How to choose new Mosfets for Amplifier on 500Khz

Discussion in 'The Projects Forum' started by Patty Morr, Oct 27, 2014.

  1. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Dear All,

    I had found the schematic for a 150Watts mosfet amplifier for Hf Bands here: http://radiotengkorak.blogspot.ca/2012/05/150-watts-mosfet-irfp250-rf-linear.html

    It seems like the designer uses different transformers for DC and the Output network which seems very close to another power amplifier used for 500Wts on 2Mhz. I have reading also that using DC coupling capacitor in series with drains will help for unbalance on the mosfets, the link above uses parallel capacitors instead of series. In this case I am considering to change the mosfets for this ones STP16NF06 ( 60V/16A,0.1Ohm , 315pF In , 70pF Out, Cross 70pf, Qg 13nc) or maybe this ones IPB06N03LA (25V/50A,5.9mOhm , 1995pF , 848pF Out, Cross 124, Qg 14nc). So I want to know which criteria is required to choose for the best mosfet for an amplifier in the schematic attached. My goal is to have a portable amplifier which delivers around 50 or more Watts on the 500Khz to 1800Khz using two 12 volts batteries(in series as dual power - or a single 24DC) . ( I have build a driver using the Lt1210 (10Ohms imedance) from linear which is able to provide 3.5Watts on the mentioned band with the +/- 12 Volts.

    Any suggestions are really appreciated

    Regards

    Patty
     
  2. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Forgot to include the 500wts schematic
     
  3. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    Patty, I noticed that no one has responded to your posts. So, even though your application is a little out of my normal comfort zone, I will give it a shot.

    First, your frequency range transverses the AM broadcast band. Would you mind if I asked what your intended use of this amplifier is?
     
    RamaD likes this.
  4. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Dear Lestraveled,

    Thank you for your reply. The amplifier is indeed intended for the Broadcast Band ( AM BC) and it will be use for tuning antennas. I have succeed with the Lt1210 broad band amplifier which is very robust for open load or short circuit load. Futhermore, I was able to use a special switched transformer array at the output of the LT amplifier so it is happy with what ever load I put on it on the BC band (i.e 5 Ohms or 120 Ohms Capacitive or inductive). Unfortunately 3.5watts is too low power to fight for interferences generated by other broadcasters in same area, lets remember this antennas are very large metal structures. So, basically I am looking to have a little more power to overcome for interferences and this is why I'm interested in a mid power Mosfet amplifier. At the beginning I will be happy if the amplifier can perform on the standard 50Ohms load; better yet if the amplifier will be able to perform on a 10 ohms reactive load so I can use the same switched transformer array I built for the LT amplifier.
    Best regards , Patty
     
  5. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,348
    Hello,

    Do you have a licence to transmit that amount of power?

    Bertus
     
  6. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Dear Bertus,

    The mentioned AMPLIFIER is not intended for a transmitter plant or to stay installed for long periods of time(lets say couple hours until battery dies). The amplifier will be part of the testing equipment needed to tune antennas.

    Patty
     
  7. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,348
    Hello,

    Why would you need so much power for testing an antenna?
    A low power oscillator and a SWR bridge should do it.
    It just depends on the sensitivity of the SWR bridge.

    Bertus
     
  8. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Bertus , Been in a tower site where you can see your neighbour's 8 tower array with 50Kw and 30Kw co-located on different frequencies on the AM band, and Just by touching the target tower with some tools and nothing connected to it, you will be able to see some nice sparks of Rf induction. For sure I had seen smoke coming out of low power oscillators which is not nice at all, and the worst is to have to repair it later on.

    To clarify the power I'm looking for the amplifier is not for 500Watts which is going to run the portable batteries very quick . A good starting point will be 50Wts.
     
    Last edited: Nov 5, 2014
  9. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Here is another Mosfet I tried on the Ltspice. This time I choose for a Mosfet in the Lt. library rather than importing libraries from manufacturers which leads to long time process simulation. In the schematic attached I used the STP8NM60ND, 600Volts, 7Amps 0.59 Ohm, 560pF In, 37pF Out , 4pf Cross, 22 nC. I was able to see some clean output signal of about 3V . The signal was constant from 500Khz to 2Mhz.
     
    Last edited: Nov 5, 2014
  10. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    This is a little update on my progress, while choosing for power Mosfets on LTspice I have found this one to be delivering some 30Watts to the 50 ohm load. Is the BSC340N08N53, 80V 23A, 34m Ohm, 567 pF In, 156pF out, 7 Crss, 6.8nC . It seems like two paralleled STP16NF06 I have on stock.
     
    Last edited: Nov 5, 2014
  11. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Seems like I will begin the construction of the attached schematic. The Ltspice simulation does not like the circuit response on the frequencies below at 1Mhz (just 30Volts Peak to peak Voltage); futhermore there is somehow a linear response from 1.5Mhz to 15Mhz. This time I used a set of paralleled STP16NF06 which provided some 90Volts peak to peak Voltage over a 50Ohms load(think that's power) with a input signal of 20Volts Peak to Peak. Bias is controlled by R4 , Simulation shows 3.99 Volts on all 4 Gates.
     
  12. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    I got rid of the excessive ringing by fixing the value of C5 from 680 Faraday to 680nF. Now there is better response from 3Mhz and up. The picture attached is for 10Mhz output response.
     
  13. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    By changing input and outp transformers, the amplifier is now on 1Mhz center frequency. Source of Just 1 Vpp on 10 Ohms cause the output to be 70Vpp on 50 Ohm and 60 Volts at 500Khz and 1800Khz. If input signal is increased from 1Vpp on the whole band, the signal saturates. Bias was increased adjusting R4 from 3.9Volt to 4.01V. I Will need some help now with the output transformer to be very efficient for a 10 Ohms load. What do I need to consider for the output impedance of the Mosfets in this configuration?
     
  14. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Please correct me if I found the wrong information for the output stage. It seems like L4 and L5 is a DC supply choke for RF and it needs to be at least 50 Ohm reactance on lowest frequency which the actual value of 16uH is just fine. For the circuit's push pull output impedance, RL = 2 x (0.9 VDD)(0.9 VDD)/PO (forgiveme not to use sqr) , with a Vdd of 24 Volt, Requiered PO of 50W , the output impedance should be RL = 18.66 Ohm , which is fine for now, because I am not looking for the standard 50 Ohm load; so 6uH makes 19 Ohm @ 500Khz and I can use a 1:1 transformer( or a 3uH for 10 Ohms with 100Wtts). Think this is perfect because I will need to protect the amplifier with an isolation transformer anyway with the arrangement for different impedances a BC's tower antenna may offer at different frequencies. Now I have to choose for a ferrite material core, have on hand a 77 material of 1" OD, 0.6" ID 0.3" Ht, or a 43 kind material 1.2"x0.7"x0.652.
     
  15. Patty Morr

    Thread Starter New Member

    Oct 24, 2014
    12
    0
    Seems there is another approach for the output transformer at the end of this post:
    http://forum.allaboutcircuits.com/threads/how-can-stabilize-the-drain-current-help.38174/

    Using Output V = √ (PoW x ZL Ω) With Po 50W, ZL 10 ohms and VDC 24V a 1:2 ratio is needed. Secondary XL= 10times 10 Ohm and rounding for 500Khz L=40uH on secondary; so it means the primary will be looking at 10uH , far away from the 3uH suggested in my previous post (or even far from 6uH on 19Ohm). Maybe the safe way to go now is for the highest value of L.
     
    Last edited: Nov 8, 2014
Loading...