Home Review Jcat Master OCXO Clock – Don’t Be Square!

Review Jcat Master OCXO Clock – Don’t Be Square!



  • Easy installation
  • Solid hardware
  • Fairly immune to power supply


  • Pricey
  • 20 MHz is of limited use

Price: € 1200

Build quality


We can confidently say that Jcat offers some pretty nerdy products. USB cards, Ethernet interfaces and now even a Master Clock to externally clock the USB XE and perhaps later a NET Card XE (?). The target market for these is, of course, very, very small. But for the tweaker, these are products that can put big dot on the “i”. Although they are not cheap. We take a look at the 1200 Euro Jcat Master Clock.

As in our lives, a clock in a digital device indicates time. However, a digital device does not care what time it is; it wants to see a constant rhythm. In other words, what matters in a digital device is that the “pulses” are delivered very constantly. In audio, short-term stability is important. After all: we listen to music where it is all about transients and impulse behavior. The long-term variation of a clock is then not important. Unlike clock in GPS satellites or the ‘atomic clock’ where drift over longer time is extremely important.


That a clock in a d/a converter plays an important role, we know by now. It is the conductor for the dac chip…. If the clock is not stable – shows jitter – we can hear that.

But what about on a USB card? That’s a particularly good question. Frankly, we were wondering the same thing. After all: USB is package-based. And we remain in the digital domain where jitter doesn’t really play a role at all…. Only when converting to analog is it an issue. (Or in the studio at analog to digital conversion). But as long as we stay in the digital domain, jitter is not a problem…. So not even with USB-Audio where the data is sent in packets to the DAC.

What we suspect is that the noise decreases the moment a clock does its job without noise. After all: a clock also generates noise, as you will see further on. The very slight deviations – jitter – don’t matter much to the data stream. And if it is “out of specification,” we do hear a tick or a drop-out. What then remains is that the noise at the output of the usb port lowers.

That would also explain, why a difference is audible, if we change power supplies on the usb cards. And on switches…

20 MHz

Jcat’s Master Clock does its work at 20 MHz. That’s a bit of a shame, because it’s a frequency we don’t see very often. Just like 25 MHz in switches. Now if it had been a 10 MHz clock, we could do a little more with it. Now it is only suitable for the USB card. And that while the Master Clock offeres two outputs.

Now Jcat in itself can’t do anything about that of course; the USB chipset will need 20 MHz. However, it might also have been possible with a 10 MHz baseclock and a 2x multiplier. Ah well… a good DIY’er might also be able to make a divider network.

Type test
Single Test
  • None
  • None
Product type
    Production country


    1. I posted it on your FB page, but will post here as well 🙂

      Thank you for the review 🙏 There is an upgraded version of the USB Card coming in June – USB XE EVO – which will accept the Master OCXO Clock via SMA connector. There are other improvements too (to be disclosed when the card becomes available). One small comment on the Master OCXO Clock review: the product can be ordered at a custom frequency (e.g 10MHz). We inform about it on our website. Again, thank you for taking the time to review our product

      Regarding the question from Gordon: the chipset that’s used on the card requires 20MHz. There is no PLL on the USB Card XE.

      Best regards,

    2. Jaap,
      The really big question here is why 20Mhz? I have designed 100’s of USB products and they pretty much all fall in the 12Mhz or 24Mhz (early ones 6Mhz). Using a PLL on 20Mhz to align with the required frequency for a USB card would not be hard. All theory aside the idea that the out going frequency of a PLL inherits the jitter of the main clock is never a reality.
      I say the same thing to all these companies using 10Mhz clocks in audio. Really??? There are probably some real nice oscillators out there at 24Mhz that could be divided down to 12Mhz if required that would do a real nice job if used with a low noise power supply and good lay out techniques.