Moderm RF transistors in SMD packages.

[EFR]

I have been banging my head to desk last couple days.

I want try build SMD rigs, if others can do them, why I cant!

I did choose one of my old VCO+driver stage as test, desining PCB part for driver was easy, I did choose biggest SMD parts from local shop, easier to etch it and solder them.

But then the shit hit the fan, what I use to replace 2N3866 in nice SOT-89 package?!

I have been reading datasheets, only one what I did find was BSP42, ft 100Mhz.
It would be best replacement on HF bands, but what I use for VHF?

Is anyone made listing for these or anything?

Yeah, old fart is going to learn something.

[EFR]

Aaaaand it looks like BSP42 is out of production.

[jvok]

Rd01mus1 is the go to

Aft05ms004 and rfm04u6p look usable but never tried them.

[EFR]

I want stay transistors, just quick try to SMD work by usingmy big hands.

And on my mind, easier operate on linear range.

[Albert H]

BFG135 works well. We use a pair of them in a push-push circuit for 1.5 Watts easily - running cold.

[jvok]

BFG135 is obsolete now too

RF power BJTs seem to be going the way of the dodo, everyones gone to mosfets now. Higher gain, more robust against SWR. Only downside is fairly high spread of threshold voltage and transconductance means you usually need to make bias adjustable with a pot, where you could normally get away with fixed bias with BJTs

[1608cc]

I am currently working on similar design. The VCO will base on BFR520 and BB133 for modulation and tune. PLL on TSA5511 or 5512, configured from Attiny25. Then driver also on 520, and 2x BFG135 or 197 (can put BFG591 here) final stage producing 1 - 1.5W (I hope). Ofc LPF following the PA output. I get rid of regular LC circuit in VCO and put RG178 or RG316 in its place. I did tests with 15cm and 20cm coax and works pretty well. There will be second board connected in stack (sandwitch desing) with MPX encoder based on Attiny828.

We use a pair of them in a push-push circuit for 1.5 Watts easily - running cold.

Could You share idea with us? (I mean schematic - will be helpful to pick the values of impedance matching circuits )

[Albert H]

Word of advice with the coax resonator for the VCO - be careful where to route the coax. You have to bear in mind that it can be really microphonic! The slightest tap on the ox (or the vibration from a fan) can easily appear in the modulation.

Regarding the twin BFG135 stage, I'll scribble it down for you and put it up here shortly.

[1608cc]

Thanks Albert - I wait patiently (not true )

Coax will be clamped to pcb with U-shape ears / or glued.

[EFR]

Thanks Albert - I wait patiently (not true )

Coax will be clamped to pcb with U-shape ears / or glued.

One commercial transmitter had VCO semirigid coax between two 5mm thick pieces of Armaflex like foam.

Foam was glued to the pcb, and second piece was glued top of it.

[reverend]

There's a PKBee design that uses a piece of coax for the resonator. The circuit of the oscillator is below. Never quite sure why they use so many varicaps to tune it.

Screenshot 2025-01-31 103507.png

[EFR]

Quick Google search: 8x varicap voor een hele mooie sound!

But these are just on PLL side...

[1608cc]

Coax LC circuit has very high self-resonance frequency vs its length.

obraz_2025-01-31_170320854.png

To get 88-108 range using only coax it gonna be ~30cm. So my guess is they use shorter cable (because the available space) with many varicaps to drop down the oscillator frequency with way larger capacitance ratio in circuit.

obraz_2025-01-31_170300357.png

[yellowbeard]

This guy has a similar procedure but the coax is actually coiled up, which is microphonic like Albert says - vibrations from fans and whatnot get into the audio. Bang it with a trim tool I says!

http://mkmbroadcasting.blogspot.com/201 ... m-vco.html

[reverend]

My point about the multiple varicaps was not that it didn't somehow 'geeft een erg mooie sound' but that surely it would just be easier to use less varicaps each with a higher capacitance value!

[1608cc]

it would just be easier to use less varicaps each with a higher capacitance value!

They are hard to get - even BB809 has ~40pF which is quite high value such for a VHF varicap (way higher than BB405 fg.). Higher capacitance varicaps are intended for AM tuners tunning, so they are almost obsolete nowadays.

[rigmo]

it would just be easier to use less varicaps each with a higher capacitance value!

They are hard to get - even BB809 has ~40pF which is quite high value such for a VHF varicap (way higher than BB405 fg.). Higher capacitance varicaps are intended for AM tuners tunning, so they are almost obsolete nowadays.

as Albert H say 30V varicap supply solve all problems..

[1608cc]

as Albert H say 30V varicap supply solve all problems..

Where? I don't see any post about that here (or you mention another topic?).

Anyway - varicap voltage high as 30V is practically useless. Basing on BB809 characteristic, 1-10V range give us 30pF deviance of capacitance, where 10-30V only 8pF. Also the highest capacitance (equals to lower freqency) is at lower bias voltage.

obraz_2025-02-01_002913162.png

[EFR]

Old TV tuner varicap BB639 goes to 36pF, and is around 70cent each here.

BBY62 is douple varicap, seems to give 16pf so both caps together 32pF, price also around 70cent.

BB809 seems to be around one 1€ here, well known shop sells 2€/each, and I belive these are genuine.

In Holland, everything look like even cheapper, but shipping for some transistors and other stuff is very expensive, 45-75€ for 150g parcel...

[1608cc]

Smd close equivalent of 809 is BBY40 or BB133