Choosing Transistors is Bewildering

Discussion in 'Homebrew and Kit Projects' started by N5HXR, Jul 30, 2020.

ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: L-MFJ
ad: Left-2
ad: FBNews-1
ad: Left-3
  1. N5HXR

    N5HXR Ham Member QRZ Page

    I've received such great help from this forum, so thanks in advance!

    I'd like to learn about amplifiers. I have built a few class A amplifiers on a breadboard using the ubiquitous small signal choices (2n2222, etc.). But I understand these are not ideal. I'd like to put together a BJT-based class AB amplifier to experiment with, where I can change parameters and measure things.

    A lot of the textbooks, guides, or circuits I find specify transistors that are no longer made, have to be obtained from specialty shops, were harvested from a dead CB radio, or the author says are less than ideal, etc. I saw that digikey has a category for RF transistors, which seems like a good place to start, but a lot of them seem to be intended for UHF and beyond.

    So, what's better?
    1. Try to hunt down these obsolete or unusual transistors other people use
    2. Try to find a non-RF transistor with the parameters I think I need
    3. Use these UHF transistors for my little HF experiments
    I find #1 pretty frustrating, and I honestly don't know if people just use those parts because they have them, or if they really are ideal somehow. For #2, I don't know what parameters really matter (I know gain bandwidth product... but are there other showstoppers?). For #3, I just don't know if it's bad to use something with a transition frequency of 8GHz for a 10MHz project.

    Or, maybe put another way, what in-production transistor should I buy from digikey that will not shoot me in the foot trying to build a 0.5-1W amplifier to learn from?
     
    N0TZU likes this.
  2. VK4HAT

    VK4HAT Ham Member QRZ Page

    With BJT transistors, you want a minimum FT of Frequency * gain, this will equal the potential bandwidth of the circuit. So say you want an amp stage that has 10x gain and a max frequency of 30Mhz, 30mhz * 10 == an FT of 300. But 10x gain might not be enough for an IF stage, you might need 20x gain and that means an FT of 600 at least. There are some BJT's for cents a piece on digi/mouser with an FT of 1000, that is where I would start if you want broadbandness.

    Mosfets rated for UHF and up will work fine at HF, even though they are not characterized for those frequencies. They are cheap, buy a few different ones and then characterize them at HF yourself. Pick the ones that are cheapest and work the best.

    People use parts because they have them, others used them, they are cheap, not always because they are ideal. Once i got away from emulating other peoples schematics and started to understand how to work out some things for myself, life became much more enjoyable. Think toriods and wire sizes and the seemingly random way people choose core and wire sizes and if you dont know, you end up copying what the schematic says for no real reason. I have trays of parts on the wall behind me from doing things like that. LOL take the 2n3904 2n2222 bc547 etc etc, all exactly the same transistor. A lot of this only comes with asking lots of questions and reading books.
     
    Last edited: Jul 31, 2020
    KN4LGM, N0TZU and NQ8J like this.
  3. K6BSU

    K6BSU Ham Member QRZ Page

    These days, you have a choice of thousands of transistor types. If you are not up on all the latest, you should research designs similar to what you're looking for,. Then use whatever devices are specified. It saves a lot of time and mistakes.
     
  4. PY2RAF

    PY2RAF XML Subscriber QRZ Page

    Hey Josh! Congratulations for running the extra mile experimenting with amplifiers!

    I would only chime on item #3. This comes from RD70HUF2 datasheet (an VHF/UHF MOSFET power transistor):
    I'd also suggest you a few resources:
    1) This NICE "Understanding RF Data Sheet Parameters", gives a few hints, insights helping you make the right decisions: https://www.nxp.com/docs/en/application-note/AN1107.pdf
    2) Facebook Amp Builder community; plenty of high power pr0n over there, both vintage and solid-state latest components: https://www.facebook.com/groups/593515501007579/
    3) The obligatory MRF101 starter kit: https://www.digikey.com/product-detail/en/nxp-usa-inc/MRF101AN-START/568-15353-ND/10136497 . Read the Datasheet, provides instructions on how to use the board, the right place to put the capacitors, inductors etc from 1.8-250 MHz

    I really hope that helps - and you get to a successful build!!!!

    - RF.
     
    KA0HCP likes this.
  5. N5HXR

    N5HXR Ham Member QRZ Page

    OK, so should I infer that FT is the most important aspect? I mean, I know that max power dissipation, voltage specs, and current details are important too, but I feel comfortable reasoning through those. So In general, if it's a BJT, has an appropriate FT, then it should be OK?

    It does seem difficult to find one that can take a watt or so, has FT that high, and costs less than, say, $5 apiece. Is there one (or more) that you might suggest?
     
  6. N5HXR

    N5HXR Ham Member QRZ Page

    Would that same caution apply to a BJT? Right now I'm trying to understand the BJT, and am trying not to look at FETs yet (Experimental Methods in RF Design treats them separately, and I'm trying to go in order!). I found the BFU590G -- a $0.35 BJT, can dissipate up to 2W, etc., but it's got an FT of 8.5GHz. That kinda' made me wonder if it would be a mistake at HF. None of the books I've read have cautioned about FT that's too high.... but then, they often leave other things out I've had to learn the hard way.

    And that MRF101 looks really neat, but at $16, each, I can't really say that fits in the experimental budget at the moment!
     
  7. N5HXR

    N5HXR Ham Member QRZ Page

    Actually, after a few more Google searches, I see a few parts on the EMRFD group suggesting that the BFU590 may be a suitable alternative to the 2n5109 that shows up in the book. So maybe that is an ok choice after all.
     
  8. VK4HAT

    VK4HAT Ham Member QRZ Page

    Of course, you would take all things into consideration, but with BJT's i would only use them in small signal amps situations. For making power, anything say 0.5w and larger, then mosfets are the go. Even the IRF510 fets will get you to 5 to 10w and are cheap as chips and good to 30mhz with good design. Fets like the RQA0009 can be found for 50 cents each and are good to a couple of watts linear. Its a VHF fet used in the baofang and will do 5w in non linear modes. With the RF mosfets pay attention to supply voltage, many of them will require odd ball voltages like 7V.

    After those, you are then looking at $5 and more for rf fets. What you will learn from the IRF510 will translate to everything else. The IRFZ250 will get you to 100W + on 40m and it costs less than a dollar. If you can make those work, then you will make more expensive fets like the MRF101 work.

    Rob.
     
    N0TZU likes this.
  9. K6CLS

    K6CLS Ham Member QRZ Page

    Heh, I think I know which book you're working from ;)

    Good to know about BDU590s, thanks.

    If you want 2N5109s, Anchor Electronics in Santa Clara has NOS for not too much $. They also had J310s.

    I had some success with IRF510s, but some frustration too. They are static sensitive so be careful when prototyping. Also seem to be delicate in some other ways, heat, Vdd, current, etc..

    Enjoy!
     
  10. KB0MNM

    KB0MNM Ham Member QRZ Page

    Mr. Stone- See post #4, item 1- ( That appears best ) or "Secrets of RF Design" by Joseph Carr- Tab Books. Maybe the library system there in Livingston county or elsewhere could get you a copy to look at. If you think that you have all of the fundamentals of BJT circuit configurations down, it may be time to look at some related factors: capacitors as a means to change the gain of circuits, effective series resistance ( actually part of 'lumped' circuit values - which applies to almost everything- that resistor will also have a certain inductance ), source/load impedance for standing wave effects/distortion, noise created by thermal conditions, etc. As one of the others stated here on 'the Zed', "Theory only works in a vacuum". While not entirely correct, it conveys that nothing here on earth is perfect. I sometimes have to explain "Pobody's Nerfect" ( No one is perfect ) because there is a trinity also known as The One. That is a different story.
     

Share This Page