I have been working on a project that utilizes a 10MHz Oscillator and 28V input power. At the start of my project, I used a DC/DC converter which took my 28V line and moved it to a 5V line. I then used this line to power my Oscillator. After hooking this up to my scope, I saw that the noise really affects my signal after the Oscillator, its only slightly noticeable, but the part this plugs into expects a very accurate 10MHz line.. .

In my original design I had my ground of the 5V grounded up with the 28V line, but I was told that I should not do this, and instead have a small local ground to the oscillator. Additionally I was told that I should keep the wires as close to each other as possible to reduce some more noise. I redesigned, and it worked, but very minimally did it change the amount of noise.

I was told that using filters is the way to go, but everwhere I look I can't really see what exactly I'm looking for or what I should be doing exactly.

Here is my schematic at the moment:

enter image description here

With the imoportant aspects of this project:

DC/DC Converter: XP Power JCM2024S05

  • Variable Input, Output 5V, 4A Max draw output

Oscillator : Abracon LLC AOCTQ5-X-10.000MHZ-I3-SW

  • 5V input, 10MHz Out, Pulls 5W

Could anyone point me towards a nice source on the subject or some nice tips/pointers/help on my specific problem? Anything would be so much appreciated.


The simplest thing to do to start with is to add some smoothing caps on the 5V line - a 10uF ceramic and something like a 47uF tantalum would help to keep some of the lower frequency ripple under control.

If you can adjust the voltage of your DC-DC, one option would be to increase the voltage to say 5.3V, and then use a low noise 5V LDO (e.g. TPS7A85) which will help to clean up any switching noise. Setting the DC-DC voltage to just greater than the drop-out voltage of the LDO will help to keep losses and heat generation to a minimum.

Additionally you can look at splitting your supply regions up using feed-through capacitors. These act like L-C filters and have great noise rejection. By placing one between the main power supply rail and sensitive components, you can isolate some of the noise in the system.

Finally for sensitive components, you can add more than one ceramic cap close to the power pins. If you use a series of values in parallel, such as 0.1uF + 10nF + 1nF, with the smallest value closest to the power pins, you will increase the effectiveness of the filtering over a larger range of frequencies.

As an example, we are working on a 12-channel 370MSPS ADC board with ultra low jitter clocks. As it stands that board has 12 amplifiers, 6 ADC ICs and a clock generator. Each one has local power planes, separated from the main power planes using feed-through capacitors to isolate noise. We have ended up with 43 power regions, supplied by 12 LDOs which are in turn supplied by 7 DC-DC regulators. On that board there are over 600 capacitors alone!

The one thing we have done on this board is to not split up the ground planes. There is a nice application note (I'll see if I can find the link) which explains how actually splitting up the ground plane can cause more issues than it solves from a noise perspective. The key thing is to keep an eye on the placement of components - for example keeping noisy things like switching regulators away from sensitive analogue circuitry. By keeping an eye on when the current return paths will be, you can keep noise contained even on one single ground plane.

  • \$\begingroup\$ Nice! 12 channel 370 MS/s - am I correct to think of MIMO digital comm systems? \$\endgroup\$ – Marcus Müller Dec 22 '16 at 22:23
  • \$\begingroup\$ @MarcusMüller In fact it will be an analogue front end for a multiplexed ultrasound system (background reading in case you are interested). But in essence you could quite easily use a similar design of board for digital comms systems. \$\endgroup\$ – Tom Carpenter Dec 22 '16 at 22:28
  • \$\begingroup\$ Not at a PC where I have full IEEE access, but from the abstract: niiiice. But I guess that basically means that you get to recover the 12 signals that were time-multiplexed at the catheter head, if I understand correctly, limiting each to around 30 MHz of bandwidth (assuming we're talking about signals that are converted to some complex baseband and IQ sampled, not directly real-value sampled) \$\endgroup\$ – Marcus Müller Dec 22 '16 at 22:34
  • \$\begingroup\$ But that still doesn't change the overall bandwidth of the muxed signal. How does clock distribution to the sensors work? Is there timing recovery? Will definitely have to read that paper \$\endgroup\$ – Marcus Müller Dec 22 '16 at 22:36
  • \$\begingroup\$ @MarcusMüller it's an open-access paper, so you should be able to download it without an account. The multiplexing scheme is quite basic (direct sample as you say) due to size and complexity constraints, but seems to get the job done. \$\endgroup\$ – Tom Carpenter Dec 22 '16 at 22:36
  1. get rid of the dc dc converter. This module generates radiated noise and noise on the output. The module excepts 15V input voltage. If you are not able to change the 28V, use a 15V linear regulator with cooling. During warmup of the module the regulator needs to be able to get rid of a little less than 5W.

  2. Add common mode and differential mode filtering to the power input connector.

  3. Use shielded cables and a shielded enclosure and connect the shield of the cable direcly to the chassis at the point the cable enters the enclosure. Let all cables enter the chassis near each other and connect it directly to the pcb with short leads. Connect the pcb ground plane with the chassis at the point where the cables enter the chassis.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.