MX60 leaking oil

Patman3
Patman3 Solar Expert Posts: 62 ✭✭✭✭
Hi All, Great Forum! I keep a paper logbook under my MX-60 to record the data since it only holds 64 days. Anyhow, the paper had a small maybe 1 inch circle oil spill on it and I felt the bottom left side of the MX60 and it was slightly oily. This unit has been in operation for several years now. I hoping its from the fan? and not from capacitors?? Any thoughts/ideas greatly appreciated - THANKS...

Comments

  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil

    First thing that I thought of was capacitors. What you might do is slide the MX out of the chassis and take a look down the top end of it to see how the end cap looks at least. Press on the top of it if you can get your finger in there and see if it is mushy or solid on the capacitors' top. I just changed a fan in an MX60 that had been in operation for about 5 years and the cap tops were kind of bulging. No leaks though. There really isn't much in the fans so I kind of expect capacitors...

    Shouldn't be too big of a deal to fix though if it's just the caps that need replacing.
    The MX60 actually is specially designed for longer capacitor life. The caps would not last nearly as long without that design feature. Some other charge controllers, like the ones from up north, do not have that design. It wasn't evidently cost effective for the short term.

    BTW, what is the nominal input and output voltage of your MX60 system ?

    boB
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: MX60 leaking oil

    Yikes ! Several years ago, there was a counterfeit cap company, that hosed a whole years production for several computer motherboard factory's. Now mainstream RE products are having cap failures? This is not good.

    Caps have 2 enemies, heat and ripple, both of which should be easily controllable. Purchasing the proper cap, (avaib up to 2,000 hrs @ 125C) would be key. But say running a 125C cap at 85C, that de-rate only gets you to 20,000 hours which is less than 3 years.
    Caps do have a design life, they wear out, and can go for much longer, if they are not working hard. (low ripple) In a high power application, with any appreciable heat, it's going to be tough to get a long lifetime of > 5 years.

    So what does a MPPT controller need caps for? I'd guess the output of the buck stage inductor. Needs enough to knock down the switching noise to something that the FCC won't come looking for. The big battery bank will sure soak up a lot of ripple, and not have trouble. (wiring ham radios direct to battery, not fuse block in car). So choices in the inductor frequency would be needed to made, to work at a frequency where caps are effective, and not see too much ripple. As freq goes down, caps get more effective, but then the ripple goes up, so choseing a balance would be needed to reduce warranty claims and product repairs. User replaceable caps would be a benefit, and if the PCB could be laid out so it's intuitive and feasible, liability should be low.

    Anyone else haveing capacitor troubles with large MPPT controllers? A long turnaround cycle would be very bad. (even a short turnaround would be bad)
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil
    mike90045 wrote: »
    Yikes ! Several years ago, there was a counterfeit cap company, that hosed a whole years production for several computer motherboard factory's. Now mainstream RE products are having cap failures? This is not good.

    No, this is not the same thing. That was a pretty weird problem though,huh.

    The caps (which are necessary to store energy), just get hot because of environment and ripple current. Some charge controllers are designed to minimize this (like the MX60, FMxx, and Classic)...

    Unfortunately, nothing lasts forever... 125 degree C or 10,000 hour, either way, the caps are bound to fail some time. Just replace them when necessary. It's maintenance just like maintaining your car by replacing the brake pads every so often and changing the oil and battery and bunches of other stuff. You get so many miles out of it.

    Hey, look at Enphase... They're mounted on the back of a solar module and have electrolytic caps. Enphase says their inverter has a MTBF of something like 300 years !! Right ! ;)

    boB
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: MX60 leaking oil

    now that's interesting how they warrant for longer than even controllers and other inverters are warranted for. i seem to recall initially seeing a 20 year warranty on it, but it's now down to a 15yr warranty. i guess by year 6 they'll drop it to 5 years and you're screwed. i even had previously questioned their ability to warrant something that long under adverse conditions. i doubt the parts are that good even if quality parts are used, and for a 331yrs mtbf?
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil

    It WAS 10 years and went to 15 years.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: MX60 leaking oil

    ok, but that's still a long time for any electronics let alone some out in the elements. i have my doubts that the parts will last 15yrs under such conditions.
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil

    Paytman...

    How long has that MX60 been in service ? Years ? Or, maybe, how many kilo-Watt-Hours have been accumulated on this unit ?

    I am trying to get an idea on how many hours we might be able to expect to get on a charge controller of a particular design, etc... Also, what is the serial number of
    that MX60 ??

    Thanks,
    boB
  • Brock
    Brock Solar Expert Posts: 639 ✭✭✭✭
    Re: MX60 leaking oil

    On our MX-60 with a 48v bank when we had our array wired 5 panels in series (88 volts) I noticed the MX-60 fan ran a lot more. I re-wired the array to 4 panels (70 volts). The overall wattage went up slightly and the fan runs a LOT less. Since the voltages are closer is this also easier on the caps? I would think heat wise it would be better for sure.
    3kw solar PV, 4 LiFePO4 100a, xw 6048, Honda eu2000i, iota DLS-54-13, Tesla 3, Leaf, Volt, 4 ton horizontal geothermal, grid tied - Green Bay, WI
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil

    The absolue "easiest" on the filter capacitors is when the input voltage is exactly double that of the battery voltage... So, in your case, it may not be easier on the caps, BUT, at the same time, as you noticed, the fan runs less and there is less power wasted inside the controller in the FETs. That lower inside temperature also helps the life of the caps of course.

    It's all a compromise, but in your case, you ended up with more power with less PV Voltage. Lowering the PV input voltage is what one ~should~ do when your array isn't ~too~ far away from the controller.

    boB
  • lorelec
    lorelec Solar Expert Posts: 200 ✭✭
    Re: MX60 leaking oil
    mike90045 wrote: »

    Caps have 2 enemies, heat and ripple, both of which should be easily controllable. Purchasing the proper cap, (avaib up to 2,000 hrs @ 125C) would be key. But say running a 125C cap at 85C, that de-rate only gets you to 20,000 hours which is less than 3 years.
    Caps do have a design life, they wear out, and can go for much longer, if they are not working hard. (low ripple) In a high power application, with any appreciable heat, it's going to be tough to get a long lifetime of > 5 years.

    So what does a MPPT controller need caps for? I'd guess the output of the buck stage inductor. Needs enough to knock down the switching noise to something that the FCC won't come looking for. The big battery bank will sure soak up a lot of ripple, and not have trouble. (wiring ham radios direct to battery, not fuse block in car). So choices in the inductor frequency would be needed to made, to work at a frequency where caps are effective, and not see too much ripple. As freq goes down, caps get more effective, but then the ripple goes up, so choseing a balance would be needed to reduce warranty claims and product repairs. User replaceable caps would be a benefit, and if the PCB could be laid out so it's intuitive and feasible, liability should be low.

    The fortunate thing about electronics used in solar applications is that they're not operating at full tilt 24/7...so it wouldn't be unreasonable to expect 5 years of life out of some electrolytics used in a conservative design. Two well-known cap manufacturers make low ESR electrolytics designed for switching power supplies that have a 10,000 hour rated life at 105degC. Unless you're always running the capacitor at max rated temp and ripple current, they should last substantially longer than their rated life. What some designers fail to take into consideration is the internal heating that higher ESR causes. It's usually necessary to parallel several caps to achieve an ESR that's low enough, even though the ripple current may be well within spec.

    With a buck topology, the input needs a large amount of capacitance owing to its square-wave nature. The output sees only the p-p inductor current ripple, and this is usually quite small, so the output of a buck converter needs little capacitance. Switching noise usually needs to be addressed closer to the point of generation (the switching node) -- otherwise it will radiate and become very difficult to attenuate. This is often accomplished with snubbers, electrostatic shielding, and/or strategically placed bypass caps. To clarify what Bob said: the most stressful situation for the input caps from an internal heating perspective is when the duty cycle of the converter is at 50%...because this is when their RMS current is the greatest and the most heating occurs. This corresponds to an input voltage that is twice the output voltage.

    Marc
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil
    lorelec wrote: »
    To clarify what Bob said: the most stressful situation for the input caps from an internal heating perspective is when the duty cycle of the converter is at 50%...because this is when their RMS current is the greatest and the most heating occurs. This corresponds to an input voltage that is twice the output voltage.
    Marc


    Normally this 50% Duty is most stressful. In the MX60, FMxx, Classic (and I think, the Apollo controller design), 50% duty ratio is (theoretically) 0 (zero) ripple current because of the buck converter design.

    boB
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: MX60 leaking oil

    Ripple in of it self has little bearing on cap life, its the heat that the current swing of the FETS switching which causes a voltage drop due the ESR of the cap that creates heat, similar to a wire heating when a voltage is applied on a wire and the resistance will heat the wire. This heat, in combination of the ambient temperatures the components are mounted in is what determines if the parts is being used within specifications, not the ripple its being subject too.

    The XW is Superior to the MX/Flexmax in its thermal design as the all the heat generating parts, inductors, FETS and wiring is all above where the electrolytic caps are located.

    The XW also via the heat sink wicks the thermal energy to the ambient air external to the enclosure, so the actual temperatures the caps are subject to is about 20-30C lower at full power than what the MX series sees

    The MX has the caps sandwiched between the Fets and Inductors and also are at the Top of the unit where the highest temperatures are located. The XW, the Inductor and FETS are mounted to the massive finned heat sink, The MX/Flex use a small fan to pull the hot air out from inside the unit where everything is subject to the elevated temperatures.

    The XW has four, 3300 uf, 105C HC temperature rated caps to the MX, 3 2200uf caps ( XW 13,200 uf to MX 6600uf ) which is well with the rating of the caps with a 60 amp single phase switcher. The XW was extensively thermally modeled and tested, it never ever exceeds or even any where near the 105C/ 10,000 hour rating of the parts.

    Yes the XW is a single phase buck switcher, the others are multi-phase ... but under no condition is the MX ( or Flexmax ) better thermally on the components in this designer opinion.
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil
    Yes the XW is a single phase buck switcher, the others are multi-phase ... but under no condition is the MX ( or Flexmax ) better thermally on the components in this designer opinion.

    And you are certainly entitled to your opinion, Henry...

    However, I'm not sure what controller, in the field now, is of your hardware design ?

    Unless, maybe, Xantrex listened to you when they incorporated those cost saving design compromises into the XW controller.

    One thing I agree on, in the XW, is the huge heat sink to get rid of the fans. However, that does not get rid of the heat from the middle of those filter capacitors that are going to (most likely) fail early. That's OK.... It may just have to be serviced earlier (caps changed out) sooner that some other controllers out in the field today.

    boB
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Re: MX60 leaking oil

    " It's maintenance just like maintaining your car by replacing the brake pads every so often and changing the oil and battery and bunches of other stuff. You get so many miles out of it."

    I once asked a famous boat maintenance author about what he did for preventative maintenance. He said, "when it breaks I have to fix it"

    At $500 for a MPPT CC I keep a spare and when one breaks I sell it after it is repaired. There sure alot of fan failures.........
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: MX60 leaking oil
    boB wrote: »
    Unless, maybe, Xantrex listened to you when they incorporated those cost saving design compromises into the XW controller.
    boB

    Like Outback listened to you for the Flexmax "improvements" and tracking software

    Disappointed in you boB ...

    SG
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil
    Like Outback listened to you for the Flexmax "improvements" and tracking software

    Disappointed in you boB ...

    SG


    I had nothing to do with the Flexmax and its tracking software. And, I do not defend the Flexmax tracking software either. Just the basic hardware design of the buck converter itself.
    There is no question in my mind that that particular part of its design is going to help capacitor life, which is what this whole thread is about.

    There are several flaws in the XW controller design. Capacitor life will most certainly be one of them, fan or not. A fan would help the life of the caps in the XW, but that would only make it take a wee bit longer before servicing will be necessary.

    Also, I haven't been involved with OutBack's designs since summer of 2005. :D

    boB
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: MX60 leaking oil

    Hey - When is the new mid-range MPPT charge controller coming out. I've gotten tired of holding my breath - gotta breathe soon....
    relaxing .. relaxing .. getting the last bits of O2...
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil
    mike90045 wrote: »
    Hey - When is the new mid-range MPPT charge controller coming out. I've gotten tired of holding my breath - gotta breathe soon....
    relaxing .. relaxing .. getting the last bits of O2...

    Whos' ? MidNites' ?

    This fall. Late September, October I think.

    boB
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: MX60 leaking oil
    boB wrote: »
    I
    Also, I haven't been involved with OutBack's designs since summer of 2005. :D

    boB

    Ditto for me on Xantrex , last was Fall of 2005 ...
  • halfcrazy
    halfcrazy Solar Expert Posts: 720 ✭✭✭
    Re: MX60 leaking oil

    I think we can all agree Xantrex went down a different path and definitely had just one thing in mind $$$$. I will say it is a decent piece if you happen to get one of the good ones. From my testing it doesn't seem as productive as the FM60 but it doesnt get stuck or stay awake at night either. I think the 4 wire design will cause them many heartaches though.
  • nigtomdaw
    nigtomdaw Solar Expert Posts: 705 ✭✭
    Re: MX60 leaking oil

    My XW60 been up and running for over a year now, no problems does not get very hot even when its in float for more than half a solar day which is most days now the long hot summer is here.

    Looking forward to the launch of the Classic-Clipper

    Great Forum, Interesting as always !
  • lorelec
    lorelec Solar Expert Posts: 200 ✭✭
    Re: MX60 leaking oil
    boB wrote: »
    Normally this 50% Duty is most stressful. In the MX60, FMxx, Classic (and I think, the Apollo controller design), 50% duty ratio is (theoretically) 0 (zero) ripple current because of the buck converter design.

    boB

    OK, I didn't realize the MX60 was dual phase.
    Ripple in of it self has little bearing on cap life, its the heat that the current swing of the FETS switching which causes a voltage drop due the ESR of the cap that creates heat, similar to a wire heating when a voltage is applied on a wire and the resistance will heat the wire. This heat, in combination of the ambient temperatures the components are mounted in is what determines if the parts is being used within specifications, not the ripple its being subject too.

    I think we're talking about the same thing: ripple due to the switching action of the FETs. Without that ripple, there would be no current through the cap and so no heating.

    Marc
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: MX60 leaking oil
    lorelec wrote: »
    I think we're talking about the same thing: ripple due to the switching action of the FETs. Without that ripple, there would be no current through the cap and so no SELF heating.
    Marc

    There, I fixed it for you

    It makes no difference where the heat comes from, ripple or adjacent components. You could have a 100 phase converter with zero ripple, if the caps are in a space with elevated temperatures it can fail just as fast as a single phase units higher ripple and self heating

    The key is keep what is hot from what is not
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: MX60 leaking oil
    There, I fixed it for you

    It makes no difference where the heat comes from, ripple or adjacent components. You could have a 100 phase converter with zero ripple, if the caps are in a space with elevated temperatures it can fail just as fast as a single phase units higher ripple and self heating
    The key is keep what is hot from what is not

    Say, this is a pretty hot topic ! :D

    Yes, of course internal controller temperature will heat those caps up and reduce their life, ~BUT~, high capacitor ripple current heats them up from the "inside".

    The internal capacitor temperature will be higher than the outside capacitor surface temperature, (measured at any one spot its surface). The more ripple current, the higher that internal capacitor temperature will be.

    But, yes, high ambient controller operating temperature, from FETs and inductors, etc, also contributes to the filter capacitor's faster demise.

    A fan (unfortunately they can fail too), will help cool the capacitors as well as the other components. I don't think, (although now I will have to look again), that the XW CC has its main filter capacitors heat sinked ?? If so, they might even be heated MORE from the FETs, and get hotter, faster. It's good to get the heat out of the whole unit if possible.

    boB