Phase Perfect dpc-a10 - Page 2
Close
Login to Your Account
Page 2 of 6 FirstFirst 1234 ... LastLast
Results 21 to 40 of 110
  1. #21
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Looks like my goodies package from Digikey will be arriving today (a day early) so I will have one more go at repairing the unit. If that doesn't work (and it most likely won't) I am just going to work on drawing a schematic until I get the other unit from thermite. At that point I can start probing the two units next to each other with the oscilloscope and try to see the differences. I also have a friend with an old 30hp unit that is putting out a nasty peaky wave and sending it back out through the input but I am guessing that is most likely from bad caps. It still works unless he runs it through a boost transformer to run 440 volt machines, in that case it starts to buzz, hum and shake badly. One project at a time.

  2. Likes RustySparks liked this post
  3. #22
    Join Date
    Jan 2003
    Location
    Canada
    Posts
    11,202
    Post Thanks / Like
    Likes (Given)
    5278
    Likes (Received)
    3355

    Default

    Spikes and too much feedback are probably the filter caps, which are supposed to get changed every few years and checked regularly.
    Which reminds me its time I check mine again.
    Last edited by SND; 08-23-2019 at 11:26 AM.

  4. #23
    Join Date
    Sep 2011
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    26,787
    Post Thanks / Like
    Likes (Given)
    7561
    Likes (Received)
    8377

    Default

    Quote Originally Posted by SND View Post
    Spikes and took much feedback are probably the filter caps, which are supposed to get changed every few years and checked regularly.
    Which reminds me its time I check mine again.
    Manual on newer ones call for a 3-year change interval - which is only about 40% the lifespan of typical VFD. Clearly, Phase Technologies are not "joking"..

    Now... "working with" our newly-volunteered solder Meister...

    "Back in the day", we usta routinely bypass a fat electrolytic with a smaller solid Tantalum, then bypass that in turn with a smaller-yet Silver-Mica. Or synthetic equivalent.

    Paralleled band-pass rig, trying to keep the sharpest rise-times finding easier paths through the most appropriate and most enduring capacitor of the array.

    All-hands lasted a great deal longer that way, and it was not terribly costly.

    Anybody have "numbers" for this? Does it make sense, here?

  5. #24
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Quote Originally Posted by thermite View Post
    Manual on newer ones call for a 3-year change interval - which is only about 40% the lifespan of typical VFD. Clearly, Phase Technologies are not "joking"..

    Now... "working with" our newly-volunteered solder Meister...

    "Back in the day", we usta routinely bypass a fat electrolytic with a smaller solid Tantalum, then bypass that in turn with a smaller-yet Silver-Mica. Or synthetic equivalent.

    Paralleled band-pass rig, trying to keep the sharpest rise-times finding easier paths through the most appropriate and most enduring capacitor of the array.

    All-hands lasted a great deal longer that way, and it was not terribly costly.

    Anybody have "numbers" for this? Does it make sense, here?

    Sounds like it might be worth looking into. IIRC the smaller the capacitance value the higher frequency it can smooth out. What I do know is electrolytics are a necessary liability and any time you can take some load off them the better you will be in the long run.

  6. #25
    Join Date
    Sep 2011
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    26,787
    Post Thanks / Like
    Likes (Given)
    7561
    Likes (Received)
    8377

    Default

    Quote Originally Posted by Hardplates View Post
    Sounds like it might be worth looking into. IIRC the smaller the capacitance value the higher frequency it can smooth out. What I do know is electrolytics are a necessary liability and any time you can take some load off them the better you will be in the long run.
    VFD, P-P, or the relatively fewer DC Drives that use caps at all it isn't the clean sine waves that do-in the caps. Look how long motor-run caps last on 1-P MOTORS, utility-mains sine-wave AC

    The high "freguency" here is just the fast-RISE (or fall..) time off the pass-element switching. Caps are THERE to integrate and smooth the greater chunk of energy "under the curve" as well as the vestigial switching artifacts, but the small tail can unduly affect the life of the larger dog..

    Sharing the challenge out could perhaps push "3 years" out to mebbe 6, 9, or 12 years?

    Phase-Technolgies is paying any attention to we chik'ns and our commercial needs, cost of DOWNTIME? An "extended life" option would already have sales poop touting it and justification for its premium costing, whether at point of initial purchase, or as a premium cap replacement kit that is also an upgrade/life extender.

    Not to forget, "affordable" caps also hit the next barrier - degradation with age alone off their chemistry, even left on the shelf, so there is a break-even point in there as well.

    Anything less than "orbital" - or even a Voyager deep-space craft, new caps will at some point become less-costly than protection that costs far more.

    I just don't accept that "3 years" is that point! Seven to nine or twelve years of a VFD seems reasonable, even though they are being worked differently - VFD on input, P-P on output.

    Budgeting, predictability, and downtime thing, IOW.

  7. #26
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Georgia
    Posts
    134
    Post Thanks / Like
    Likes (Given)
    60
    Likes (Received)
    56

    Default

    Quote Originally Posted by thermite View Post
    "Back in the day", we usta routinely bypass a fat electrolytic with a smaller solid Tantalum, then bypass that in turn with a smaller-yet Silver-Mica. Or synthetic equivalent.
    I like to use tantalum where possible, and this bypass idea sounds good to take some of the fast rise times away from aluminum electrolytics, but tantalum is expensive and may not even be made in the voltages required. I suppose these AC line input filtering capacitors you are supposed to replace every three years, must be similar to AC capacitors used in motor start circuits? So there is another problem for tantalum, you would need to put two in series in such a way as to make your own non polar cap to put across AC line, then resistors across them to help balance things.
    Otherwise sounds like a great idea for lower voltage DC circuits.

  8. Likes JST liked this post
  9. #27
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Quote Originally Posted by RustySparks View Post
    I like to use tantalum where possible, and this bypass idea sounds good to take some of the fast rise times away from aluminum electrolytics, but tantalum is expensive and may not even be made in the voltages required. I suppose these AC line input filtering capacitors you are supposed to replace every three years, must be similar to AC capacitors used in motor start circuits? So there is another problem for tantalum, you would need to put two in series in such a way as to make your own non polar cap to put across AC line, then resistors across them to help balance things.
    Otherwise sounds like a great idea for lower voltage DC circuits.

    Looks like 75 volt is the highest economical tantalum and 125 volt get very pricey. Perhaps a stack-up in series and parallel would work. Mica on the other hand seem to be cheap and can be had up to 10kv...…….

    The AC caps are just standard non-polar run capacitors.


    We are dealing with a machine worth several thousand dollars, would a couple hundred not be a wise investment as preventative maintenance being it seems like when the filter caps go it can cascade into more difficult to diagnose and repair components?

  10. #28
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    17,877
    Post Thanks / Like
    Likes (Given)
    1813
    Likes (Received)
    2987

    Default

    The replacement interval on the caps is no doubt because they are being "worked hard".... in other words they are being hit with a lot of "ripple current",

    "Ripple current" is just the current which the AC component of the voltage across the capacitor drives through it. It causes heating of the part, and heat is a prime factor that kills electronic parts. Yes, the high frequency content, such as repetitive spikes, etc, is the part that boosts the riple current, because a capacitor passes higher frequencies better. *

    The PP folks reckon that in 3 years of continuous use, the probability of failure has gone up enough that you need to replace the parts. The real average life may be 4 years, or 5, but they very reasonably suggest a shorter time so that very few parts will fail before their suggested replacement time.

    PP could have reduced the problem by using several smaller parts (same value) in parallel, but apparently made the decision to do what they did, possibly on grounds of cost, space needed, etc. I do not know their reasons. Several smaller parts would have overall had a considerably larger current capacity, due in the most part to better cooling (more surface area total), and also because the parallel paths would likely have reduced the overall impedance, plus each part would have less inductance and resistance to start with.

    It would be a wise move to just replace those generally on the schedule suggested, realizing that the hours of working per day will affect the time. It's like an engine overhaul, you don;t need one as soon if the engine is operated much less of the time, vs one that goes 24/7. But the consequences of capacitor failure may be damaging to the uit or to other devices, so it is better to be safe than sorry with the replacement of relatively cheap parts of an expensive device.

    The net value of capacitance is likely to be critical, especially for any parts used in a "sine filter" to reduce EMI. So use of a different value part would need enough of them to make up the same overall value. It is likely that only the bux capacitors are critical for ripple current, the filter capacitors in EMI usage are much less likely to see high currents, due to the other filter components. That is very dependent on the filter design, however.

    * Motor run capacitors are not usually very good at passing high frequencies, since they are not designed for that. But PP must have selected a type that is better for that purpose. Capacitors are not "purely" capacitive, they also have an inductive and a resistive "componnent" as well. The resistor part develops heat, and the inductive part reduces the efficiency at high frequencies.

    It is possible to reduce the inductive and resistive portions of the capacitor by design, the issue is that many motor run parts are not made to a design that reduces them. This is because a motor does not need that, and it can cost more.

    BTW: The old technique of bypassing the large capacitor with another one for high frequencies has been largely discarded. Problem is that the smaller bypass capacitor may resonate woth the inductive portion of the larger part, and cause a worse problem than existed to begin with.

  11. Likes RustySparks liked this post
  12. #29
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    YouTube

    It appears as though I've reached a point where percussive maintenance makes a difference......not that that's a good thing. I am having a tough time tracing it down...….maybe a cracked ceramic capacitor from the hot air rework station.....?

  13. #30
    Join Date
    Sep 2011
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    26,787
    Post Thanks / Like
    Likes (Given)
    7561
    Likes (Received)
    8377

    Default

    Quote Originally Posted by JST View Post
    BTW: The old technique of bypassing the large capacitor with another one for high frequencies has been largely discarded. Problem is that the smaller bypass capacitor may resonate woth the inductive portion of the larger part, and cause a worse problem than existed to begin with.
    I suspect more than a few of us knew that. Which is why I asked if anyone "had numbers" for it, given MY use was on lower voltage goods by a LOT lower. Also nominally DC to begin with.

    That said a P-P is a fixed-frequency device, so it should be computable. Also easly vetted and adjusted, if need be, predictably stable, thereafter.


    Page Two:

    Indeed, an array of smaller (physically) caps might make good sense - even better sense that add-on bypass.

    P-P use-up MOST of the space in either blue or white enclosures, but I'm not opposed to hanging a backpack, upper or lower "story" onto that apartment house if there's gain to be had off the extra hassle.

    So long as the leads themselves do not become resonators nor transmission lines, electrons don't much care if the blankets they share are in the same bedroom, let-alone same bed.

  14. #31
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Georgia
    Posts
    134
    Post Thanks / Like
    Likes (Given)
    60
    Likes (Received)
    56

    Default

    Quote Originally Posted by Hardplates View Post
    YouTube

    It appears as though I've reached a point where percussive maintenance makes a difference......not that that's a good thing. I am having a tough time tracing it down...….maybe a cracked ceramic capacitor from the hot air rework station.....?
    If you haven't already, be sure to take a close look at the legs of the surface mount chip (DSP ?) near where the percussion occurs in your you tube video. Be sure all legs are soldered securely. Also suspect whatever it uses for clock crystal.

  15. #32
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Quote Originally Posted by RustySparks View Post
    If you haven't already, be sure to take a close look at the legs of the surface mount chip (DSP ?) near where the percussion occurs in your you tube video. Be sure all legs are soldered securely. Also suspect whatever it uses for clock crystal.
    Funny you should mention the clock, when I touch the output of the 15mhz clock with a floating scope it trips the unit out. I was going to try and take a differential reading (to get rid of noise) with 2 ungrounded channels but it trips out even when you touch an ungrounded probe to the clock output. I know very little about clocks and DSPs but that doesn't seem right...….?????

  16. #33
    Join Date
    Sep 2011
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    26,787
    Post Thanks / Like
    Likes (Given)
    7561
    Likes (Received)
    8377

    Default

    Quote Originally Posted by Hardplates View Post
    Funny you should mention the clock, when I touch the output of the 15mhz clock with a floating scope it trips the unit out. I was going to try and take a differential reading (to get rid of noise) with 2 ungrounded channels but it trips out even when you touch an ungrounded probe to the clock output. I know very little about clocks and DSPs but that doesn't seem right...….?????
    "Ground referred" is squirelly with all manner of parts, these units. Best to figure it is in float w/r OUR Earth, has references to its own independent planet but not-only. See also faulty bonding and grounding, but make NO ass-u-me umptions. "With care", IOW.

    Capable of surprising a person, IOW. Nasty-capable. Very.

    Select test gear and use it accordingly.

    As 350 VDC plus-plus serious motor/regen and switching spikes are part of my playground, I have US-made HV probes on all four channels of my newer toy, a Rigol digital. Those probes cost more than the 'scope did, but I don't have to rob the old dual-channel Fluke, the 'scope yet lives... and so do I!

    There are bold electronics techs. There are old electronics techs. There are damned few old, bold electronics techs. This is not an effing Pee Cee.

    Vintage of 1945, and still counting.

    Go ye forth and be paranoid, likewise!

    Consumately lazy, Iyam, in me dotage, but even so, I don't want to put-off a delivery run because ye've rectumfried yerself right off the dance-card.

    Get tired, get frustrated, or get luck-happy and cocky, any of.. down probes, walk away, take a break, run your brain a while. Come back refreshed, and cold-blooded CAREFUL.

    These units were never taught how to merely take prisoners.


  17. #34
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Georgia
    Posts
    134
    Post Thanks / Like
    Likes (Given)
    60
    Likes (Received)
    56

    Default

    Quote Originally Posted by Hardplates View Post
    Funny you should mention the clock, when I touch the output of the 15mhz clock with a floating scope it trips the unit out. I was going to try and take a differential reading (to get rid of noise) with 2 ungrounded channels but it trips out even when you touch an ungrounded probe to the clock output. I know very little about clocks and DSPs but that doesn't seem right...….?????
    Don't worry about it, I don't think this means a problem in this unit. You were looking for a mechanical fault that caused the DSP to branch into some other instruction set when you whacked the board. Does it trip out when you touch the same point with something like a fiberglass rod? A non-conductive tapping tool would be best here.

  18. #35
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    When probing high voltage boards like this I use a battery powered 100MHZ siglent handheld scope. The only thing not floating on this scope is the two channel grounds in relation to each other.....they do make an orange version that has isolated grounds but I just don't hook the grounds to different potentials and myself and the scope live to fight another day X10 probes also

  19. Likes RustySparks liked this post
  20. #36
    Join Date
    Sep 2011
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    26,787
    Post Thanks / Like
    Likes (Given)
    7561
    Likes (Received)
    8377

    Default

    Quote Originally Posted by RustySparks View Post
    A non-conductive tapping tool would be best here.
    May have a teflon or PET one for him. Maybe even a vintage "Orange stick". Mah Bell thing, so New Jersey, not the fruit!

    Haven't worked up the energy to climb the attic stairs yet.

  21. Likes RustySparks liked this post
  22. #37
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Quote Originally Posted by RustySparks View Post
    Don't worry about it, I don't think this means a problem in this unit. You were looking for a mechanical fault that caused the DSP to branch into some other instruction set when you whacked the board. Does it trip out when you touch the same point with something like a fiberglass rod? A non-conductive tapping tool would be best here.
    I have been tapping around with a plastic rod and can't seem to locate the fault. Often if I twist the board it does the same thing but you only have a short window when the lights are flashing quickly. Once they slow down no amount of tapping will help. I've been working around high voltage for some years, long enough to respect it and know what it can do but not be afraid of it, just like a gun or any other tool the operator is the one in control.

  23. #38
    Join Date
    Sep 2011
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    26,787
    Post Thanks / Like
    Likes (Given)
    7561
    Likes (Received)
    8377

    Default

    Quote Originally Posted by Hardplates View Post
    I have been tapping around with a plastic rod and can't seem to locate the fault. Often if I twist the board it does the same thing but you only have a short window when the lights are flashing quickly. Once they slow down no amount of tapping will help. I've been working around high voltage for some years, long enough to respect it and know what it can do but not be afraid of it, just like a gun or any other tool the operator is the one in control.
    Twist the board sez cracked trace, burnt trace, or faulted plated-through hole to inner layer on a multi-layer board, 7 layers the worst I EVER saw. Three is what I'd expect, here - inner one perhaps common ground and planar. Four layers only maybe. Strong - really strong - light should tell the tale.

    Transient detection one can do, de-powered board, of course, with a digital R/S flip-flop. One DIP package, simplest of NAND gates, 7440? Small battery. LED, reset pushbutton, and good to go. Intermittent open or short, either one. Thermal glue it on it's back like a dead cucaracha, no need of a socket, just wire-wrap or solder to the splayed out (for clearance) legs.

    Old Day Job, Chief Eng wanted me to patent one I had built to test the wires embedded into eyeglass frames, bi-lateral & CROSS hearing aids. Folding the temples stressed the wire, eventually failed it intermit - wearer got painful cracks and pops, the gain and power we had to run. Meter wasn't good enough to catch the fast action, R/S latch was VERY fast.

    He had over 2 dozen patents, but I thot he was nuts. Too easy for there to NOT be prior-art!


  24. #39
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Quote Originally Posted by thermite View Post
    Twist the board sez cracked trace, burnt trace, or faulted plated-through hole to inner layer on a multi-layer board, 7 layers the worst I EVER saw. Three is what I'd expect, here - inner one perhaps common ground and planar. Four layers only maybe. Strong - really strong - light should tell the tale.

    Intermittancy detection one can do, de-powered board, of course, with a digital R/S flip-flop. One DIP package, simplest of NAND gates, 7440? Small battery. LED, reset pushbutton, and good to go.

    Old Day Job, Chief Eng wanted me to patent one I had built to test the wires embedded into eyeglass frames, bi-lateral & CROSS hearing aids. Folding the temples stressed the wire, eventually failed it intermit - wearer got painful cracks and pops, the gain and power we had to run. Meter wasn't good enough.

    He had over 2 dozen patents, but I thot he was nuts. Too easy for there to NOT be prior-art!

    I was hoping a cracked solder joint but couldn't find one, you are probably right, I hope you are right and it's not a problem inside the dsp. Although not common I have ran into transistors and ICs that responded to percussive maintenance.

    Had a intermittent display issue with a Crest ultrasonic once and it turned out to be a loose socket on the main IC. A zippy tie around the outside of the socket was all that was needed to make sufficient contact. Point is far more often than not I find issues to be something very simple in nature, but every so often a "proper" problem pops up.

  25. #40
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    312
    Post Thanks / Like
    Likes (Given)
    78
    Likes (Received)
    104

    Default

    Quote Originally Posted by RustySparks View Post
    Don't worry about it, I don't think this means a problem in this unit. You were looking for a mechanical fault that caused the DSP to branch into some other instruction set when you whacked the board. Does it trip out when you touch the same point with something like a fiberglass rod? A non-conductive tapping tool would be best here.
    No it does not trip out when touched with an insulator...so I guess it is that sensitive that touching it with even a small conductor changes something like the capacitance or inductance and causes it to fault out? The two pads for the oscillator go to two pins on the DSP. Now to try and find a mechanical fault such as a trace...….


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •