Simple fix for Bus Pirate power supply oscillations

in Bus Pirate by DP | 17 comments

8503210456_806d594d89_o

Jeff of Mightyohm wrote an article about power supply oscillations on the Bus Pirate:

After removing my device under test, I quickly narrowed the problem down to noise on the Bus Pirate’s 3.3V rail. I initially suspected that this might be some kind of noise coming from the 5V USB supply, so I added an additional 10uF cap in parallel with the 1uF already on the board (C1). The sawtooth increased drastically in amplitude and went down sharply in frequency. That’s when I realized that this wasn’t noise, it was a power supply oscillation.

This entry was posted in Bus Pirate and tagged , .

Comments

  1. Markus Gritsch says:

    Yup, I can confirm the oscillation. My freshly built PB v3.8 has a similar sawtooth on the 3.3 V supply with a frequency of about 24 kHz and 170 mVpp. A higher frequency noise of about 130 kHz and 700 mVpp is superimposed.

  2. JBeale says:

    Is this new? That article is more than a year old. The issue was discussed here at that time, see for exampe: http://dangerousprototypes.com/2013/03/06/possible-issue-with-voltage-regulation-on-some-bus-pirates/

  3. JBeale says:

    More than a year old… search the forum here for “Possible issue with voltage regulation on some Bus Pirates” posted on Feb 28, 2013.

  4. oliver says:

    a link would have been helpfyl. was searching the forums nut could only find 1 post where the guy replaced the entire regulator.

  5. Drone says:

    Yeah, this problem has been around forever – and DP is still allowing Seeed to pump out BP boards with this problem still? If so, it should go well together with the outdated firmware Seeed is probably shipping.

  6. Niklas says:

    The possible issue with too low ESR of the output capacitance has been mentioned before but I think that Ian also should have a closer look at the BOM specs. Y5V is not a very good dielectrica and the datasheet for the MIC5205 specifies a minimum of 1 uF at the output for stable operation

    Y5V means -20% to +80% of capacitance change over the operating temperature range and with 0 Volt biasing, ie worst case range is 0.8-1.8 uF. The datasheet from Kemet does not mention any manufacturing tolerance, otherwise that would have been added to for the calculations. To that you will also have to add the biasing effect on the effective capacitance. For the 10 V rated C11-C13 with Y5V this is quite significant even at 3.3V. By significant I mean more than 50%. The result is an effective output capacitor with much less capacitance than expected and a possible root cause for the oscillation.

    If you are going too stick with Y5V and 10 V rating, uprate the capacitance to 10 uF. A better alternative would be to change to >= 2.2uF, >= 16V and X5R or X7R instead.

    A note about ceramic and tantalum capacitors and their rating:
    Murata has some good info in their datasheet catalog for MLCCs about voltage derating with respect to operating temperature. For temperatures above 70′C this must be considered. I have seen X5R caps going to short circuit due to this when operated close to the rated voltage.
    Kemet has info about recommended voltage derating on tantalums. 50% of rated voltage up to approx 70-80′C followed by a linear derating above that. At 110′C a 35 V rated tantalum is not recommended to have more than approx 12 V applied. Failure modes include short circuit with risk of fire with open flame.

    • Ian says:

      In the original v2go and v3 I specced a mic part at recomentation of Nate at sparkfun. They used ceremic caps but I noticed the datasheet did spec tantalum or electrolytic and used it on the original design. When I redid the v3.5 we swapped to a new vreg that works with the cheaper and more socially kind ceramic caps. In most boms I spec sot23-5 3.3volt instead of micxxxx because there’s no real standard name for these little vregs like 7805, etc. You have to see what is available and check the datasheet.

  7. jaromirs says:

    That is how it ends when you don’t read datasheet during designing a product and assume “throwing ceramic cap anywhere on power line doesn’t hurt”.

  8. Kaz says:

    I kind of have to agree with the other comments. It’d be nice to hear if this is going to be addressed in some future revision. As it stands, it comes off as “Someone found a problem… BUY IT ANYWAY!”
    To be honest, it hasn’t impacted any of my projects (even when I’ve used the 3.3V supply), but that could just be due to being very tolerant to voltage instability on the input. I still think the Bus Pirate is an absolutely amazing product, but will this be improved in new ones shipping from Seeed someday?

  9. Ian says:

    This is a really old dupe, not sure why it’s here.

    Jeff’s article applies to the Spark Fun Bus Pirate, though it was found in Seeed’s Bus Pirate too.

    It was only a small section of Bus Pirate’s that have this problem. When we sourced the parts for the Bus Pirate years ago the MIC VREGs were expensive/not available in china and we used an alternative. That alternative supported ceramic caps according to the datasheet, so we switched from tantalum because it’s a conflict mineral. At some point new sourcers/suppliers at Seeed reverted from the alternative to MIC5205 probably because they got big and were able to do a deal. We swapped that out after the problem was found last year but I’ll go check again.

    Sorry it’s taken so long to reply, the great firewall has throttled DP down to a drip and VPNs are no help at all. I have to snatch a random moment when the firewall is feeling generous or run to Hong Kong to reply. I think they’re unhappy about “Hacker Camp SZX” even through its a celebration of engineering here :(

    • Kaz says:

      Thanks, Ian. I (and probably others) didn’t notice it was a year-old post on Mightyohm. Glad to know the problem is (most likely) resolved. I’ll have to find some time and measure my 3v3 supply.

  10. Adrian says:

    My Bus Pirate V3a complains that the 3v3 supply isn’t working when I enable power. the self-test reports it as 2v2. I guess I’ve blown up the regulator – what’s the best choice for a replacement ?

    • Ian says:

      I’ve been using the Microchip regulators
      Lately. Very cheap and plentiful. Something like McP1703 maybe.

    • Tony says:

      If replacing the 3v3 regulator doesn’t work, also check if there’s something on the 3v3 rail thats gone dodgy and is pulling the voltage low. Capacitors sometimes go short-circuit when failed.

  11. Markus Gritsch says:

    It would be nice, if the Partlist here http://dangerousprototypes.com/docs/Bus_Pirate_v3.6 would be updated to contain the recommended regulator/capacitor combination. Currently it just says MIC5205-3.3YM5 regulator and 0.1uF C603 capacitor. Not mentioning either to use non-ceramic caps or a different regulator.

    People (me included) relying on the current partslist building their free PCB Bus Pirate get easily trapped, otherwise.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.