building ATF09 as predrive for SDR

[Robin Ivetic]

Hello

I would like to build pre driver for Razvan amp 600W or your 101 amp, using your AFT09 5W amp, now I don't know what to choose between MMIC or without MMIC version, except  (correct me if I'm wrong)  logic tells me that MMIC  option will generate less heat, and PCB and elements will last longer.

 At Rf Input side would be attached SDR cards, Lime, hackrf, both can give max around 15dBm between 1-30MHz. If I'm using MMIC version need to put attenuator, I learn that can use ceramic resistor, SMD. I'm looking online calculator what option I need to choose Pi Attenuator calc???, and what I need to enter 15dB, 50ohm impedance, something else, for example if I choose option 15dBm, I will get R1 =71.629 and R2 =136.140. Where I need to solder those resistor with two legs each on predicted place  R8,R9 and R10, for MMIC version R8 is populated with CR1206-J/-000ELF.

My whole connection schematic would look like this:

SDR card Rf out ---coax---Rfin--ATF09—Rfout-----coax---Rfin-amp600 or your 101 (I don0t need to use attenuator on Apm600 side, because it's less or equal 5-6W)

Don't know if you remember, but you find some part, which is manufactured by another code PHA-101, and purpose is the same as your AFT board, but for NXP 25W MRFE6VS25NR1. You also looked photos from that amp, and you said that your AFT09 board can replace that IC  PHA-101, but to operate he uses also couple of other elements which are  soldered on PCB, if I'm using your ATF09, I just connect your board true coax, or I need to remove or bypass those already populated elements on PCB board.

I read couple of posts regarding ATF09 topic, and question would summarize what I like to know, because I'm layman;

Robin.

[jcve...@gmail.com]

Hi Robin,

The decision to use the MMIC pre-driver depends on how much gain you need. If you need 5W to drive the Razvan amp that's 37 dBm if you SDR makes 15 dBm you need 22 dB gain. 22 dB is more than the ATF09 stage alone but less then the MMIC + the AFT09 so you will need an attenuator, 15 dB sounds about right. A PI attenuator has three resistors. One 136 ohm and two 71.6 ohm for 15 dB according to your calculator. You must remove R8 and replace it with the 136 ohm and add the two 71.6 ohm resistors R9 and R10. You can use the nearest standard values.

Your routing sounds correct but you will need to switch the amplifier out of the path for receive.

The PHA101 is not compatible with the 5W driver board because it required 9V and the driver board provides 5V  for the MMIC. The  MRFE6VS25NR1 is a 50V part and also not compatible with the RF Power Tools driver amp.

Regards,

Jim WA2EUJ

[Robin Ivetic]

Hello Jim,

Technically Razvan says that I need "drive stage with 20-27dB of gain", means that I can boost that just with ATF09 (without thinking about attenuation), but another limit that I read in assembly instruction with red markup text, Never exceed of drive to the amplifier 5-6W, means that I can also use MMIC version which I like to choose, but need to handle input gain, schematics says that I mustn't exceed 10dBm, if we talking here about linear measurement, I need to attenuate only 5dBm, correct me if I'm wrong cause max output that SDR can give is 5-15dBm!!!??? from 0-30MHz. so which values need to calculate for Pi Attenuator calc, 15dBm -10dBm=5dBm or 15dBm, which you confirm in my first mail, this 15dBm that we mention in first mail is just for input size or it's also for output, because your math suggesting that you attenuate 15dBm, because MMIC can give 37dBm wich is more that mine gap of 22 dBm  or shortly do I need to be afraid that I will over run main amplifiers fets, and above all what would be optimal values for attenuators considering 15dBm max that can SDR give, and max around 37dBm that can main amp handle.  

I read somewhere in post that one issue was with over heating, now I don't know if I'm going to have issue with that; I learn something from you, and also reading around net about that topic, but you mention that you suggesting for inRf and outRf cables like RG-316 or RG-174, I'm having now RG-58, can I expect any issue with this!!!??

Robin

[jcve...@gmail.com]

Hi Robin,

If you are concerned about overdriving the MRF-300 amp, which is a valid concern, leave it out of the circuit. Drive the AFT09 into a 50 ohm load and measure the power. In theory a signal MOSFET amp with a 4:1 output transformer running on 13V can only make 6.8W no matter how hard you drive it. Drive the AFT09 amp with your SDR and see how much power comes out. then figure out how much attenuation you need so that the max output from your SDR gives you about 5W output. When you are comfortable that you have the drive level correct so that the max drive is about 5W then connect the MRX300 amp and start with the power as low as the SDR will go and slowly increase the power to see what you get..

73,

Jim WA2EUJ

[Robin Ivetic]

Hello Jim;

Understood everything, except I don't have power meter, think that I have somewhere dummy load 50 ohm. I have pair of multi meters, do you know at which point I can connect positive probe on ATF09, closest to RFout, and that I don't burn something, so that I can read current at that point at that specific time, voltage is not issue to measure;

Robin

[jcve...@gmail.com]

Hi Robin,

Attempting to develop, build, tune and operate RF power circuits without a the ability to measure forward and reflected power is like trying to weld with you eyes closed. I would recommend that you not even try to build an RF power amp until you have the ability to measure RF power.

N5ESE has an excellent website on building and using an RF probe:

https://www.n5ese.com/rfprobe2.htm

VA7OJ also has a nice page about measuring RF power:

https://www.ab4oj.com/test/pwrmeas.html

Another easy to build wattmeter:

https://www.qsl.net/z33t/rf_powermeter_0-2ghz_eng.html

Here is an instructable:

https://www.instructables.com/RF-Load-With-Power-Meter-for-HF-Radios/

Please before you start working on an amplifier buy, borrow or build a wattmeter.

[Robin Ivetic]

Hello Jim,

If I will not buy it, maybe I can ask someone for help, but this one you can help me because mouser.eu don't have this item on stock, backorder with expected date around 24-07-24, I tried to dig similar item, and anything close to this by foot print is this, and still is double than original, other values except inductance are not even close, can you offer me substitute for inductor on Mouser, that I don't wait until end of July.

Sorry for bunch of mails, I'm study this for couple of hours, and like to end it. You are safe with me, until I get shipment from Mouser;

Thanks, Robin. 

[Robin Ivetic]

Hello Jim,

Do I need to isolate this part when applying with Al radiator, regarding it's +/- from supply and RF in/out, with termal pad, or just don't apply this side with radiator at all

Also c14 and c15 need ti be populated because I'm choose pre driver and R8,R9 and R10 are populated

Can I put this replacement ( CDRH4D18NP-220NC ) for Coilcraff inductor, because foot print is similar, and values except resistance!!!???

Robin

[jcve...@gmail.com]

Hi Robin,

Yes you do need to isolate that part of the PCB from the heatsink. I looked at the datasheet for the  CDRH4D18NP-220NC all of the spec that they give look OK. They do not list the SRF (self-resonant frequency) which is >75 MHz for the Coilcraft part. If the coil is self-resonant at a lower frequency and you inadvertently operate at that frequency, the amplifier will oscillate and may destroy the MMIC.

73,

Jim WA2EUJ

[Robin Ivetic]

Hello Jim,

I can't get more than 0.85A at VR1, multi meter is set to 2A, do u have suggestion what I need to look first, think that resistors, capacitors and dc-dc regulator are fine.

Robin.

[jcve...@gmail.com]

Hi Robin,

The bias is supposed to be set to 0.1A, why do you think that 0.85A isn't enough? I would advise turning the bias down to 0.1A.

Jim.