Controller fuse burns

All kidding aside.

Diagnosing or troubleshooting a problem like this is a lot like being a police detective solving a murder and arresting the guilty. He has the physical evidence, a list of suspects and so on, from there its a "process of elimination."
If this was a service call that I was making, not knowing anything about the system or any thing about the cc, just relying on experience, some of the very first things I would check would be;

Verify the correct type of fuse and its ratings, according to Condumex, is being used.

That each pair of positive and negative wires (pv to cc, cc to batt.) are the same length and same type of wire. Also that it is the correct size and type of wire.

Verify polarity throughout the system.

All terminations are clean and tight.

Verify the ratings of the cc as compared to the pv. In particular the Isc. (Spec. sheets for both would be nice here.) Don't forget the 156%. (The Evergreen EC-102-GL has an Isc of 7.28 amps times the 156% means there is a potential of 11.36 amps according to UL and the NEC and this is where calculations for sizing electrical must start.)
http://homepower.com/view/?file=HP79_pg110_CodeCorner
Does a load come on during those times that the fuse "burns up" such as a thermostat controlled fan or something? (Oh! Don't assume that because I elaborated on this one that I think this the problem.)

I had a similar problem with a Trace C-30A+ cc of my own way back when they first came out. A simple on/off cc. It controlled, but the fuse got hot. I never did figure out why. The best I could come up with was that there was a problem on the pc board. What every it was, the damage had been done. I have to admit though, in those days I was a bit of a mad scientist, experimenting with loads, pushing the limits and all. I just don't know which experiment it was that corrupted it. I might have even reversed polarity by mistake at one point or another.

Once those clips that hold a fuse in place get hot they have a tendency to get worse and worse from then on. The metal loses its finish, the metal gets softer, if they are soldered to a pc board, one or both can lose continuity. It may feel and look like a good tight fit when its cold but as it heats up it gets loser and heats up more. (I am sure you are knowledgable of the properties of metals and how, when a metal is heated up and cooled off slowly it gets softer as opposed to if its heated up and cooled off quickly it gets harder.)

Would you provide a link to specifications on your particular Condumex pv cc?
(a link that doesn't require a code name and secret password

)� �

« Last Edit: Mar 31, 2010 06:02 am by Thomas Allen Schmidt »

220 Posts

Apr 4, 2010 02:23 pm

surprised we have not heard back from george, he may have it sorted out?

not much info on those Condumex controllers around either..saw some talk of one called the CX7 (7amps) if we are using that one then bingo.

"Once those clips that hold a fuse in place get hot they have a tendency to get worse and worse from then on. The metal loses its finish, the metal gets softer, if they are soldered to a pc board, one or both can lose continuity."

yup, seen that problem. seems that once those holders overheat even once, things are never the same again. have even gone as far as soldering the fuses into the holder as a workaround.

curious that george mentions the "spring" output. same thing here..the last two weeks of march and the first two weeks of april is when i record the all time highest outputs on my small pv array. (highest wattage not highest daily production)

cheers, dave

6 Posts

Apr 4, 2010 10:31 pm

Hi, sorry for not getting back, I'm having problems with internet connection and haven't found any information on the condumex controller (I got it used, with nothing written). Meanwhile, "my" engineer who installed the system says to use a "bigger" (i.e. 30) fuse.

Also, besides the spring phenomena, this time it was an "after cleaning the panels," bright, bright sunshine, plus spring phenomena....

thanks for your comprehensive suggestions... will work on those and get back to you if/when I get more information.

george anna

6 Posts

Apr 8, 2010 07:56 pm

This is the closest to any type of information that I've gotten on the controller - http://www.solar1.com.mx/calentadores.swf

click on "equipos voltaicos" and then "accesorios sueltos" and you get a "ta-da..." photo of a similar controller. The dissimilarity being crucial... the fuse that burns out is the one that ISN't shown in the photo--- it is placed to the right of the red (low) battery light, and the left of the dial... it also says 20 amps, as does the fuse to the far left (which you do see in the picture)... the one to the left would be for loads on the system. The one on the right (the one that gets hot) is coming in from the panels.

So to answer some of your questions to the best of my ability... I assume I should be using a 20amp fuse. Otherwise, no idea of ratings.

Wiring, polarity and tightness seem to be as they should be.

I don't think it has anything to do with loads, especially since i don't have any loads at noon, except a 3w/hr antenna phone, no fans coming on, I have the equipment in a little earthbag room, with a "green" roof, and it which keeps temps much better than the 100ºF we're reaching these days.

I cleaned off the panels again today, (btw, the panels were practically smoking, very, very hot) had a 25amp fuse, and it got a little warm, but no melting, at least. Maybe I should put a 30amp fuse in. Is there a downside to using a 30 amp fuse?

and I really dislike gender stereotyping, but notice the Anna as my second name. I'm not stupid, just ignorant, and electricity doesn't come easy to me. Didn't know about the metal melting issue, so that was good to know, and I'm not sure about what is meant about the continuity problem. I used to have a problem with getting the caps to turn so that the fuse "connects" and now it seems I don't even need a cap (on some of the fuses)-- I can get them in and out ok, but they're not loose anymore. Is this good, bad or indifferent? If/when they get loose their respective light goes off... I check that as well as fiddle with them and they stay tight.

One of the panels seems to charge lower than it should --- its respective fuse (on the cc) does NOT heat up.

So... with all this information that I don't have, any other suggestions? Otherwise, I'll go back to bugging my eng., who is a man of very few words...

Thanks so much for your time, I really appreciate your effort.

George Anna (it's a southern tradition, what can I say?)

220 Posts

Apr 9, 2010 05:50 pm

Hi George Anna,

please accept my apologies for the gender confusion on my part,<-insert hat tip and southern bow..

consider the following as opinion/speculation on my part.

we have problems here..i would advise taking that controller off line, as there may be some risk of things going real bad real fast.

if you can work on that controller or have someone else work on it (back to the factory would be best) besides checking/refitting that fuseholder. i have a VERY strong suspicion* that we are backfeeding into the pv module from the battery bank. (caused by a failure of the blocking diode(or equivalent)in the shorted mode..this is very rare)

looking at the link you provided and reading some more leads me to believe you have 15amp controllers protected by 20amp fuses...i hate that style of fuseholder. (those are a poor choice for anything that carries even a modest current) still can't get into the spec sheet without the secret password. @!*.*.RHUBARB!

what i might do is see if there is voltage on the pv INPUT side when it's dark. do you have a meter?

while waiting to get that controller fixed i would be tempted (not suggesting you do this) to run both modules on one controller..this will put it over it's rating! (assuming it's the 15amp model)

wish i could be of more help.

kindly keep us posted.

dave

* lots of clues pointing to backfeeding here.

6 Posts

Apr 12, 2010 07:06 pm

Hi Dave, et al,

This is what I did... I checked the voltage from the pv panels on the controller with my meter at dark. Is that right? and there was basically no read out... at times it switched from 00 to .01. Same for both cc.

I see that I erred in a previous post, it is indeed the pv panel hooked up to the controller with the fuse that burns that has a lower input. Yesterday I checked it at about 1 pm and it was at 13.80 checking the box behind (below in this case) the panel. Strangely enough at the controller the input measured 13.30. Compare that to the other panel which measured 17.90 both on (behind/below) the panel and at the controller. Right now at 6 pm they're both measuring a slightly higher charge- measured only at the c.c. (I understand that the extreme heat at midday results in a lower charge.??)

I do appreciate so very much everyone's suggestions, sorry for slow responses, internet signal not cooperating...

thanks,
george anna

220 Posts

Apr 14, 2010 03:24 am

some good info there George Anna,

for others looking on, some more tiny pieces of information on what that controller may be.

http://www.alternativasolar.com/catalog/index.php?cPath=24_29&osCsid=9c733b1a7cc4b347602898c87dc6a4ff

http://www.casolar.net/LineaEnergiaAlternas.php

some translation: (thanks bable fish)

In English

Controller. Measurer for systems Photovoltaic Model CMCX.12-15-20 Description: The controller of load with voltameter, model CMCX-12/15/20F provides in efficient form the protection and signaling adapted for photovoltaic systems that operate with batteries, as much for the apparatuses connected as to the batteries and the same users. In load control he in series consists of a limitor by relay towards the solar modules and another one for disconnection of the loads fed by low voltage of battery. He includes luminous indicators of normal battery, battery low and module in load, fuse, of protection when coming out and selector to inhibit the disconnection by low voltage. The expanded scale, voltameter (10 to 15 volts) this divided in colors to facilitate the user to determine the state of load of its battery (ROJO= low battery, AMARILLO= load A battery average. VERDE= battery with acceptable load). The controller - measuring it is lodged in an attractive aluminum cabinet enameled in black, in whose in front the functions above described are indicated with clarity, besides being the fuseholders, selector of disconnection for cables until caliber 10 with or without connectors. Characteristics: - Nominal Voltage 12 volts. - Current principle of solar modules: 15 amp. - Current principle to the fed loads: 20 amp. - Permissible maximum Voltage: 22 volts. - Levels of operation voltage (batteries seal with lead-acid). - Disconnection of modules (loaded battery): 14.8 + 0,2 volts. - Reconexión of modules 12,9 + -0.2volts. - Disconnection of loads by low voltage of battery: 11.8-0.2 volts. - Reconexión of loads (automatic): 13.2 + -0,2 volts. - Interrupting to select to the automatic disconnection by low voltage (normal) use of emergencia. - Fusible type 3-Ag to the loads in bayonet fuseholders. - Protection against transitory of voltage by Varistor (MOV). - Voltameter with expanded scale of 10 to 15 volts. - Precision -5% of the scale.
___________________________________________________________
this controller looks to me like.. (others, please jump in if you see different)..a shunt type controller. sometimes called a "bang bang" controller. it goes on if the voltage of our battery drops below ~12.7 to 13.1 volts and have pv power available. and turns off at ~14.6 to 15 volts...no wonder they don't like to provide the specs!

your test measurements show that the controller that gets hot was running as it should..a bit high with the voltage drop (still gonna loose a tad more before we get to the bank) the voltage output and bank readings would be good to have as well. the one we thought was running right (the cool one) was off both times you checked it. can't be sure on that one until the battery drops below ~12.7 volts. we really would have liked to see both of them running both times you checked. that is a hysterical amount of hysterics (voltage spread) built into those controllers!

also, please confirm both controllers are going to the same battery bank.

your measurements at night show us that we are NOT shorted back to the modules while off. thats GOOD!

was thinking you might get your "talkative" friend to bring out some leads off that defective fuseholder for a nice blade fuse holder external to the controller. could probably get any audio/tv shop to do that. i would stay with the 20 amp fuse. it would not hurt to put another inline fuse for each module as well rated at 12-15 amps.

i'm not here to sell you anything but you can do a LOT better than those controllers..a nice temp corrected pwm multi stage controller (just one) would put you way ahead of what you are running now, your batteries will thank you for it and there is even a chance of reclaiming those that are scheduled for retirement.

pretty good "bang for the buck" or "pop for the peso" with this one.

http://www.altestore.com/store/Charge-Controllers/Solar-Charge-Controllers/PWM-Type-Solar-Charge-Controllers/Morningstar-Charge-Controllers-PWM/Morningstar-SUNSAVER-SS20L-20A12V-Charge-Controller/p801/

kind regards, dave

edit: "..it would not hurt to put another inline fuse for each.." should read ..it would not hurt to put another inline blade style fuseholder for each...those spring loaded type inline holders are trouble!

further edit: i see a terminology mistake here calling that a "shunt" type controller..more like an on/off series controller that leaves our pv circuit open when off.

« Last Edit: Apr 15, 2010 05:00 am by david ames »

220 Posts

Apr 16, 2010 12:25 am

some random thoughts,

a few times today the specs of this controller came to mind and i tried to think of anything good about it.

then it occurred to me that a controller such as this would be very useful in a "closed" system where we have our modules and loads matched for some kind of daily cyclic function such as walkway lighting or water pumping applications. (made to charge up then run back down each day)

in systems like ours we are missing alot of charging opportunities with these (even if they run as they should). and it would seem likely that they would "mess" with each other when we try to run them in parallel to the same bank.

"mess" <- one may come on a hair before the other has a chance to trip on and the first one on might/will keep the other from even starting up at all that day. yikes!

we did find something good to say though.

cheers, dave

6 Posts

Apr 17, 2010 03:59 pm

Hi Dave,

I'll answer as I can, I suppose the cc isn't exactly the same, because it refers to an amarillo (yellow) colored bulb as indicating average battery charge. In these ccs they have the green v red bulbs indicating battery charge (good v bad). And they have a yellow one which indicates whether the panels are doing their thing or not.

controllers are most definitely going to the same battery bank, which consisted of four batteries, now two.

thanks for the advise about the morningstar cc, as I remember, it wasn't working and what was easily available were condumex... hope to get the US soon and will probably pick it up.

regarding your last note, the system is for general use in the house, both day and night. Although it is a very small system. thanks again...

ah, but what I don't understand is...

voltage output and bank readings would be good to have as well. the one we thought was running right (the cool one) was off both times you checked it.

when do I check the batteries? what time of day, I mean. And where do I check the voltage output? and how do you know that the cool one was off both times I checked it...

thanks again,
George Anna

220 Posts

Apr 19, 2010 01:50 pm

hey George Anna,

he.he..was just re-reading that babel fish translation and having a chuckle, not quite english no longer spanish..reminds me of that movie "spanglish"

might even be easier for you in the original format..down the page a bit.

http://www.casolar.net/LineaEnergiaAlternas.php

looks like the amarillo was in reference to the meter colors. that is a snazzy looking meter. bet that is a big selling point. we can add that to our list of "things that are good".

there can be little doubt that the 17.90 voltage is the open circuit voltage of the module, you confirmed this with the foresite to take two voltage readings at different points in the line..both readings the same means no power going through the wires.

with just the few measurements you made at 1pm, if someone were the curious type they could extract the module operating temperature and apply the corrections to all of the performance data provided at standards to plot a new real time performance chart for the module.. with your readings of the "running" module one might extract the line resistance (for that run), make a guess at the wire/distance used, and get an idea of the battery state/rate of charge. (current readings are nice too but a bit more complicated and dangerous so we can guess at those for now)

the data chart for the ec-102-g

http://www.evergreensolar.com/upload/pdf/us/DS20-007024_v3.8%20Manual_Cedar.pdf

just for amusement, exchange where 16.7 is on the lower line of the i/v chart for 13.2 then move from the 13.2 (the point that was 16.7) to where 13.8 is to get an idea of where we are running. our main goal here is to stay away from the big dropoff on the right of the chart. looks ok so far (but well on our way down), still have room for about 20*f more on the temp as well (before we fall like a six flags rollercoaster)... (but it would still be nice to reclaim most of that 1/2 volt line drop)

i always get a kick out of comparing the real running measurements against the "paper" calculations after a system is up..sometimes we can pick up on a bad connection or problem that we can't see by just looking.

if you take any more voltage readings while running- a set of readings taken about the same time at the pv back, the controller pv input, the controller battery output and right at the battery terminals will give us a better overall view of the system. battery voltage readings are always good to have throughout the day. two that would be good to know is the voltage at the bank terminals right before sunup and right after sundown.

still should address that faulty fuseholder.

this next bit of advice would most always be considered bad...but, it may be worth while to bring those other two batteries back online and move that second controller (the cool one) over to that one to see if it will start up and run without interference from the other (hot one). in effect two mini systems?

question: have you ever seen the voltage on "the one that was high" aka "the cool one" down into the operating range? 13-15 volts?

are we having fun yet?

kind regards, dave

6 Posts

Apr 19, 2010 03:44 pm

ah, Dave, I'm not sure if we're having fun, yet, but your last comment certainly gave me a chuckle.

All rightie, will carry out assigned tasks soon, will also try to catch my engineer and see if he has any idea why the last controller didn't work. (sorry, that wasn't clear on my last message ... I originally had another controller- I think it was a morningstar - but don't remember well. I had problems with the system in general and brought the engineer in who cleaned up the system, and changed the controller...)

thanks again, and I'm learning lots! I think....

and are you sure that's the right link for the condumex controller? I get a page that refers to condumex, but it's all about (the website) solar water heaters? Huh

George Anna

Thomas Allen Schmidt

679 Posts

Apr 26, 2010 05:53 am

"Didn't know about the metal melting issue, so that was good to know, and I'm not sure about what is meant about the continuity problem."

In this case, continuity, is just a word used to describe the flow of electricity. A free flowing continuity is good. Continuity that is not free flowing can cause heat. Think of it in terms of a creek flowing unobstructed and a creek that beavers have damed up. The beaver dam creates resistance to the flow of water and floods. In an electrical circuit resistance to continuity creates heat.

I see from David Ames recent post that a complete spec. sheet and manual of the Condumex CMCX 12-15-20 is still not attainable but I do see the specs. as advertised and think I understand most of them.

-Nominal voltage - 12 vdc
-Maximum amperage from PV modules - 15 amps.
-(This one I am not sure about) - 20 amps.
-Maximum open circuit voltage - 22 vdc
-For use with lead acid batteries.
-Disconnects PV from batteries at 14.8 vdc
-Reconnects PV to batteries at 12.9 vdc
-Disconnects load from batteries at 11.8 vdc
-Reconnects load at 13.2 vdc
-(Not sure)
-Something about a blade type 3 amp fuse.
-Lightning protection.
-(Not sure about this one) - 10 to 15 vdc
-Efficiency factor - 5%

From what I can tell as compared to the PV module specs. for the EC-102-G, all of that looks OK. There is still the question of that 20 map rating. A 15 amp charge controller with a 20 amp??? Load control???
(This is where that "mad scientist" in me might have tried both PV modules on one charge controller just to see what happned. Supervised of course.)

"I used to have a problem with getting the caps to turn so that the fuse "connects" and now it seems I don't even need a cap (on some of the fuses)-- I can get them in and out ok, but they're not loose anymore. Is this good, bad or indifferent? If/when they get loose their respective light goes off... I check that as well as fiddle with them and they stay tight."

I suspect from this statement that, damage to the fuse holder from excessive heat has happend already.
Is there any way to un-install this fuse holder and re-install a different type of fuse holder?

I wish I could actually see, feel and, smell it during operation, that would tell me conclusively what the problem is but not so much what caused the problem to start with so that it doesn't happen again.

Thomas Allen Schmidt

679 Posts

Apr 26, 2010 06:25 am

It just occurred to me that, in all that I have seen about these Condumex charge controllers, I haven't seen the UL listed "stamp" of approval.
I might be asking a dumb question George Anna, but is that Condumex CMCX 12-15-20 UL Listed? UR Listed? Or any of these to follow in this weblink?
http://www.safetylink.com/

Posts

Jun 11, 2010 05:51 am