Listen to Latest SolderSmoke Podcast

Tuesday, May 31, 2016

NP0 Is the Way to Go!

Much to the consternation of Pete "Digi" Juliano, I have been working on analog LC VFOs for simple superhet receivers.   As described in earlier posts, I recently converted an old Barebones CW superhet to 40 meter SSB.   At first, the VFO (2 -2.3 MHz) was not stable enough -- it would slowly drift in frequency. ("We have a solution for that," chuckled Pete.)   My first effort at stabilization involved replacing the toroidal coil.  The material in the core is sensitive to temperature changes and this can lead to instability.  I found my traditional cardboard tube from a coat hanger, and made a coil of the needed inductance (you can see it in the pictures).  This yielded some improvement in stability, but it was still drifting.  

Next I tried taking out all the silver mica and disc ceramic caps in the LC circuit of the oscillator and replacing them with NP0 ceramic caps.  The feedback caps are in the box below the tuning cap, but you can see some of the little NP0s on the outside of the box, connected to a rotary switch.  This serves as the equivalent of variable "Bandset" variable cap, with the tuning cap serving as the "Bandspread."  I have seven switch positions, each covering about 40 KHz (with some overlap).  This gives me all of the phone band and the bottom 30 kHz of the CW band.  

Switching to NP0 caps really did the trick.  The receiver is now very stable. When I told Farhan about my VFO woes, he mentioned that he'd had very good stability results with surface mount caps.  I wonder if this success has more to do with those caps being NP0 than with their surface mount configuration.  

Here is a good description of NP0:

NP0 Ceramic Capacitors are single-layer ceramic capacitors made from a mixture of titanates.
A NP0 ceramic capcitor is an ultrastable or temperature compensating capacitor. It is one of the most highly stable capacitors. It has very predictable temperature coefficients (TCs) and, in general, does not age with time. 
NP0 stands for negative-positive 0 ppm/°C, meaning that for negative or positive shifts in temperature, the capacitance changes 0 part per million, meaning that it has a flat response across a wide range of temperatures; the capacitance of the NP0 capacitor stays constant (at the same value) despite variations in temperature.  


But I think it is a stretch to claim that these marvelous caps do not "age with time!"  That would be a really astounding property of the titanium dielectric.  That would be a Negative-Positive Zero FLUX capacitor, right?


  1. NP0 are the ones to use to ensure stability with changing temperature.

    However, lets not forget that the variable tuning capacitor will also exhibit drift with temperature - likely a small negative change with rising temperature. In days of yore designers of analog VFO's would spend much time trying to match all of the pluses and minuses of temperature effect by mixing NP0 as well as Pxxx and Nxxx temperature coefficient capacitors to try and get as near a perfect zero drift with temperature. The Nxxx and Pxxx (xxx is a number such as 100 to indicate the change with temperature). These N and P capacitors can still be found but are not near as common as they once were. check out the vfo schematic for the old Kenwood TS-520 or Yaesu FT101 for examples of mixing N and P capacitors in a VFO circuit.

    Also, you can get phenolic and ceramic toroid forms both of which will far better temperature characteristics than the powdered iron types while allowing you to keep the toroid form factor. W8DIZ sells some of the phonelic types.

    cheers, Graham ve3gtc

  2. Bill,

    I played with making a VFO using NP0's a couple years ago. I didn't go the next step of trying to add positive or negative coefficient capacitors to truly zero out drift. But, I was very pleased with the end results. My little direct conversion 40m receiver has very little drift after a ~10 minute warmup phase. Something on the order of 1 Hz/hour. It was so small as to not be noticeable on lengthy QSOs. For my money it wasn't worth trying to make it any better than that.

  3. Negative-Positive Zero FLUX capacitor...that's, if we can just feed it with 1.21 gigawatts and get this thing up to 88 miles per hour!

    Good Stuff!
    Bill McMillan


Designer: Douglas Bowman | Dimodifikasi oleh Abdul Munir Original Posting Rounders 3 Column