RDVV Big Sprog Problem

Everything technical about radio can be discussed here, whether it's transmitting or receiving. Guides, charts, diagrams, etc. are all welcome.
User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

RDVV Big Sprog Problem

Post by Maximus » Sat Apr 15, 2017 11:59 am

Hi necks. I've got an rdvv 8w notune exciter.
Since day one it's had problems with sprogs.

When the frequency is set to 96.0 there's a massive sprog at 102.5 with roughly the same power as the main frequency. There's also another one at around 110mhz.

Thank goodness I decided to put it on an analyser before plugging an aerial in.

Obviously it can't go on air in this state. Any help will be greatly appreciated.

Included a couple of pictures of the board before it was finished.

Image

Image

Image


Sent from my iPhone using Tapatalk
Last edited by Maximus on Sat Apr 15, 2017 11:59 am, edited 1 time in total.

nrgkits.nz
big in da game.. trust
big in da game.. trust
Posts: 55
Joined: Fri Oct 17, 2014 10:35 am

Re: RDVV Big Sprog Problem

Post by nrgkits.nz » Sat Apr 15, 2017 1:13 pm

Is anything showing up on 89.5 ?

Also noticed you've used metal film resistors, I'm not sure how inductive yours might be, this could be a problem. Some one once told me not to use metal film resistors with RF, and I've used carbon film ever since. NRG boards all use carbon film.

I'm sure Albert will be around soon and will comment.

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Sat Apr 15, 2017 1:48 pm

Hi mate. Thanks for the quick reply.

Yes there's another harmonic on 89.5mhz. This one is roughly half the power.

The lot came as a kit and assumed that the supplied components should be ok. It's also sat inside a tin can to shield from the amp.

The harmonics change freq as expected when playing around with the coil, due to the pll obviously only locking to the desired frequency.

Anyhow I think you're probably right about the components.

Luckily there's nothing on the harmonic frequencies around here. It just seems silly to waste the power and have a radio station on two frequencies covering a small town.


Sent from my iPhone using Tapatalk

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Sat Apr 15, 2017 7:55 pm

Thought I'd add a circuit diagram for you as well. It not quite the correct one for this version of the board but does give you a good idea.

Image


Sent from my iPhone using Tapatalk

nrgkits.nz
big in da game.. trust
big in da game.. trust
Posts: 55
Joined: Fri Oct 17, 2014 10:35 am

Re: RDVV Big Sprog Problem

Post by nrgkits.nz » Sat Apr 15, 2017 8:30 pm

This sounds very much like intermodulation, your main fundamental is on 96 but there's also strong unwanted resonance on 102.5. The two are mixing to produce intermods on 89.5 and 109. You may find if you lower the resolution bandwidth there will be intermods also on 83 and 115.5. Be vary careful with the 115.5 if you're near an airport.

What happens if you remove it from the can and disconnect the amp?

From the photo it looks like the 1971 is not sunk down to the pcb as far as it will go. With RF try and keep lead lengths as short as possible, sinking the 1971 down to the pcb should also increase the power out slightly more.

Albert H
proppa neck!
proppa neck!
Posts: 682
Joined: Tue Apr 05, 2016 1:23 am

Re: RDVV Big Sprog Problem

Post by Albert H » Sat Apr 15, 2017 11:50 pm

You seem to have something self-oscillating at about 6.5MHz and mixing with the carrier to give those spurs. The 7809 could be a culprit. Try adding a couple of 100n caps input leg to ground and output leg to ground. These should be as close to the regulator as possible. The comment about the 1971 is also worth investigating.

Sent from my SM-G930F using Tapatalk

Is it meant to smoke like that? :shock:

MiXiN
tower block dreamin
tower block dreamin
Posts: 448
Joined: Sun Sep 07, 2014 4:20 pm

Re: RDVV Big Sprog Problem

Post by MiXiN » Sun Apr 16, 2017 12:27 am

This is from the Amateurradioshop.nl shop in the Netherlands, right?

I'm sure RF Head said some of his gear was pretty sproggy, but it was a while ago.

I've bought a couple of kits from this same shop, and although the PCB's are really decent - He's supplied a few snide components in the kits, like fake Chinese Transistors (2n4427 and a MRF237).

Albert H
proppa neck!
proppa neck!
Posts: 682
Joined: Tue Apr 05, 2016 1:23 am

Re: RDVV Big Sprog Problem

Post by Albert H » Sun Apr 16, 2017 12:52 am

Cheap counterfeit transistors don't generate spurii!

Sent from my SM-G930F using Tapatalk

Is it meant to smoke like that? :shock:

MC Spanner
who u callin ne guy bruv
who u callin ne guy bruv
Posts: 17
Joined: Thu Mar 23, 2017 12:01 pm

Re: RDVV Big Sprog Problem

Post by MC Spanner » Sun Apr 16, 2017 1:10 am

That circuit diagram is not the same as the board - the board has better matching between the BFR96 and the 2SC1971 and has a more sensible class C bias arrangement for the PA. I also have an idea that later versions binned the printed PA matching and filter inductors and went back to wound coils.

I'd guess you have a layout / earthing / decoupling problem. This board does have modest earth stitching in sensible looking places. Check you installed all the decoupling caps properly, and other stuff like making sure your can is soldered properly to the board all the way around. I am assuming that your 2SC1971 is now mounted correctly. This is VERY important! Leads as short as possible and bolted firmly down to a heatsink which is well connected to the same earth as everything else. I think the transistor is supposed to go underneath with its front facing the back of the PCB. If you bolt it down on top of the PCB, then its proximity to the printed inductor feeding the base may have an unpredictable effect.

Can you try a couple of things? Move the frequency by say, a couple of meg to either 94 or 98 MHz and see if the sprogs move by the same amount. If they do, you've got low frequency parasitic oscillations (6.5 MHz in your case) and if you identify the affected stage (probably the PA) then there will be various things you can try. Second, can you try and get a feel for how the presence of your sprogs varies with the setting of the blue "power" control. If the sprogs aren't there at low power levels and it works nicely, turn it up until the sprogs start, and then turn it back down below that point and see if the sprogs stop. If not, it's more than likely the PA you need to look at as opposed to the driver stage and a LF damping network or "stopper" resistor may sort it. But give the layout and construction some serious scrutiny first. And as ever, sorry if I'm telling you stuff that you already know or have tried. :)

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

RDVV Big Sprog Problem

Post by Maximus » Sun Apr 16, 2017 4:27 am

Thank you everyone for your suggestions. It's given me plenty to look at after a good sleep.

The transistor is correctly mounted under the pcb with short legs.

Yes the kit was bought from amateurradioshop.nl


Sent from my iPhone using Tapatalk

MiXiN
tower block dreamin
tower block dreamin
Posts: 448
Joined: Sun Sep 07, 2014 4:20 pm

Re: RDVV Big Sprog Problem

Post by MiXiN » Sun Apr 16, 2017 4:09 pm

Albert H wrote:
Sun Apr 16, 2017 12:52 am
Cheap counterfeit transistors don't generate spurii!

Sent from my SM-G930F using Tapatalk
I know mate, but I'm just throwing it out there per chance something else is amiss.

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Fri Apr 21, 2017 11:41 am

Found the problem. It's something to do with the power adjustment!

When I adjust the power the sprog frequencies also change. The sprogs disappear at almost full power.

The only problem is it can't be ran at full power as it'll be overdriving the 170w pa.

Have a feeling it's something to do with the BD139 circuitary


Sent from my iPhone using Tapatalk

nrgkits.nz
big in da game.. trust
big in da game.. trust
Posts: 55
Joined: Fri Oct 17, 2014 10:35 am

Re: RDVV Big Sprog Problem

Post by nrgkits.nz » Fri Apr 21, 2017 10:29 pm

The power adjustment is varying to DC supply to the driver only according to your schematic, the power should be adjusted by varying the DC supply to the final and driver stages so that the impedance matching remains the same as much as possible.
Z = (Vcc * Vcc) / (2 * power out).
Credit to Albert for explaining this formula to me some time back.

I would probably bypass the BD139 to get the same DC voltage to the driver as well as the final, then reduce the output of the power supply if possible, or use a RF power attenuator to resistively loose a few watts at the output.

I have been experimenting with adjusting the power output on my new design and the main problem with using an onboard linear regulator/darlington is the heat generated. There won't be much heat with the schematic in this thread because it's not varying the DC to the final, hence a lot less current to deal with.

Albert H
proppa neck!
proppa neck!
Posts: 682
Joined: Tue Apr 05, 2016 1:23 am

Re: RDVV Big Sprog Problem

Post by Albert H » Fri Apr 21, 2017 10:45 pm

I never liked the RDVV way of varying the power, because the interstage match will change. I'm surprised that the sprogs are big, but the way recommended above is probably best. Run the exciter at full power. Try reducing the supply voltage and see if it stays clean. Also try a resistive "pad" to lose a bit of the excess exciter power. One other thing that I've done in the past is also waste some of the power in a critically tuned filter to make the drive absolutely clean!
Is it meant to smoke like that? :shock:

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Fri Apr 21, 2017 10:55 pm

Thanks for the replies gents.

Do think that replacing the 150 ohm supply resistors for higher value ones to the bfr96 would help?


Sent from my iPhone using Tapatalk

Albert H
proppa neck!
proppa neck!
Posts: 682
Joined: Tue Apr 05, 2016 1:23 am

Re: RDVV Big Sprog Problem

Post by Albert H » Sat Apr 22, 2017 10:11 am

Maximus wrote:
Fri Apr 21, 2017 10:55 pm
Do think that replacing the 150 ohm supply resistors for higher value ones to the bfr96 would help?
To prove it quickly and easily, just unsolder one end of one of the 150Ω resistors and pull it out of the board. This will increase the resistance to 150Ω (it's 75Ω at the moment with the two resistors in parallel). The problem that you might run in to is that there may be insufficient drive for the final transistor - Class C stages need a certain amount of power to force the transistor to switch. There's no harm in a quick test, though.

If you find that you don't get much output with 150Ω there, solder the resistor back in and test again at full power to make sure that you haven't done any damage.

Have you measured the current consumption of the board? If it's 1A or less, get a 1N4001 diode and connect it in series with your positive supply connection. If it's more than 1A, use a 1N5401 (though I can't imagine that it will be). This will drop the supply by about 0.65V. Measure the output power (at full power) again, and you should find that it has reduced a bit. Try two diodes in series for a 1.3V supply reduction, and test again. You may find that reducing the supply like this - to the whole board - reduces your output power whilst staying clean because the "power control" section of the board is not doing anything.

If you find that everything's clean and happy at reduced voltage (and it should be), you should add a LM317 regulator circuit to the positive supply input. You'll then be able to set the output power by setting the supply voltage. You won't be able to reduce supply much below 11V because of the 78L09 feeding the oscillator stage - this may necessitate further modifications if 11V still gives you too much power.....

Give it a try and let us know how you get on!
Is it meant to smoke like that? :shock:

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Sat Apr 22, 2017 1:07 pm

Thanks for the reply Albert.
Disconnected one of the resistors and the output lowered to the appropriate level. However the sprogs have reappeared. Looks like there's something more sinister happening here.

The exciter is powered from its own 7812. However I haven't connected the negative. It's fed from the grounded casing which is bridged at the 24v smpsu if that makes any sense. Basically the gnd and negative have been linked with a bit of wire and it's going through the casing.

Also the amp pushes full power when it's cold but goes down to 100w after a few minutes (only started doing this last week out of the blue). It goes up when I touch the input matching caps and the harmonics disappear at the same time.

Perhaps I should be looking there as well. Will try connecting some caps and try to alter the matching.


Sent from my iPhone using Tapatalk

Krakatoa
ne guy
ne guy
Posts: 5
Joined: Tue Mar 07, 2017 8:57 pm

Re: RDVV Big Sprog Problem

Post by Krakatoa » Sat Apr 22, 2017 3:11 pm

Just a suggestion as it seems you have the trouble isolated: try disconnecting the 10nF capacitor that goes to ground from the junction of the 2x 150 Ohm resistors and 10uH choke coil that feeds the BFR96 transistor.
This way, the choke will be damped by the series 75 Ohm resistor, and will lower its Q and the risk of self-resonance.

I never liked undamped choke coils, they tend to resonate with whatever they have around...

And while you have the soldering iron hot... solder that 10nF cap from the base of the BD139 to GND.

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Sat Apr 22, 2017 4:40 pm

@Krakatoa you are the man!
Removing the 10nf cap has solved the problem!

I'm forever grateful to you sir :)

Here's another picture of the trace underneath if anybody requires it for anything.

Cheers!

Image


Sent from my iPhone using Tapatalk

User avatar
Maximus
tower block dreamin
tower block dreamin
Posts: 285
Joined: Wed Jan 21, 2015 1:29 pm
Location: Where you don't need licenses

Re: RDVV Big Sprog Problem

Post by Maximus » Sat Apr 22, 2017 4:46 pm

It's even solved the problem with the 170w amp losing power after 5 mins.

Also the SWR has nose dived from 1.4:1 to 1.15:1

Today is finally turning out good after all :)


Sent from my iPhone using Tapatalk

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest