Home Forums General DIY Pedal Discussion Clipping Diodes and Curve Tracing

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #14959
    Cybercow
    Moderator

    To prove my point to some of the newcomers to pedal building in a couple of the FB pedal building groups, I gathered, photographed and curve-traced several different diodes (excluding LEDs) that are typically used in clipping circuits. The results are self-evident in the image below. I wanted to focus on the dvl-black-banded NOS Ge #2 diode as they seem to be quite special. First, they sound smoother than any other diode or clipping device I’ve ever experimented with. And the curve trace on it demonstrates that while it posses an unusually high Fvd (about 0.829v), the curve-trace exhibits a very long (slow) rise to that value, as the current rises. The resultant “sound” of those as clipping diodes (to my ears anyway) is very smooth. (In the chart, the curve trace for the dbl-black-banded NOS Ge diode is marked “A-K-3” and is the longest light-green line in the chart. In the legend, it noted simply as “NOS Ge #2”.)

    Anyway, here is the screen shot of the curve trace results – and the photo following that is one of the dbl-black-banded NOS Ge diodes that SmallBear occasionally has for sale. I only have two of them and would like to find more of them.

    DIODE CURVE-TRACING CHART:

    Here is the photo of the dbl-black-banded NOS Ge diode that SmallBear intermittently stocks.

     

    #14963
    wilkie1
    Moderator

    Mommy, Mommy!  Buy me some of those!

    #14974
    Barry
    Keymaster

    Very informative.

    I am afraid I ate all of mine.

    #14983
    wilkie1
    Moderator

    I’m keeping that plot to use when selecting diodes.  This is great, Mark!

    #14993
    Cybercow
    Moderator

    Thanks fellas! Feel free to copy and share.

    I found the consistency, even with the Ge diodes to be quite solid. Before generating, saving and posting the chart above, I first tested 10 of each of the same part number – except the blue-band NOS Ge #1 diodes I got from Barry. I only had two of those. And even those two were identical to one another. All those for which I had ten or more of showed no deviances in the curve-traces. So it seems that Ge diodes tend to be quite consistent in their curve-traces also.

    I tried doing some curve-tracing on the P-N junctions for some NOS Ge BJTs; however, due to leakage, gain & temperature sensitivity factors, the results were all over the map; varying as much as 20% for the same part numbers. Even those within a known same batch. So I’ll NOT be posting those results.

    In the original post, I was somewhat focused on the MA856 diodes because of how highly they are cherished in the pedal building community for KOT circuits. However, as close as they are to 1N914 & 1N400x series diodes in the curve-trace & Fvd, I was somewhat disappointed regarding the ‘special attention’ and regard that specific part has been given. Genuine MA856 diodes are indeed hard to find, but I was able locate some after a long search from a reputable vendor.

    After doing more research in the various pedal building forums, MA856 diodes are a cork-sniffing, wild goos-chase. Turns out the original manufacturer of the KOT pedal simply had tons of MA856 diodes at their avail. The curve-traces prove how similar the MA856, 1N914 & 1N400x series are in their response. The comments more recently found demonstrated that others had recorded side-by-side KOT circuits identical in every way except for the diodes. And the OPs and responses all agreed there were zero admitted discernible aural differences.

    We live, we hopefully learn, and then we get build the next pedal.

    #14996
    Barry
    Keymaster

    OK I am confused.

    I have blue band ones.

    I am hording them till I can find more. Are these the ones and which do these represent on your chart?

    #14998
    Cybercow
    Moderator

    Barry – those are the “NOS Ge” – the red trace on the chart. I made an error initially in my OP and have since corrected it. It’s those dbl-black-band clear glass “NOS Ge #2” that created the long graceful (lt green) curve up to its 1.4v Fvd at 12mA. Sorry for the confusion. I thought I had gotten them from you, but dug around my records and purchase orders and discovered that I had gotten them from SmallBear. SmallBear never provided an actual part number for them. They’re just labeled as “Germanium Diode Hi-V Fwd”, describing it as “A generic part in a clear, DO-7 glass package.”

    The blue-banded ones you have for sale from time-to-time are still quite nice with medium Fvd and nice soft knee to its curve.

    #15168
    brd
    Participant

    They look a lot like these. I purchased these from Small Bear years ago and were sold as “point three +”. I believe I got 3 matched sets out of a lot of 10. I don’t have notes on which builds they ended up in except that the last set measured around .35 and were recently used in an MKC build that sounded sweet. I have also been looking for replacements, and just purchased some Amperex brand from Small Bear that were advertised as high V fwd. I haven’t had time to test them yet and I doubt I have the resources to measure and produce a diode curve tracing chart for them. I have an Atlas DCA55 and a Fluke meter, if there are any measurements you would be interested in on any of the diodes, just let me know.

    #15213
    Cybercow
    Moderator

    brd – visual comparison of this particular diode can be confusing. My best guess is that as production went on, the die some changed and the specs began to drift until they eventually replaced the die. And TBH, I’m not sure of the manufacture process of those diodes (or any others for that matter). But the poor is that while they “appear” identical, they may well not be. At least according to the curve-trace testing I’ve done. I found a couple more of those same “looking” diodes buried away in another box I had and found they were not the same. The ones I most recently found measure in at ~0.36v (@5mA) as opposed to the 0.87v (@5mA) diodes I charted in the OP.

    When it comes to Ge diodes, I no longer depend on the specs provided by the vendor and test & record the Fvd values of them before committing them to solder.

    Unfortunately, the Peak Atlas DCA55 doesn’t come with the PC software for curve-tracing.

Viewing 9 posts - 1 through 9 (of 9 total)
  • You must be logged in to reply to this topic.