Convert your 48V charger into 100V for 24s for CHEAP !

I am tired of parallel charging 2 x 12s .. that make alot of aditional connections and contact resistance.

A great cheap and easy solution i was thinking about for long time and finally decided to test is teh use of a 1:1 ratio DC-DC converter.

Teh idea is to connect the output of a 48V dc-dc in serie with his input to double the voltage. Tmost of the dc-dc have an isolated output vs input.. SO YOU CAN connect the output with the input!

Few dc-dc that take 48V input have variety of output from 3.3V to 48V. by connecting a dc-dc with a 48V in and 48V out you can get 96V out.. and play with teh trim pot to rise the total output to teh desired voltage around 100.8V is you need for 24s lipo... 8)

For sure the total power output will decrease a bit and you might loose aroud 7-8% total power due to the efficiency of the dc-dc that is around 90%.. but it's just for half the voltage ( the 50% that represent 48V out of 96 of the dc-dc output)..

Let say you bought a 48V 10A ( 500W) meanwell psu. by connecting the 48-48v dc dc with it you will get around 100V at 4.5 to 5A out... for around 450W..

but.. you will be able to keep your pack connected in 24s!.. no more need additional serie parallel connection for 2 x 12s !!

I will test that and comeback with review once i'll get the dc-dc arrived from ebay.

DC-DC are great cause you can accehive the dedired voltage from any lower psu by connecting their output in serie to the input!


I already did that to convert my 1500W meanwell 48V to a 66Vout for my friend bruno for chargin his battery with two BXB150 48vin 5v out in serie to acheive +10V to the output. ( the 48V meanwell can do up to 56V out.. so 56+10=66V

If you search on ebay you wil lfind CHEAP dc-dc.. not the one you see at 75-100$ but most around 25-30$ each.

That is a good solution for compact size charger. and for those who dont want to have to buy another psu connected in serie with the first!

I found some that output 250W out for 20$ !

Doc

This is sorta the out Im looking for but hell Im thinking about just getting another meanwell to series with my current hacked one... a few bucks more I can just get another power supply. Im not knocking your find i think its awesome but man ur like 10-20 bucks away from doubling raw power.. Seem like there would be some bottle necking and not being able to unleash the full potential.

icecube57 said:

This is sorta the out Im looking for but hell Im thinking about just getting another meanwell to series with my current hacked one... a few bucks more I can just get another power supply. Im not knocking your find i think its awesome but man ur like 10-20 bucks away from doubling raw power.. Seem like there would be some bottle necking and not being able to unleash the full potential.

Click to expand...

yeah.. but.. i forgot to mention that in my case, doubling power... mean from 1500w to 3000W.. adn the little most availlable 120V 15A socket are hard to find for 3000W :lol:

that's why this solkution is the best..

problem is that for 100V, power supply like meanwell are very rare for the power we need and at decent price!

They generally only make up to 48V...

and for those who use 1000W+ power supply, doubling power can not be acheived with normal 15A ac plug... unless you use 220V.. but most common socket are 120V and when you travel with your ebike and want to stop for charging the 120V socket are the most common...

Doc

You are a very creative man docbass -
I'm curious how this works out!
What dc-dc did you order?

auraslip said:

You are a very creative man docbass -
I'm curious how this works out!
What dc-dc did you order?

Click to expand...

19$ for doubling voltage of your 48V psu :wink: for up to 500W total!... or 38$ for 1000W total... not bad!


Here is the link: http://cgi.ebay.ca/ws/eBayISAPI.dll...794695&ssPageName=STRK:MEWNX:IT#ht_2457wt_905



Doc

hey doc,
I was thinking about this today. Specifically, I had in mind using these with lukes infamous 3kwh 48v HP server power supply that goes for $40....

If the dc-dcs are voltage adjustable and will limit the current, you could build a very, very powerful and flexible charger for less the $100.

Better than using a bunch of meanwells in series and cheaper too.

Just stack all the dc-dc that you need to get the current and voltage you desire.
I suppose the same could be done with recycled (free) 12v computer PSUs.

It would be nice to current limit them to where you could run two dc to dcs off 1 48v Meanwell. Even if the output was 250w total I would be happy with such a flexible power supply.

icecube57 said:

It would be nice to current limit them to where you could run two dc to dcs off 1 48v Meanwell. Even if the output was 250w total I would be happy with such a flexible power supply.

Click to expand...

If the Meanwell has the current limiter set where you want it, it should work with the dc-dc converters over some voltage range. I'm not sure what the dc-dc converters do when the input voltage sags, but most of them will try to compensate. If the main supply goes into limiting, it will drop it's output voltage. As long as the pack voltage is not too far below the final output voltage, it should be able to limit OK.

My Meanwell has a current limit of 10A are you saying that I may be able to series one of these DC to DC with one mean well and have a low amp high voltage charger. Im looking at the data sheet and depending on what voltage its outputting it limits the current in the same ratio certain ratio.

This converter is not a constant power device – it has a
constant current limit. Hence, available output power is
reduced by the same percentage that output voltage is
trimmed down. Do not exceed maximum rated output current.

The converter is rated for a maximum delivered power. To ensure
that maximum rated power is not exceeded, reduce maximum
output current by the same percentage increase in output voltage.

Current limit 5.31 Min 5.99 Nominal 7.29Max Amps Output voltage 95% of nominal

Output Voltage Programming
The output voltage of the converter can be adjusted or
programmed via fixed resistors, potentiometers or voltage
DACs

In theory this could be used to boost voltages and current limit at the same time and stay under the power limit of 1 meanwell PSU.

It really work when using them with a current limited power supply. Best when connecting the dc-dc in serie with the power supply is to have DC-Dc with higher amp than the power supply... and also to have protecting diodes in parallel with the power supply and the dc-dc... just like BMSed battery connected in serie. This will avoid that one of the source in serie ( dc-dc or psu) goes in reverse current due to overload caused by of the overall current on the circuit..

Here is one simple i made to boost my 48V 1500W charger from 56V ( max set possible) to 66V for my friend's 16s Konion battery.

We was able to cary the same charger but to charge both of our battery ( my 2x 12s to 24s battery) and his 16s battery .

It is made of two of the batch of BXB150 artesyn 48V to 5V 150W converter.

I protected them with diode in parallel ( the to-220 case you see) and adjusted the output voltage to 10V total ( 5V each)

In this exemple you see that i connected that device to a popular 16s LiFePO4 cahrger of 59V output.. and that i get 69V at output.

So you need a current limited supply?

I think the absolute best thing we could do would be to make a setup that'd work with these $30 3kw sever supplies.

That's a ton of power! How would we go about current limiting these?

Same thing as usual with these, adjust the feedback pin based on the voltage across a lowside shunt. For a 3Kw PSU, you would need a pretty high current shunt

Just bought 2 3kW HP power supplies and 3 of the 48V to 48V DC/DC that Doc posted. Need a 90V max charge so we'll see what I can come up with. I'll post a new thread when I get started. More reading to do in the mean time...

Great threads Doc, makes me hopefully that I can find a cheap, powerful charging solution!
-Kyle

How are you planning on limiting the current?

The psu will limit the current. and since the dc-dc can hold more current than the psu will give in serie, the dc-dc will never reach their max current but the psu will

Dont forget that a part of the total current that the psu deliver WILL supply the DC-DC input.. so if you have let say 30A out of the psu, you'll have like 15A in the power supply and 15 for the output..... but the psu wil still see 30A.. so current limiting will be made by the psu.. usually they are power limited.. ( my meanwell RSP 1500 is...) but some are current limited already...

Doc

Sorry if I wasn't clear doc; I was asking nuts&volts how we was gonna limit the current on those new 3kw hp power supplies he ordered. 6kw should be enough to melt residential wiring and the battery pack let alone the dc-dcs!

That's one of those things I'll have to figure out. I'm hoping there is some kind of mod that I can do to limit the current, that's what im going to continue reading about. If I can make both work together or one work with the DC/DC I'll have high power (1.5-3kW) charging for ~$120 which is $200 less than my other options.

I also read that the PSU will act as a CV charger once the delta_V between the PSU output and the battery pack is low enough.

If anyone has any suggestions for limiting the current then let me know.

Nuts&Volts said:

That's one of those things I'll have to figure out. I'm hoping there is some kind of mod that I can do to limit the current, that's what im going to continue reading about. If I can make both work together or one work with the DC/DC I'll have high power (1.5-3kW) charging for ~$120 which is $200 less than my other options.

I also read that the PSU will act as a CV charger once the delta_V between the PSU output and the battery pack is low enough.

If anyone has any suggestions for limiting the current then let me know.

Click to expand...

These psu type are usually power limited in addition to CC-CV(sometime)

In every of these AC to DC high power supply you will have a hall current sense or a little coil with one big wire crossing it. That's what sense the current of the HV dc section.. ( around 340Vdc usually) since the voltage of the HV DC section remain the same ( direct ratio relationship to the AC input) the easy way to sense the power that goes in the chager is to sense the current of the HV dc section.

If you have 3000W and want to use it on a 120Vac 15A circuit, you'll have to make it believe it is at full power ( 3000W)

The way to do it is to modify the current sense section.

you can add a turn to the wire that cross the "coil" or hall sensor... that will double the amount of current measured so your psu will "think" it drive 3000W while it will be doing 1500W..

or you can look closer and find the voltage divider around the hal sensor that measure it and modify the resistor value.

The current sense usually look like this or similar shape and colors:





The wire that cross the center of that O ring is usually connected to the positive or negative of the big HV capacitor.

Doc

Thanks Doc. Modding the current sense circuit makes perfect sense. I think I will attempt to install a pot on the voltage divider circuit once I tear open the PSU and study the PCB layout some.

PSU should be in tomorrow, will start up a new thread then which can document the mods (i dont think i have seen a thread dedicated to those 3kW HP PSU that LFP introduced??).

Thanks
Kyle

How ever you do it let us know because for the price those can't be beat. combimed with a trimable dcdc and you have the ultimate cjarger for the price and modilarity. Might be the next big thing.


A ghettoway to limit current would be a meanwell in series.... that might be my next chargimg setup.

So my Meanwell RSP1000-48 arrived today. I also have 3 48V DC/DC wired in parallel like the diagram below. I trimmed the MW to ~51V out. I have the PR pin shorted on all 3 DC/DC to allow current sharing. The 3 trim pins are also shorted and then connected thru the 3.9K resistor to +OUT to get an output of 38-39V

The problem; I am only getting 23V out of the DC/DC and can only trim the voltage down with any added resistors. I attempted to connect a 8ohm resistor across the output, but this sends the DC/DC into limiting and 0V out. Any ideas?? Do I need to try loading the DC/DC before I turn everything on?



Thanks
Kyle

Figured it out. I can not connect the trimming pin to the same resistor. So i need to run a 3.9K resistor on each DC/DC to trim each one to the 38-39V.

Edit: Ok so a 3.9K resistor drops me to about 41V so im going to put a 330 in series to try to get closer to 39V. Now when i hook up the DC/DC converters i get quite a bit of buzzing coming from the DC/DCs. Is this normal? I assume it is the circuitry PWM to limit the output voltage.

Also connected the 8 ohm resistor across the DC/DC and they went into current limiting, output voltage went to 24.4V. I'll test the setup with batteries soon and see if all works out.

I think i let some magic smoke out

First dumb thing i did was i hooked up the PSU, DC/DC setup up backwards on the battery (+ to +). The meanwell RSP is fine and think the DC/DC simple blocked any current from flowing so nothing, but heat was generated here.

So I flipped the connections and checked that the DC/DC had a voltage output, which it did. I had about 94V from the PSU and DC/DC side and wanted to connect it to a 84.5V battery. When i did this thou (both with the input power on or off) the DC/DCs went into limiting (lots of buzzing) then after awhile something started to smoke a little bit. That could have been some solder or insulation, but the DC/DC were just heating up so im not entirely sure whats going on.

Any ideas what i should try. I may trying removing the current sharing function OR remove the trim resistor, lower the V on the meanwell and see what happens.

edit: pretty sure the 3 DC/DC are cooked. I get no voltage out of them. I have 4 others that I will be a little bit more careful with

Nuts&Volts said:

I think i let some magic smoke out

First dumb thing i did was i hooked up the PSU, DC/DC setup up backwards on the battery (+ to +). The meanwell RSP is fine and think the DC/DC simple blocked any current from flowing so nothing, but heat was generated here.

So I flipped the connections and checked that the DC/DC had a voltage output, which it did. I had about 94V from the PSU and DC/DC side and wanted to connect it to a 84.5V battery. When i did this thou (both with the input power on or off) the DC/DCs went into limiting (lots of buzzing) then after awhile something started to smoke a little bit. That could have been some solder or insulation, but the DC/DC were just heating up so im not entirely sure whats going on.

Any ideas what i should try. I may trying removing the current sharing function OR remove the trim resistor, lower the V on the meanwell and see what happens.

edit: pretty sure the 3 DC/DC are cooked. I get no voltage out of them. I have 4 others that I will be a little bit more careful with

Click to expand...

Sorry to see you got smokes instead of power :x

Maybe next time try first with a ballast resistor or load in serie to avoid excess of current and damaging something..

Than reduce progressively the ballast resistance while everything is still fine until you are directly connected.

What i recommand is to put some blocking diodes in parallel to teh DC-DC output!.. that will avoid reverse current in case where they switch on and off while the psu is still giving current...

Yo can see them on the little +10V prototype i made in teh first page:

Doc

Doc, I wonder if you could give me a little newbie advice about choosing a suitable blocking diode for DC converters? I'm building a charging system using 48V to 3.6V DC/DC converters for a LiFe pack. The converters are rated at 20A output. Do I just need a diode rated at better than 3.6V/20A? It looks as if you used a Schottky diode. Is that important for low resistance or speed of response, or was it just something you had lying around?

Malcolm