of the offset, to decide which way round the electrolytic should go.

[mixfm]

is that lazy b. from birmingham by any chance;

[Shedbuilt]

Sorry. I didn't mention the obvious in my previous post. It could of course, just be a noisy pot (if it's that kind of crackling)....

[LAZY]

Your right Builtinashed. I've changed the source audio device and it sounds spot on! I will however try adding a cap in series so that I can use whatever I like as audio in without worrying. With regards to tuning the coils can you give me a brief guide on what's doing what?

Does one coil filter one end of the band and so on.....

I have now managed to get at least 6w at the upper end of the band and more at the lower end, although it dips around 89-90mhz. Then rising to 10w at 98mhz. I seem to be moving this flat spot up and down the band.

So far I think I've resoldered every coil at least once where the solder joints given up. With L3 I've removed it altogether and rewound it because it'd deformed from repeated bending.
Could the issue be where I'm stretching the could at the top and not the bottom? Would it be better to remove and makes sure the coils are identically spaced top and bottom?

Having what seems like an inaccurate meter and not having a spec is making this quite hard, there's only so much I can do with an SDR.

Thanks again. Lazy.

[LAZY]

BTW. You were also right about the power meter. 75ohms button in. 50ohms button out.

[Shedbuilt]

Must admit, it's a tricky one. Especially without knowing how another one behaves.
I would probably start by setting all of the output filter coils (L7 - L10), to the specified lengths. Check that all the output filter and output match caps are in the right places, and none are damaged (leads of ceramic caps can break away). Check the inter-stage too (between the sga device and FET).
If all seems ok, as a curveball, see if adjusting L2 alone has any effect on the frequency of the dip. If it does, you really want to try and push it below the bottom of the band. If it doesn't significantly affect the dip on it's own, I would tune the output match (L3 and L4). Ordinarily, I'd agree with Mauri, and tune for the top of the band (on the basis that the gain of the active devices falls with frequency). Then tune the inter-stage (L2). If results are the same, try peaking for the middle of the band, and reassess results across the band.
Coils should ideally be uniform, and if you're replicating a design (like the output coils), they really should be, but spreading and / or compressing just the top, will reduce and increase the inductance. In theory, you shouldn't need to adjust the output filter coils, so I'd at least try to optimise the matching circuits first....

[Shedbuilt]

Also meant to ask. If you use lower bias, and / or less drive, is the pattern across the band still the same ?

[pjeva]

I tried mauri's designs and LPF acts ok without any coil spreading (it could be even better but you cannot do it without spectrum analyser). LAZY, try to make new coils, and try to make them exactly like in manual. Do not spread them as much as you can, make little spreads like half millimeter. Try to spread only first and last coil and only first and last turn on them. Do not touch output coil on RD06, it is probably perfect match as it is....

[Shedbuilt]

I tried mauri's designs and LPF acts ok without any coil spreading (it could be even better but you cannot do it without spectrum analyser).

Yes. In an ideal world, you'd really want a network analyser too.....

[pjeva]

I tried mauri's designs and LPF acts ok without any coil spreading (it could be even better but you cannot do it without spectrum analyser).

Yes. In an ideal world, you'd really want a network analyser too.....

Actually, network analyser for home use can be built quite easy. You can use old cheap oscilloscope with x-y inputs, oscillator from this transmitter, ramp generator and directional coupler...

[LAZY]

Winding wire has been ordered and tomorrow i'll be ordering a new power meter (think i'm going to order the Daiwa C102L).

I've managed to sort the dip in power at 88-90 mhz and it now rises at the lower end and dips at the top. But not 100% what power it is making.
I've made a DIY attenuater from a BNC Tee piece by removing the center pin from T section and connected the power meter and dummy load one end and my RTL832u SDR on the other end. Setting the SRR to 0dB gain and turning the power on the TX up until the fundamental frequency reaches 0dB (approx. 3w on the transmitter) I can just barley hear/see the 2nd harmonic -(45dB ish, more If I apply audio) and nothing visible/audible for the 3rd harmonic. I can't see anything in band which is good??

Unfortunately If I turn the gain up on the SDR I can hear / See it easily. I take it this is just amplifying the received signal?
When you guys are testing, can you still hear/see the 2nd harmonic?

Lazy.

[Gum]

Not sure but those RTL dongles overload easy and when they do, you start seeing the strong signal all over the place... It's pretty easy to get them to do this with fairly low power PMR signals from a fair old distance as you increase the gain, so you may need to do a bit of trial and error before you start to feel you know what you are looking at.