diy solar

diy solar

Multiple charge controllers on a single battery 'fight' each other.

K

Kcp

New Member
Joined
Nov 2, 2019
Messages
56
True and false. I never believed this as I've run a Midnite Kid MPPT off 3 155w 12v panels in series with a Xantrex C40 PWM with 2 150w and a 100w in parellel with an Airbreeze 12v wind turbine with its own controller built in all to the same battery bank and everything has been fine- amps are where they should be.

Then someone in the shop started spouting heresy that chargers will conflict and I poo-pooed the idea endlessly.

Until our test bench at the shop. Which uses a Zamp PWM controller that's no longer avaliable due to quality issues- I tested 3 Midnite Kids, all had no amp output. Disconnected the Zamp controller and retested the Kids all came alive with amp output. I claimed vocally 'WTF!' The heretic was dissapointed I didn't believe him. Just an FYI.
 
People put multiple controllers on the same batteries all the time. It should not be an issue.

So what was the issue with the Zamp? A clue might be in this statement: "no longer available due to quality issues"

If the voltages are set with different end-of-charge voltages, the chargers with the lower voltage settings will quit charging before the other chargers. This is not really a problem, but to eek the most out of all of your sources, it is best to set all the chargers to the same voltages if you can. (Keep in mind that we are talking about the end of the charge cycle, so we are not talking about a lot of 'wasted' power generation.... we are just talking about witch charger drops out first)

If the Zamp was driving a high voltage on the batteries, the other controllers may have decided the batteries were fully charged and no longer drove current.

I am not familiar with Zamp. I'll have to go look it up.... but from what you are saying.... it will be hard to look at it in a positive light.
 
Kcp said:
True and false. I never believed this as I've run a Midnite Kid MPPT off 3 155w 12v panels in series with a Xantrex C40 PWM with 2 150w and a 100w in parellel with an Airbreeze 12v wind turbine with its own controller built in all to the same battery bank and everything has been fine- amps are where they should be.

Then someone in the shop started spouting heresy that chargers will conflict and I poo-pooed the idea endlessly.

Until our test bench at the shop. Which uses a Zamp PWM controller that's no longer avaliable due to quality issues- I tested 3 Midnite Kids, all had no amp output. Disconnected the Zamp controller and retested the Kids all came alive with amp output. I claimed vocally 'WTF!' The heretic was dissapointed I didn't believe him. Just an FYI.
Click to expand...
Some people think the earth is flat too.... some are just beyond help, we can't get hung up on those characters. The Midnite Kids can interlink too which is something many SCC's can't which offers a few extra perks too LOL. They are quite an impressive SCC http://www.midnitesolar.com/pages/kid/index.php

That being said, along with what FilterGuy mentioned, the SCC's have to be programmed properly AND be voltage corrected as well, which is something many folks miss. Voltage Correction being that you read the Volts at the battery terminals and then at the SCC terminals and compensating for any voltage drop / difference @ the SCC. Sometimes the drop is minimal or seems so but in LFP / Lithium Land the voltage readings are ever more crucial and that should be correct so proper charging and cutoffs are observed. The SAME applies to Inverter / Chargers, they should also be voltage corrected.
 
The Zamp is a ZS-30AP, however it's rebranded according to a Samlex controller we have that looks just like it and no one is saying where it came from. It's a small 30 amp PWM meant for RV use not residential. It's been discontinued because looking at it wrong could pop a capacitor, it's flush mount so tightening a corner screw too much somehow bends and breaks the circuit board apparently and it likes to freak out and just charge batteries at 15.3 volts all the time for the hell of it. We have about 40 left from a delivery 3 years ago and out of the box 1 out of 3 works.

Midnite controllers can be voltage corrected but that's just a compensation in the controller and I never did that with the Kid and never had problems.

The heretic discovered the conflict with his Blue Sky setup crashing when he ran his generator. Been on a rampage ever since.

The Zamp on the test bench was acting like any other controller, 13.6v @ 4.4a just doing its thing. The Kids came back from a mobile internet install thing so simply wired battery to posts on the table, PV to other posts, RESTING 0 amps. All 3. Took the Zamp out of the system and all 3 Kids entered bulk at 5+ amps.

Conflict does happen and I was shocked to see it.

I'm a RV installer so most systems we install are *bam* all in ones, all matching or solar add-ons exceeding capacity of current SCC, install a bigger one.

But my rig, other DIYers, that's totally different- why this is a thing and is in this advanced forum. Still no idea why it happened.
 
Weird. Did you measure the voltage when the kids and zamp were together? It would be interesting to put an O-scope on the circuit to see if there is any kind of weirdness showing.

Whatever was going on to cause the 'conflict' is the exception. People have been ganging up different charge controllers in boats and vans for a long time. Your description of the 'quality' of the Zamp makes me believe it was doing something it should not have been doing.

Kcp said:
The Zamp is a ZS-30AP, however it's rebranded according to a Samlex controller we have that looks just like it and no one is saying where it came from.
Click to expand...

Be carefull about assuming the guts are the same, some knock-offs can look identical to the original. Even if it did come from Samlex, it could be a totally different controller stuffed into the same case.
 
Sorry to say but this is a real thing. I've done it with zero problems. After doing a T/S on a rig for something else found out his inverter/charger will start charging immediately plugged in but takes 2 minutes clicking on and off before charginging when the generator is on- probably a frequency thing. Think the two are related in theory but way out of my field, just seen them happen.

And honestly wonder if some people may not be seeing a conflict because "it's always been that way" no ever realizing there's a conflict.
 
Kcp said:
Sorry to say but this is a real thing. I've done it with zero problems. After doing a T/S on a rig for something else found out his inverter/charger will start charging immediately plugged in but takes 2 minutes clicking on and off before charginging when the generator is on- probably a frequency thing. Think the two are related in theory but way out of my field, just seen them happen.

And honestly wonder if some people may not be seeing a conflict because "it's always been that way" no ever realizing there's a conflict.
Click to expand...
What is T/S? (I am acronym challenged)

Could you provide more info on what you saw? How was the generator hooked to the system?

An inverter-charger has to synchronize with the AC power and a generator can often be *way* out from what the inverter is doing *and* it cand drift arround. When this happens it can take quite a long time for the Inverter/charger to sync up with the generator frequency and phase. (This is normal) This process is totally unrelated to what is happening on the DC charging side. If that is what you saw, it is not conflict of chargers on the battery.

Note: @Carly found a good video from Victron that talks about how their inverter/chargers work. It covers some of this 'syncing' reasonably well.
 
I have seen the "Out of Frequency" issue when using a regular Mod-Sine Genset (not a Pure Sine Inverter Genset) like those used for construction etc. For example, the Samlex EVO tolerates 57 to 64hz by default (it can be programmed to different values) but once the frequency range goes out of that range, the Inverter/Charger will disconnect from the genset. * I don't know about all brands, I do know that my cheapo Yiyen APC3024 Inverter/Charger did not have any feature to work with the frequency range. Now I have both Inverter & Non-Inverter type generators, the inverter type puts out between 59.7 to 61 hz , while the construction genny put's out 61-66hz (hence where the out of frequency issue popped up). Most "Stand-By" type gensets used for homes & businesses are dialed in to 59-61hz and put out Pure Sine (that's where they get expensive).
 
I would think it's fairly obvious,
The voltage regulator in the MPPT can't get a clean DC reference voltage with a PWM spiking voltage up and down continuously...

To a constant flow regulator, a PWM looks like the battery is over/under charged several times a second.
It simply can't get a clean reference signal so it doesn't connect.

The continuous power chargers will work together since the battery voltage (reference/signal) rises or falls SLOWLY so the regulator can adjust.

Two of the same brand chargers in parallel from the same panel string, one will start charging before the other, and the fist to charge will adjust to compensate for the power the second is producing.
Depending on internal resistance and charge limits, both might produce a full power to charge a low battery, or tickle off to almost nothing on a charged battery.
The regulators 'Seeing' the total voltage/resistance of the battery will work together.

PWM is nothing more than a AMPERAGE/current regulated, full power until limits exceeded, then off, then back on when lower threshold is rached.
That will drive a continuous flow regulator crazy, it sees one high spike, followed by a low spike, then another high spike and can never get an accurate reading.

Why do you think the continuous sample/correction regulators have taken over the market?
 
It is an interesting point that an MPP controller is more likely to put noise on the battery circuit. It would be interesting to put an O-scope on a few different set-ups and see what is really happening on the battery side of the controllers. I suspect there is definite ripple, but I wonder how large it really gets. The batteries and natural impedance & capacitance of the circuit will tend to smooth it out. I also suspect that the ripple is worse when the batteries are nearly charged and not taking all the current you can throw at them.

It would also be interesting to put the O-scope on the solar side of the controller. I would not be at all surprised if the ripple is much worse on that side. The power and current capability on the panel circuit is typically far less than the battery circuit. When the MPPT switches on, I would not be surprised if the voltage on the panel side drops far more than the voltage on the battery side goes up.

The fact that sooo many working mobile systems have multiple chargers tells me a problem would be the exception, not the rule. I also suspect that at least some claims of 'fighting controllers' have been mis-diagnosed and the problem was something else.
 
I couldn't run my PWM charger with an MPPT direct to grid inverter. It drove the inverter psychotic trying to MPPT while a PWM was cyclically switching into and out of the batteries. Pretty much to be expected I suppose if you consider how MPPT works.
 
Keith C said:
I couldn't run my PWM charger with an MPPT direct to grid inverter. It drove the inverter psychotic trying to MPPT while a PWM was cyclically switching into and out of the batteries. Pretty much to be expected I suppose if you consider how MPPT works.
Click to expand...
How was your system wired up? I am not picturing how you would mix a MPPT direct to grid inverter and a PWM charger. Were they both hooked to the same panels?
 
FilterGuy said:
How was your system wired up? I am not picturing how you would mix a MPPT direct to grid inverter and a PWM charger. Were they both hooked to the same panels?
Click to expand...

Hi F guy. 24V, LA, 300Ahr, (2) 265W panels in series, a Xantrex C35 PWM SCC.
The grid-tie was hooked directly to the SArray as was the Xantrex which then fed the LA bank.

I figured that since the batteries were simply topped up kept floated in case of a power outage I could run the grid-tie 99% of the time with its MPPT doing whatever it needed to to deal with however much was left over from floating the batteries.

Unfortunately the Xantrex apparently hooks the SArray directly to the batteries about ever 5 seconds... More than long enough for the grid-tie to completely scratch its current MPPT values and start to hunt for new ones. This caused a lot of clicking and general unhappiness between the two. It's been long enough ago that I don't recall exactly which did what. I think the Xantrex made lots of noise and the grid-tie idiot-lights went in circles constantly.
 
Yes, I can easily believe that it would be a problem if two controllers were hooked to the same solar array. In that case they would tend interfere with each other on the solar array circuit. However, the OPs original comment was about fighting each other on the battery circuit.
 
FilterGuy said:
.....snip.......
Be carefull about assuming the guts are the same, some knock-offs can look identical to the original. Even if it did come from Samlex, it could be a totally different controller stuffed into the same case.
Click to expand...
A few years ago I was looking at Epever/Epsolar SCCs. They were the same and from the same parent company....
Reading different places where people bought a different Branded Tracer SCC, but they were not all the same.
Not all the Tracer series SCC are just a rebranded Epever SCC.
IIRC a Renogy SCC that looks on the outside like a Epever one, is not the same. You can't interchange their MT50 remote meters, either.
I'm going off what I remember from a few years back and may not be remembering it 100% right.
 
Different controllers use different parameters to decide which charge phase they are going to be in. Some have a timed absorb, some measure tail current, some base a time on bulk time. These can certainly cause interactions. Even with two identically set up controllers one will always reach float before the other, when it does it can easily put the other controller back into bulk phase and it itself will contribute nothing, if the solar now drops you could actually use the output of that controller but it isn’t available as the other one is keeping voltage above float but below absorb. You wouldn’t normally notice this unless you were actually looking or logging but it does very definitely happen. A quick heavy load can often trigger the system back into full production by pulling battery voltage down below the restart bulk setting. I have found a proportional dump load to be a good way round this so the dump load is used to keep the controllers in bulk mode by holding the voltage just below absorb so the float algorithm doesn’t cut in. The dump controller senses full charge and then uses the dump load to pull the voltage down to float level. This way you harvest the max power the panels can produce.
 
Older charge controllers did have more of a tendency to fight. I have seen it. I haven't used multiple newer controllers so I can't speak to that.
 
Some modern ones can have data links between them so they can communicate and work together rather than risk fighting and falling out with each other. It’s never a pretty sight seeing a controller in sulk mode after losing a fight while the sun is out and the batteries are still hungry and wanting to be fed.
 
You must log in or register to reply here.
Back
Top