diy solar

diy solar

Shore power for DC Air conditioner (or other constant DC loads)?

jameshowison

New Member
Joined
Jul 30, 2021
Messages
184
I'm looking at an RV build with a DC powered air conditioner (48v Hotspot), but I think the question is relevant for the 12v DC powered air conditioners as well, or any constant draw DC load.

Often we will be connected to shore power (think 30amp 120v). For the sake of argument let's discuss a draw of about 400W to run the HVAC (~8.5 amps at 48v), but it could be more like 700W (~15 amp at 48v).

So, if I understand correctly, the flow would be: 120v --> 48v charger --> 48v batteries --> 48v HVAC.

A few questions:

1. Will the charger be noisy delivering power?
2. How does a constant draw application like this relate to battery charging cycles? Would it just charge to full then bypass? Or does it charge/draw/charge/draw?
3. Is there any point in thinking about bypassing the batteries at all (if that even makes sense?)

How might this set up affect the choice of separate charger/rectifier vs inverter/charger arrangement?
 
I'm looking at an RV build with a DC powered air conditioner (48v Hotspot), but I think the question is relevant for the 12v DC powered air conditioners as well, or any constant draw DC load.

Often we will be connected to shore power (think 30amp 120v). For the sake of argument let's discuss a draw of about 400W to run the HVAC (~8.5 amps at 48v), but it could be more like 700W (~15 amp at 48v).

So, if I understand correctly, the flow would be: 120v --> 48v charger --> 48v batteries --> 48v HVAC.
the charger, load and battery are all in parralel in a typical setup.

Code:
dc_core {
    pos {
        |<->battery.pos
        |<-charger.pos
        |->dc_air_conditioner.pos
    }
    neg {
        |<->battery.neg
        |->charger.neg
        |<-dc_air_conditioner.neg
    }
}
A few questions:

1. Will the charger be noisy delivering power?
depends on the charger.
2. How does a constant draw application like this relate to battery charging cycles? Would it just charge to full then bypass? Or does it charge/draw/charge/draw?
Charge current will go to the load first.
The excess if there is any will go to the battery.
If the load exceeds the charge current then the deficit will come from the battery.
Its actually a bit more complicated but that is the simple version.
3. Is there any point in thinking about bypassing the batteries at all (if that even makes sense?)
No.
How might this set up affect the choice of separate charger/rectifier vs inverter/charger arrangement?
What do you mean by rectifier in this context?

Most Inverter/chargers can either charge or invert but not both at the same time.
The only brand that I know can do both simultaneously is Victron.
 
I'm looking at an RV build with a DC powered air conditioner (48v Hotspot), but I think the question is relevant for the 12v DC powered air conditioners as well, or any constant draw DC load.

Often we will be connected to shore power (think 30amp 120v). For the sake of argument let's discuss a draw of about 400W to run the HVAC (~8.5 amps at 48v), but it could be more like 700W (~15 amp at 48v).

So, if I understand correctly, the flow would be: 120v --> 48v charger --> 48v batteries --> 48v HVAC.

A few questions:

1. Will the charger be noisy delivering power?
2. How does a constant draw application like this relate to battery charging cycles? Would it just charge to full then bypass? Or does it charge/draw/charge/draw?
3. Is there any point in thinking about bypassing the batteries at all (if that even makes sense?)

How might this set up affect the choice of separate charger/rectifier vs inverter/charger arrangement?
The charger will be noisy if the work required to convert 12v to 48v is at or near the capacity of the converter.
 
The charger will be noisy if the work required to convert 12v to 48v is at or near the capacity of the converter.
Right, makes sense. So if the charger has, say, 30 amp capacity and it is delivering 15 amps it'll be much less noisy than if it had 15 amp capacity. (in this case we would be converting 120vac to 48vdc, not 12vdc to 48vdc, but it's the same principle, and maybe you meant 120vac)
 
the charger, load and battery are all in parralel in a typical setup.

Code:
dc_core {
    pos {
        |<->battery.pos
        |<-charger.pos
        |->dc_air_conditioner.pos
    }
    neg {
        |<->battery.neg
        |->charger.neg
        |<-dc_air_conditioner.neg
    }
}

Ah, thanks, that's helpful. In my text drawing I made it seem as though it was a series, but now I'm realizing that these will go to a pos bar and neg bar.

btw, is this a standard notation (what's the significance of the arrows, I'm guessing they show flow direction)

depends on the charger.

I'm looking at the Victron MultiPlus-II-48V-3kVA-35-50-120V https://www.victronenergy.com/upload/documents/Datasheet-MultiPlus-II-48V-3kVA-35-50-120V-EN.pdf

Charge current will go to the load first.
The excess if there is any will go to the battery.
If the load exceeds the charge current then the deficit will come from the battery.
Its actually a bit more complicated but that is the simple version.

Makes sense, so in the context of the Victron Multiplus-II would that all be automatic? The charger has a capacity up to 50A at 48vdc so if the batteries are full would it automatically match the DC loads?

And it sounds like in that situation there would be no cycle implication for the batteries.

No.

What do you mean by rectifier in this context?

I had heard AC-->DC conversion called "retifying" being the opposite of DC --> AC being "inverting". And given that this is AC-->DC but not going into the battery I thought maybe it would be called that. But clearly I have no idea :)

Most Inverter/chargers can either charge or invert but not both at the same time.
The only brand that I know can do both simultaneously is Victron.
Well, we would be running an induction hotplate and microwave (and possibly a 120v hot water heater) as well, so it would be crucial that didn't conflict with the DC HVAC (and sounds like with the Victron the charger would produce the DC for the HVAC, not drawing on the batteries).

Hmmm, but under Shore Power (so battery charging) I'm guessing those would be operating on the AC passthrough rather than the inverter. Although they would be wired into the AC side connected to the inverter (AC1?), so I don't know if those get passed through or run through the inverter. Reading about Power Assist, sounds like AC1 gets grid power directly, then the inverter can switch on to help if the loads get higher than the grid can supply. I don't see that being a big issue for us (perhaps if we were plugged into a 15amp 120v circuit and ran both the induction and microwave).
 
btw, is this a standard notation (what's the significance of the arrows, I'm guessing they show flow direction)
My own invention.
Yes the arrows signify current flow.
That is top tier hardware.
Makes sense, so in the context of the Victron Multiplus-II would that all be automatic?
Its just the nature of a charge sources, loads and batteries being in parallel.

The charger has a capacity up to 50A at 48vdc so if the batteries are full would it automatically match the DC loads?
I think what you need to know is how charging works and how current flows through the topology I described above.
I'm not in shape to do that right now, sorry.
Anybody else want to take a crack?

And it sounds like in that situation there would be no cycle implication for the batteries.
Not sure I understand.
I had heard AC-->DC conversion called "retifying" being the opposite of DC --> AC being "inverting".
Not a commonly used term here.
You've taught me something.
Well, we would be running an induction hotplate and microwave (and possibly a 120v hot water heater) as well, so it would be crucial that didn't conflict with the DC HVAC (and sounds like with the Victron the charger would produce the DC for the HVAC, not drawing on the batteries).
Those would be ac loads.
When the inverter is in bypass mode "PowerControl and PowerAssist" and power assist as described in the datasheet you linked apply.

Hmmm, but under Shore Power (so battery charging) I'm guessing those would be operating on the AC passthrough rather than the inverter.
Yes if those are ac loads and the inverter is in bypass mode then they will be powered by shorepower or generator ac but "PowerControl and PowerAssist" apply.

Feel free to ask follow up questions.
Me and my fellow forum members will be glad to follow up with answers.
 
Thanks.

Yes, I'll look around for how charging works. I tend to think with a water analogy which is useful sometimes but yesterday I had to search "What happens to solar power when the battery is full" thinking that the incoming water continues so it must cause a flood somewhere :) So that's a limit of my analogies.

So I'm thinking it's more complex than pushing current at the battery, because what would stop that? I'm guessing the BMS cuts off current and the charger notices the voltage and both stop the flow.

And for the cycle implications for the batteries I was mostly thinking of the power passing thorugh the batteries (discharge a tiny bit, recharge a tiny bit) and so having DC loads would be cycling the batteries even when they were full. But now I see that the connections are parallel, so the load is on the circuit, but flows from the charger to the load and isn't "pulled in" by the battery, so there would be no cycles (what I really mean is that the longevity of the battery capacity is unaffected).
 
This video explains current flow in the system.
Watch it first.

The read the doco I prepared on charging.
It makes sense to me but I wrote it and I'm a bit weird.

How charging works in the context of LFP batteries

LFP = lithium iron phosphate

Voltage characteristics for LFP cells
2.5 volts is dead empty.
This is a hard limit and should be enforced by the BMS

3.2 volts is the nominal voltage for an LFP cell.
This value multiplied by the cells amp hour rating yeilds its watt hour rating.

3.35 volts is the highest "safe" float voltage.

3.65 volts is dead full, this is a hard limit and should be enforced by the BMS

Constant current aka bulk
During this phase the charger controls the charge current by controlling the charge voltage.
Current flow is a product of voltage differential.
In other words there needs to be a difference in voltage between the battery and the charger to make current flow.
As the battery fills up its voltage increases.
The charger increases the charge voltage to maintain the prescribed current flow.

Constant voltage aka absorption
When the charger no longer has to adjust its voltage down from the configured charge voltage to maintain the the prescribed current flow, the charger is in the absorption phase.
During the absoprtion phase the charge current decreases as the battery voltage approaches the charge voltage.
The absorption phase ends when the charge termination criteria is reached.

Charge termination
Exposing an LFP battery to charge voltage above its full resting voltage of aproximately 3.375 volts per cell causes unnesseccary stress.
For that reason chargers have charge termination logic.
There are different methods for a charger to determine that the battery is full.

Tail current
Charge is terminated when charge current is less than or equal to a configured tail current value in amps.

Absorption timer
Charge is terminated when a configured absorption timer reaches its configured value in seconds, minutes or hours.

Voltage sense leads
This is a less common method. The charger has a separate set of leads that are attached to the batteries terminals. This allow the charger to know the voltage from the batteries perspective. This allows the charger to compensate for voltage drop over the current path between the charger and the battery and also to terminate the charge based on the voltage at the battery terminals.

Comunication protocols
Battery and charger communicate by a data protocol to control the charge process.
I can't say much more about this as I don't have any experience with it.

Float
After the charge is terminated the charger can optionally provide float voltage.
LFP batteries do not require float voltage but it can used as power assist.
At the start of float phase and depending on the configured float voltage the battery services the load and the charger stands by.
As the battery services the load its voltage is drawn down toward float voltage.
As the battery voltage appraches float voltage the charger increasingly takes over the load from the battery.
Eventualy the charger becomes the primary charge source and the battery is maintained at float voltage.
Its like a soft landing for the battery.
If the charger is shut off the battery will instantaneously service the load.
The highest "safe" float voltage is aproximately 3.35 volts which is just slightly below the full resting voltage on an LFP cell.
Engineering is all about tradeoffs, depending on your usage model it may be advantageous to configure float voltage higher or lower than 3.35 volts.

Cycling
LFP batteries don't like to be full.
Even resting at voltages above aproximately 3.375 volts is an avoidable stressor.
LFP batteries also don't like to be empty.
LFP batteries do like to cycle.
Setting the float voltage lower allows the battery to cycle.
 
I'm looking at an RV build with a DC powered air conditioner (48v Hotspot), but I think the question is relevant for the 12v DC powered air conditioners as well, or any constant draw DC load.

Often we will be connected to shore power (think 30amp 120v). For the sake of argument let's discuss a draw of about 400W to run the HVAC (~8.5 amps at 48v), but it could be more like 700W (~15 amp at 48v).

If you will be connected to shore power often, stick with an AC mini split. You will never have the conversion efficiency from a DC powered air conditioner to match a high SEER rated AC mini split.

I spent over a month looking, it was wasted effort except I learned a few things along the way. Efficiency ratings of DC powered were in the toilet. There is a lengthy thread on the topic, many discussions about DC powered but when you get down to efficiency ratings (watts consumed/btu's extracted), there really isn't any comparison, the AC mini split wins hands down.

I won't even touch upon the build quality of these units and the fact these units are fairly new to the market. (room for improvement)
So, if I understand correctly, the flow would be: 120v --> 48v charger --> 48v batteries --> 48v HVAC.

A few questions:

1. Will the charger be noisy delivering power?
2. How does a constant draw application like this relate to battery charging cycles? Would it just charge to full then bypass? Or does it charge/draw/charge/draw?
3. Is there any point in thinking about bypassing the batteries at all (if that even makes sense?)

How might this set up affect the choice of separate charger/rectifier vs inverter/charger arrangement?
 
If you will be connected to shore power often, stick with an AC mini split. You will never have the conversion efficiency from a DC powered air conditioner to match a high SEER rated AC mini split.

I spent over a month looking, it was wasted effort except I learned a few things along the way. Efficiency ratings of DC powered were in the toilet. There is a lengthy thread on the topic, many discussions about DC powered but when you get down to efficiency ratings (watts consumed/btu's extracted), there really isn't any comparison, the AC mini split wins hands down.

I won't even touch upon the build quality of these units and the fact these units are fairly new to the market. (room for improvement)
Thanks. I have seen that long thread, very useful stuff. I think you are discussing the Dometic 12v and the units from undermountac.com (possibly as well as the truck market cabin coolers). I'm planning to use a different dc system, the DC4812VRF from Hotspot Energy. See https://www.hotspotenergy.com/DC-air-conditioner/Specs-DC48.pdf

That one has been around for over 10 years, but not widely adopted in the RV market (mostly off-grid solar cabins etc). It's cooling COP is shown at 5.66 which is very high, that's significantly higher than, say, the Pioneer 120v mini-split which is more like a COP of 4.18, or the Mr Cool which is about 4.5. And then, when running on batteries, you have to subtract the inverter losses (~15%) and suffer the noise of the inverter running.

But yes, you are spot on that when running on shore power I'll take the conversion hit (and perhaps charger noise), but my thinking is I'd rather be less efficient on shore power than on batteries/solar (ie when I'm trying to boon dock).

There's actually possibly a best of both worlds, Hotspot has a "Hybrid" which runs direct of solar or AC https://www.hotspotenergy.com/solar-air-conditioner/ but that requires 220v so I ruled it out (since we are often in 30amp RV sites), but it seems worth exploring for 50amp RVs.

Of course both of these options require an HVAC tech for installation and cost over $2k ... making them equivalent in cost to the undermountac sort of systems but over double the cost of 120v mini-split options.

Anyway, I've ordered the DC4812VRF so I'm committed now ... will definitely post updates once I get it working!
 
This video explains current flow in the system.
Watch it first.

The read the doco I prepared on charging.
It makes sense to me but I wrote it and I'm a bit weird.

How charging works in the context of LFP batteries

LFP = lithium iron phosphate

Voltage characteristics for LFP cells
2.5 volts is dead empty.
This is a hard limit and should be enforced by the BMS

3.2 volts is the nominal voltage for an LFP cell.
This value multiplied by the cells amp hour rating yeilds its watt hour rating.

3.35 volts is the highest "safe" float voltage.

3.65 volts is dead full, this is a hard limit and should be enforced by the BMS

Constant current aka bulk
During this phase the charger controls the charge current by controlling the charge voltage.
Current flow is a product of voltage differential.
In other words there needs to be a difference in voltage between the battery and the charger to make current flow.
As the battery fills up its voltage increases.
The charger increases the charge voltage to maintain the prescribed current flow.

Constant voltage aka absorption
When the charger no longer has to adjust its voltage down from the configured charge voltage to maintain the the prescribed current flow, the charger is in the absorption phase.
During the absoprtion phase the charge current decreases as the battery voltage approaches the charge voltage.
The absorption phase ends when the charge termination criteria is reached.

Charge termination
Exposing an LFP battery to charge voltage above its full resting voltage of aproximately 3.375 volts per cell causes unnesseccary stress.
For that reason chargers have charge termination logic.
There are different methods for a charger to determine that the battery is full.

Tail current
Charge is terminated when charge current is less than or equal to a configured tail current value in amps.

Absorption timer
Charge is terminated when a configured absorption timer reaches its configured value in seconds, minutes or hours.

Voltage sense leads
This is a less common method. The charger has a separate set of leads that are attached to the batteries terminals. This allow the charger to know the voltage from the batteries perspective. This allows the charger to compensate for voltage drop over the current path between the charger and the battery and also to terminate the charge based on the voltage at the battery terminals.

Comunication protocols
Battery and charger communicate by a data protocol to control the charge process.
I can't say much more about this as I don't have any experience with it.

Float
After the charge is terminated the charger can optionally provide float voltage.
LFP batteries do not require float voltage but it can used as power assist.
At the start of float phase and depending on the configured float voltage the battery services the load and the charger stands by.
As the battery services the load its voltage is drawn down toward float voltage.
As the battery voltage appraches float voltage the charger increasingly takes over the load from the battery.
Eventualy the charger becomes the primary charge source and the battery is maintained at float voltage.
Its like a soft landing for the battery.
If the charger is shut off the battery will instantaneously service the load.
The highest "safe" float voltage is aproximately 3.35 volts which is just slightly below the full resting voltage on an LFP cell.
Engineering is all about tradeoffs, depending on your usage model it may be advantageous to configure float voltage higher or lower than 3.35 volts.

Cycling
LFP batteries don't like to be full.
Even resting at voltages above aproximately 3.375 volts is an avoidable stressor.
LFP batteries also don't like to be empty.
LFP batteries do like to cycle.
Setting the float voltage lower allows the battery to cycle.
Fantastic video and explanation, thanks!
 
Back
Top