RV Converter (WFCO WF-8955PEC) 120vac to 12vdc charger compatibility with LFP?

[Barry Timm]

Hi All

I have a second RV into which I'm building an ELectrodacus SBMS0 with 3 x DSSR20. and a 600ah bank of LFP prismatics @ 12v.

I want to use the existing RV Converter which normally charged the 12v FLA house batteries (and provided 12v DC to the DC panel), to now charge the 4S2P 600AH 12V LifePO4 battery bank.

I recognize that the converter cannot be directly controlled by the SBMS0 which is unfortunate, but my plan is to use a programmable DC voltage sensing relay to interrupt / stop the ac power to the converter's input terminals if the battery bank voltage reaches some voltage level, like 14.0v. The converter's charger has a bulk stage that reaches 14.4v, then switches to absorb at 13.6v until it detects no loads then switches to float at 13.2v. As I can program the converter to switch OFF at any battery voltage, say 14.0v or 13.8v to be safer (there is no cell level monitoring of course), and as we don't intend to be connected to a static ac power source for long periods of time (RV campground power pedestal or 120vac generator, I think that this converter charging should be acceptable.

As we will have a little solar (800w) through 3 x DSSR20 and also a DC-DC 12v charger from the RV's chassis battery/alternator to help charge the LFP bank as well, we will rely on this RV converter charging only very occasionally.

The main question is whether there is something with the converter's charging characteristics that make it completely unsuitable for charging my LFP bank, or not?

Can anyone say if it is absolutely not suitable or not? (given that I do understand in the ideal world that it is always preferable to have the SBMS0 directly control ALL charge and load sources, but in this case, I will be relying on the voltage sensing relay to stop the charger after LFP achieves 14v (or 13.8v to be safer)

So, not ideal by any means, but is it workable, or not?

Thanks!

Barry

Manual for the converter is here... https://wfcoelectronics.com/product/wf-8955-power-center-3/

[Dacian Todea (electrodacus)]

Barry,

What is you reason for not having the SBMS0 control this charger ?

The EXT IOx set as type 1 can control a SSR on the AC input side and stop this grid charger when any cell gets to 3.55V so battery is fully charged.

There is no impact to battery life if you top cycle the LiFePO4 it can easily last for 10 to 15 years.

On the other side doing what you intend is very dangerous as over time one of the cells can drift and exceed by quite a lot 3.6V even with voltage set at 13.8V

[Barry Timm]

Dacian, I may have misunderstood a prior reply on a similar topic, where I thought you mentioned that an SSR to do this type of switching would generate a lot of heat and require a substantial heatsink and fan to keep cool. It seemed like an inefficient application to me. Perhaps I misunderstood?

Is there a specific type/model of SSR that you might be able to recommend to control the ac switching of the input to a 120v 55amp converter from a 12v DC control source?

[Dacian Todea (electrodacus)]

I'm referring on a SSR on the AC side.  They will have a voltage drop of 0.8 to 1V so say 800W charger at 120V  = 6.66A thus at 1V drop you will have 6.66W loss on the SSR less than 1% loss (it is grid and charger is maybe just 80% efficient so it will not make a big impact in overall efficiency). The 6.66W could be handled with a passive heatsink no fan. I was probably referring at a DC SSR where yes the losses will be much more significant depending on the SSR used.

You could wait and use a DSSR50 when available on the DC side but at 55A it will have similar loss a bit above 4W so not much of a difference and worse if you consider the savings of the charger witch is fully disconnected when charging is not needed.

[Barry Timm]

Dacian, would this SSR work to switch the AC-IN to my RV converter charger ON/OFF from the SBMS0?

https://www.amazon.com/TWTADE-SSR-40-3-32V-24-380V-Solid/dp/B079BBFKP4/ref=sr_1_3?crid=VYSW5A2AKUOM&keywords=SSR%2BDC%2B12v%2B120v%2Bac&qid=1674324062&sprefix=ssr%2Bdc%2B12v%2B120v%2Bac%2Caps%2C617&sr=8-3&th=1

[Dacian Todea (electrodacus)]

Yes that one will work and the heatsink will be sufficient at just 7A or so.

You can get one with even larger heatsink if you want to keep it at lower temperature https://www.amazon.com/Baomain-SSR-60DA-Module-3-32VDC-24-480VAC/dp/B01EMYF20W

[Barry Timm]

Perfect, thx!

[Barry Timm]

Dacian, I am using the SSR you suggested above to switch the RV converter on/off through PC817 opto-isolator controlled by extio4.

I have shorted the 3 k-ohm resistor on the output of the PC817 but the SSR is not switching. The voltage appears to be too low at 1.6v on the SSR Positive Input from the PC817.

When I measure the V-out on the PC817 it is 11.4v so down from 13.3v at busbar. But the G-out on the PC817 is only 1.6v which is the same reading as at SSR positive.

If I test the SSR switching by temporary connecting 13v busbar to SSR positive terminal, everything works.

Question is why the voltage is so low from PC817 V-out to G-out with input side active?

Thoughts? Should I remove the external 1 k-olm resistor on the PC817 output Side?

On Saturday, January 21, 2023 at 11:15:06 AM UTC-7 electr...@gmail.com wrote:

[Dacian Todea (electrodacus)]

So you say there is a 1.9V voltage drop on 1kOhm resistor and then a 9.8V across the PC817 output ?

Best guess is that you did not shorted the 3kOhm resistor (not sucessfully at least).

You can measure the voltage drop across the 3kOhm resistor and it should be 0V if short circuit is done correctly.

[Barry Timm]

I measured the resistance from the leg of the component just ahead of the resistor where the trace to the resistor comes from, to the output terminal. It measured 3 k-ohm before the wire was added and 0 k-ohm after the wire was soldered across the reisitor path. I will measure the actual voltage and report back.

[Barry Timm]

Still trying to get my RV 120vac-12vdc converter working through the EXTIO4, PC817 and SSR.

These are the voltages and mA I measured.

The SSR works perfectly when I hot-wire the SSR INPUT Pos to the local 12V DC panel Pos and the SSR INPUT Neg to the 12V DC panel Neg.

----- Voltage is 13.22v

----- mA through the SSR when working is 12 mA

The SSR does NOT work through the PC817, when wired as follows:

13.26V Pos <-----> 1 k-ohm resistor <-----> PC817 OUTPUT V-term

PC817 OUTPUT G-term  <-----> SSR INPUT Pos.

SSR INPUT Neg <----->  DC panel Neg busbar.

I tried two PC817 modules wired exactly as above, but one with the internal 3 k-ohm resistor shorted with a wire and the other with the internal 3 k-ohm resistor in place, both with an external 1 k-ohm resistor between 12v+ and Output V-term.

On a PC817 module with the internal 3 k-ohm resistor SHORTED with a wire, and a 1 k-ohm resistor IN PLACE, Red jumper removed 

-----Voltage is 13.26v at the PC817 V-term and 13.25v at the PC817 G-term

------mA with SSR connected is 1.77mA (SSR is not switching on correctly)

The SSR also does NOT work through the PC817, when the 3 k-ohm resistor is IN PLACE, and wired exactly as above.

-----Voltage is 13.26v at the PC817 OUTPUT V-term and 13.25v at the PC817 OUTPUT G-term.

I am at a loss as to what to test next. 

It seems to me that the SSR is not receiving enough current when switched through the PC817 even with the 3 k-ohm resistor shorted with a wire and the 1 k-ohm external resistor in place. 

Any ideas? Remove the 1 k-ohm external resistor?

[Dacian Todea (electrodacus)]

12mA is way more than typical SSR that requires just 2 or 3mA

If the SSR requires 12mA then you will need to reduce the 1kOhm resistor value to at least 470Ohm or as you have 1kOhm reistors have two of them in parallel for 500Ohm

That way at 12mA voltage drop will be 0.012A * 500Ohm = 6V so there will be another 6V or so available for the SSR sufficient to turn that ON.

[Barry Timm]

I installed two parallel 1 k-ohm resistors between 13vPos and PC817-Out-V-term as recommended.

SSR still not switching on.

With the SSR disconnected from PC817, the voltage at PC817-OUT-V-term is 13.56v and at G-term is the same 13.56v

With the SSR connected from PC817 (but SSR not switching on), the voltage at PC817-OUT-V-term is 12.65v and at G-term is 1.647v

The current between PC817-OUT-G-term and SSR when connected is 1.835 mA.

[Dacian Todea (electrodacus)]

So you have 12.65V at V-terminal and only 1.65V at the G-terminal while only 1.835mA flows through it?

That will mean that equivalent resistance is ( 12.65 - 1.65 ) / 0.001835 = 5994Ohm that is almost double that is should be if the 3kOhm resistor was not shorted.

The ( 13.56V - 12.65V ) / 0.001835 = 496 Ohms makes perfect sense as you have two 1kOhm resistors in parallel.

So there is clearly something wrong with your PC817 board and most likely the 3kOhm resistor is not shorted and even so it seems is almost double that at close to 6kOhm thus not sure what is wrong.

Do you have enough voltage at the input of that PC817 channel ?

[Barry Timm]

OK!!!!

So I switched my testing to the INPUT side of the PC817, and removed all the other charging devices from the PC817 INPUT modules that were daisy-chained to EXTIO4 to isolate the control of the RV Converter SSR through the PC817.

The RV converter is now working (with the other two charging devices removed from the PC817 INPUT).

But, despite all the daisy-chained PC817 RED LEDs lighting up correctly when I switch EXTIO4 on and off, maybe the current or voltage for the three charging devices needs to be corrected when controlled from the single EXTIO4 wires, through the single 1 k-ohm resistor on the INPUT side of the PC817?

I have the following on the INPUT side of the PC817.

13vPos <-----> 1 k-ohm resistor  <-----> EXTIO4+

EXTIO4- <-----> | PC817-INPUT-IN1 (to control 3 x DSSR20) with internal 3 k-ohm resistor on PC817 INPUT.

      ^                   | PC817-INPUT-IN2 (to control SSR to switch RV Converter) with internal 3 k-ohm resistor on PC817 INPUT

      ^                   | PC817-INPUT-IN3 (to control Renogy DC-DC 60 amp charger) with internal 3 k-ohm resistor on PC817 INPUT

PC817-INPUT-G1 <-----> | 13VNeg

PC817-INPUT-G2 <-----> |.      ^

PC817-INPUT-G3 <-----> |.      ^

The PC817 IN ports are daisy-chained together from EXTIO4-

The PC817 G ports are daisy-chained together to 13VNeg

NOTE: The Renogy DC-DC charger is also not switching correctly, but I'm concentrating on the RV Converter issue first. I assume, at this stage, that the same problem is occurring with the Renogy at the RV Converter!

[Dacian Todea (electrodacus)]

You have 3 inputs paralleled and each input is made out of a Red LED about 2V drop on that and a infrared LED inside the PC817 opto isolator that has about 1.2V drop total about 3.2V drop meaning you are left with 10V that are divided in you case exactly in half as there are 3 channels each with 3kOhm and when in parallel they are equivalent with 1kOhm thus you will have half 5V on the external 1kOhm and also 5V on each of those 3kOhms

5V / 3kOhm = 1.66mA and that is the same current that will flow through the red LED and through the infrared LED on the PC817 input.

Now it depends on the brand of PC817 but 1.6mA will be to low as you get 2.5mA when using a single channel but even that is at the limit as you will want 3 to 5mA to be sure it can handle good current on the output.

You can reduce the 1kOhm resistor to at least a third of that value so have 3 or 4 resistors in parallel or a single 270 Ohm resistor.

CTR (current transfer ratio can be anywhere from 50% to 500 or even 1000%) depending on the PC817 manufacturer so you want an input current close to the output current times that CTR. So say CTR is 400% and you need 12mA on the output for that SSR then the input current should be at least 3mA

This is quite complicated unless you understand all this and at the same time is hard for me to give examples for all possible combinations. 

[Barry Timm]

Wow, OK, thanks. 

I did not realize that there was transfer from input to output as I assumed the two sets of circuits (input and output) were completely independent when each was fed with their own 12V power.

I had bought a whole package of many different resistors and the nearest to the 270 ohm is 150 ohm or 350 ohm. I can try both of them, but all of my resistors are only 1/4 watt resistors. Do I need higher wattage versions? I'm happy to buy new ones that are a better match.

Thank you for always going above and beyond in your support. I don't expect this level of support due to my ignorance but very much appreciate it. My other RV has been running an older SBMS0 with DSSR20s absolutely PERFECTLY with 100% reliability for over 3 years, so really want to get your system working on the new RV!

Barry.

[Dacian Todea (electrodacus)]

The PC817 as any other optoisolator has an infrared LED on the input and a solar cell powering a transistor on the output.

If there is not much current on the input there is not much infrared light to shine on the solar cell so transistor can not properly close the circuit as you observed.

So from an electrical point of view the input and output are fully isolated but there needs to be enough infrared light in order to the output transistor to be able to close a circuit with large enough current flowing through it.

A 270Ohm resistor with say 14V from the battery and the other side of the resistor shorted to ground (battery negative) will result in 14V/270Ohms = 0.052A (not a problem for the EXT IOx if that is in series when the short happens).

14V * 0.052 = 0.73W so more than 0.25W and thus after some seconds the resistor will overheat a fail like a fuse. But hopefully that never happens and in normal conditions current will be just at most 14V - 1.2 - 2 = 10.8V / (1000Ohms + 270Ohms) = 8.5mA * 10.8V = 0.092W with most of that power dissipated on the 3 parallel 3kOhm resistors and very small part on the 270Ohm resistor so basically nothing.

You will just need to make sure to have the resistor in some sort of fireproof case so if you ever make a short the resistor will just overheat and burn and nothing around can catch fire from the heat from resistor before it fails.

So you can use the 350Ohm for testing as it should work just fine or just parallel 3 of those 1kOhm resistors if you have that many.