Key differences between MPPT charge controllers

Hi Experts,

I have been comparing the new Xantrex XW to the "old standard" the Outback MX 60. It is difficult for me to understand the differences between the two and if they matter or not. Is one clearly the better product? Xantrex seems to want to give the impression that their product is more "state-of-the-art". Also, do either company have MTBF statistical data to discern with charge-controller might be more reliable?

Thanks,
Kevin

Comments

  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: Key differences between MPPT charge controllers

    The MX60 is a sweep and sleep controller ... on the firmware I evaluted, every 12 minutes so it would do a mini-sweep and once an hour a full sweep, meaning a power off, to full load sweep.

    The WX is a true dynamic mppt controller, evaluating mppt in the sub-second time domain and in power samples that are sub-watt .. similar to the gridtie mppt units, lots more going on than this, but gives then general idea.

    The performance difference is the MX-60 can sweep durring cloud events getting locked into a non-optimal until the next sweep .. also in climates with winds or intense sun, the panel voltages can exceed 4-5 volts variation per minute ( 48V nomial strings ) so energy can be lost if not dynamically adjusting the vmp point.

    The WX thermally is much better than the MX60 also WX has full Class B compliance.

    Which is better long term .... check back in a year :roll:
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Key differences between MPPT charge controllers

    Considering all that the MX60 has been around for >4 years and that the XW is new, I’d hope that the XW is a better product. Early MX60’s apparently suffered from cooling fan failures, so the XW’s passive cooling and higher operating temperature spec is an improvement.

    Both are rated at 60 A continuous and include a DC-DC stepdown feature. I like the MX60’s range of user configurability. I don’t know how the XW compares. The XW is certified to FCC class B emissions specs, while the MX60 is not. I’ve heard of but a very few EMI/RFI issues with the MX60, so I don’t know if that’s a real issue. One XW spec I've not seen is power efficiency; the MX60's efficiency graphs are included in its manual.

    Virtually all of an off-grid charge controller’s MPPT benefit is realized in the bulk charging stage. The XW’s continuous MPPT is technically superior to “sweep and sleep”, or “occasional tracking”. However, I’ve not yet seen a demonstration or a study that suggests any significant practical benefit.

    Despite being a great controller, the legacy MX60 contains a few quirks. One is that it operates at a fixed “MPP” percentage when the output current is less that 5 A. The default setting is 77%, which I think is too low, as it does not result in maximum output current, which, after all, is the goal of MPPT. But, this setting easy enough to change, and I set my MX60 at 85%, which seems to work well enough.

    The apparent major quirk, and this is a specific issue criticized by the new competition, is that the MX60’s output power drops to zero on occasion. Specifically, the MPPT output current briefly drops to zero once every three hours when operating in Mode 1 or Mode 2 to perform a calibration check.

    The "zero power" (energy) penalty for either of these two operational modes is not too difficult to gauge. Assuming that each Mode 1 or 2 restart lasts 15 seconds, and that output current ranges from zero A to 40 A in an essentially linear manner during that period, then the energy penalty will be (15 sec. / 3600 sec./hr) x (40 A - ((0 A + 40 A) / 2)) = 0.083 Ah. In a 48 V system operating at 56 V, that’s ~5 Wh. If this Mode 1 or 2 restart occurs twice during bulk MPPT mode, the penalty is 0.167 Ah, or ~10 Wh.


    Note that Mode 1 or 2 can be user set for Mode 0, which will limit the full restart sweeps to the morning wakeup routine and eliminate this penalty.

    A perhaps larger issue is that the MX60’s output current drops -- but not to zero -- briefly during its periodic mini-sweeps. Assuming using the ½-90% sweep range causes output current to vary between 25 A and 40 A, that a mini-sweep occurs every six minutes and last 10 seconds, and that the bulk (MPPT) charge stage lasts 5 hours, the energy penalty would be ((10 sec /sweep) / 3600 sec/hr) x (40 A - ((25 A + 40 A)/2)) x (10 sweeps/hr) x 5 hrs = 1.04 Ah. In a 48 V system operating at 56 V, that’s 58 Wh.

    The example above shows that the MX60’s bulk stage sweep operations exact a penalty of about 1.2 Ah/day. In practical terms, it means that a system delivering 40 A will stay in MPPT mode for an additional two minutes before the controller switches to absorb mode. Another way of looking at the example above is that the off-grid system would spend two minutes a day less time in Float mode.

    It’s also possible that the MX60’s present MPPT algorithm would operate less than optimally in dynamic weather conditions. I’ve not yet tried to quantify the penalty from such conditions. But, having endlessly poked, prodded, and observed my MX60 over the past ~3 years, my gut feel is that such an environmental penalty would be a matter of but a few Ah and a few minutes per day, on average. Accordingly, let's say there’s an ~3 Ah total penalty on a day of variable Sun, passing clouds, and variable winds in the normal occasional sweep mode.

    A similar improvement in my system, which rarely delievrs more than 20 of charge current, would amount to ~1-1/2 Ah/day. My system harvests ~100 Ah on a day that it runs runs in bulk mode for five hours, so that could mean an improvement of up to 1-1/2%. And, like like the efficiency benefits of heavy wires and VRLA batteries, every little bit helps.

    So, does continuous MPPT tracking make a difference? Yes, it does. I believe that a tight continuously active MPPT algorithm can reduce the total MPPT bulk mode energy penalty to below 0.5 Ah per day. If using a “perturb and observe” (PAO) algorithm, the penalty can’t be reduced to zero as the PAO tracking (increment/decrement PV array voltage, observe effect on output current) will cause regular although minor reductions in output current.

    From what I’ve seen, read, and calculated, I’d therefore say that the new continuous MPPT algorithms are evolutionary improvement over the MX60’s legacy “occasional” MPPT sweep algorithm. And, because they’re possible, continuous tracking algorithms are now expected in the marketplace. Plus, as we’ve seen, marketing hype will affect customers’ perceptions.

    But, does continuous MPPT render the entire MX60 production run (~30,000 units?) obsolete overnight? I’d say no. But, we’ve gone from a situation where one controller dominated the big controller field to several new continuous MPPT controllers available – or soon to be -- from Apollo, Midnite, OutBack, and Xantrex.


    Ain’t competition grand?

    HTH,
    Jim / crewzer
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Key differences between MPPT charge controllers

    Personally, I am very happy that Xantrex has finally released the XW. Competition is good for the industry ! ;)

    The once an hour MX60 full/reset sweep can be disabled. You forgot that. Also, the 12 minute interval can be adjusted too. Yes, it drops a little bit of power on each side of the max power point, but also does not get stuck on local maxima, which can happen with small dynamic sweeps if not careful. By staying on the sweet spot, which mainly changes due to module temperature and somewhat with insolation, more power can be wasted by continuously tracking and going off that power point (like I say, if not careful). There ARE compromises. Every engineering project has compromises.

    During clouds, except for edge of cloud events, there is VERY little power available during that time, so tracking the power point to a gnat's ass may not help all that much. You aren't going to get even NEAR what you've got when it's 100% sunny...... My experience was that the MX60 worked pretty darn well in cloudy/sunny transitions. What was your ENERGY advantage dynamically tracking ? Energy production is really what counts. 12 minutes is probably too long of an interval. The default was 3 minutes for the MX60.

    I know that the XW has been cheapened up from its original design, conversion efficiency lowered as a result. This doesn't mean it's a bad MPPT controller. I'm sure it will work just fine. It's just that some of the arguments try to make it seem like it's soooo much better than the MX60 , but only time and experience will tell. Supposedly, the MX and/or MF units have dynamic MPP tracking now. We'll see what that does for them.

    No fan is nice as is FCC compliance. It won't necessarily give you any better output though. The lack of any fan can make it worse. Gotta make sure a good fan is used when one IS used.

    We will see, as you say, check back in a year. I am curious how it will fair, but I don't know if there are any fair side by side comparison sites in place right now. It boils down to Watt-Hours of energy produced, given that very closely matched arrays are used for comparison.

    boB
  • nigtomdaw
    nigtomdaw Solar Expert Posts: 705 ✭✭
    Re: Key differences between MPPT charge controllers

    I think a year sat in the wings seems a good idea at this moment in time. I had cash waiting and blind faith for the Spring 2007 release of the XW, with a 1600 kw array to power at 24 volt. Spring came and went no launch countdown for the XW60. A opportunity arose to obtain 4 more 200 w PVs same as my original 8 so now got 2.4 kw to power, still waiting for XW but realised would need 2 *damm 24v*. Oct 07 hear of one XW shipping and appears faulty (this forum) Hmmm glad I didnt rush in, then opportunity favers the brave and Outback announce Flexmax 80, ahh an improved spec and 80 amp capacity one controller will do nicely !. May ship late 07 depending on trials. A opportunity arises Oct 07 for 8 x Sharp 162 watt panels, cash out and bought. Now skint no money for a flashy new MPPT CC at the moment. So the 2.4kw Isophotons will have to sit happily on my 2 Xantrex C40s. The 1.3 kw Sharps will have to be content on an Xantrex C60 I have in the cupboard and Ill sit tight and enjoy this forum, X and OB.s and chew the cud over the next 6/10 months and see if Xantrex Wins XW:D or OB OrdersBack its crown :roll:or the not so new kids on the block MN take the gold medal with the Classic.8) Having unsucessfully tried taping each company to be a beta tester (in you dont ask you dont get...I asked and didnt.. hey Im cheeky) Ill zone out on which brand MPPT CC is best for the moment ..Hey as Jim (Creezer) brought me back to reality correcting me on a 10 % rather than brochure stating 30% gain on annual benefit overall on MPPT over PWM, then now is the time to chill. Let the battle commence and dont discount OB slashing the price of the MX60 as with 30,000 + units sold economy of scale and pi..ing off your competitors is an option that it has in the bank !

    Great Forum Thanks Nigel
  • khotton
    khotton Solar Expert Posts: 38
    Re: Key differences between MPPT charge controllers

    In keeping with the theme of this post, I wonder how the Apollo T80 MPPT charge controller compares to the Outback and Xantrex models? It is the one I am currently favoring because the overall value of the included Trimetric battery monitor and 80Amp output (don't really need 80Amps, but overhead is nice). Does this unit track conditions like the MX60 or "continuously" like the XW?

    PS: thanks to all the experts on this forum. You have taught me so much. I am getting very close to "pulling-the-trigger" on buying my PV system components and the added time I have spent getting "ready" to "aim" and finally "fire" have been time well spent.

    Kevin