Can I use an old Lab Power Supply to charge LiPo directly

[ryan]

My friend has a 0 to 80V / 0 to 18A DC Power Supply salvaged from an old office. It works fine, but it's a little roughed up. It's one of those huge old rack server power supplies.

Can I just plug that into any old US outlet and adjust the power to 75v10A and let her ride?

 

[ZOMGVTEK]

It's better to refer to a supply like that as a 'lab' power supply rather than 'server'. There's quite a bit of mention of server power supplies around here, which you can typically expect to absolutely NOT charge LiPo effectively just by hooking it up.

However, an adjustable supply like that will do the job very well. That's a nice supply, assuming its the one pictured. You can set the voltage (unloaded) to your HVC, put the current dial low, connect the pack, and dial the current up to the intended level. When the current meter starts dropping, its charged.

 

[amberwolf]

Yes. Those are Lab(oratory) supplies, not server supplies. They are PERFECT for this kind of thing, because they are already designed to be settable for the voltage and current limits to whatever you need.

You would set the voltage to the exact voltage you want the pack charged up to, and the current limit to the maximum current you want to limit charging to.

As to whether you can plug into any old outlet, it depends on which model. Some require 230VAC input and some can use 115VAC.

I am not sure exactly which model it is, you might have to look on the plate on the back. but the manuals are available here:
http://www.programmablepower.com/products/Products_Discontinued.htm
for all the old DCR series units. It looks like they are 230VAC units, for the most part. Some are 3-phase only.

If it is the DCR-A series (probably not, since it's model number isn't in the list in the manual), you have to wire the power plug to the right transformer windings for the voltage you're going to plug it into. It has the instructions on how to do that here:
http://www.programmablepower.com/products/Discontinued/Downloads/DCR-A%20Series%20Instruction%20Manual-1057663.pdf
starting on page 2-4.

EDIT: I found while looking for more specific info taht the photo you used above is identical to the one on this ebay auction:
http://www.ebay.com/itm/Sorenson-DCR80-18A-Power-Supply-0-80v-0-18A-/380388132052
So I assume that you don't have a picture of the one you have there, and just used this one because it is the same as yours? If so, then the nameplate on teh back of yours will tell you what the input voltage is already set for:




BTW, if you happen to want to get rid of one of those, I can always use it here.

I can't offer much $$ for it though. And I think I'd have to come get it becuse I definitely couldn't afford to ship it. :lol:

I do have some other things I could trade, including a 900W+ networking supply with multiple voltages (but none really useful for charging, unless you need a huge single-cell charger):
http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=36266

 

[ryan]

So if I can use the back photo you referenced as a reference, regular 110-120v wall outlets should be fine?

 

[amberwolf]

If yours is the same, then yes, but as I said, you have to check the nameplate on your actual unit.

It is best to check that plate for the model number, and then find the manual for it on the page I linked above, and verify in the manual.

I have found that unfortunately with some rackmounted equipment it does not matter what plug you see on something--whoever setup the rack may have used non-standard plug on there, like using a 115VAC sized plug for 230V. Not correct, but it happens sometimes. Found that the hard way with some stuff I got at a salvage yard years back. Everything in the rack was not 115VAC powered, but they had changed all the plugs to that style, and created three "busses" inside using three regular powerstrips. One of them was 230VAC, one was 115VAC, and the last was 48VDC. But the people that moved the rack from it's original location cut all the cables from those strips off at the strip, so I had no way of knowing just by looking that they weren't all normal 115VAC strips. Naturally this resulted in some blown equipment, as the DC stuff didn't have a sufficient input protection against that.

 

[ryan]

Thanks. My "friend" was the eBay auction one! I had to play coy because last time I found a deal like that it got snatched before me by more knowledgable folks on ES.

I live in the area so I can just pick it up and avoid the crazy shipping expenses.

That said, there's one left!

 

[amberwolf]

Just be sure you go with a good solid hand-truck or dolly, cuz you ain't carrying that thing out to your car. :lol:

I have a Sorenson DCS-55-55, and it's 230VAC, and a switching supply, so its a lot lighter, and I can still barely carry it around when I have to. I think it weighs over 50lbs, maybe 75.

EDIT: I guess it's just shy of 50lbs:
http://www.radiomuseum.org/r/sorensen_power_supply_dcs_55_55.html

 

[MadRhino]

Yep, many here would jump on it at that price. I won't, because I would have needed the pair to bulk charge 24s. I suggest you buy them both, for you might want to charge higher voltage some day.

 

[amberwolf]

Well, you can use *any* isolated-output PSU in series with one of these, as long as it can supply the voltage and current you need. It doesn't have to be adjustable, or current-limited, because this one will do that part for you.

 

[ryan]

That's cool. So I could plug in some true server power supplies on the cheap like ZOM does? (should I ever need more voltage)

Another question AW, while I've got your input...

I plan on bulk charging 18S4P with this power supply. I understand my voltage should be set at 74.7V (4.15V per cell), but what about Amperage? Is that simply a function of the voltage or could I set the Amperage as low as, say, 3A and as high as 18A? If so what Amperage do you recommend?

 

[ZOMGVTEK]

The fantastic thing about a supply like this, is the current can be dialed anywhere from 0 up to its full rating. Most Sorenson supplies are calibrated to go up to 115% of nameplate ratings, so it will probably put out over 18A, its just not 'rated' for it. Slower is generally better, but as long as you keep it around 1C it should be fine. That means if your pack is 10AH, 10A charge rate should be perfectly fine. You CAN charge faster, but its not always a good idea. It just depends on your cells.

Also, I would say 4.15V/cell is a bit on the high side for this type of charging. 4.12-4.14 is better, 4.10 is best if you don't NEED the little bit of capacity you will loose. It gives more wiggle room for cell voltage imbalances, among other things.

The output of most any lab supply should be isolated, but all you really need is the rest of the supplies to be isolated and you can series them away. If you ever need a bit more voltage, just get a supply that can put out 20A+, toss it in series with this, and you have a higher voltage supply. The HP ESP135's are probably the most economical choice, since they are quite cheap and nice supplies.

At this price, this Sorenson is a steal. I'd have snagged two if they were close to me.

 

[amberwolf]

For charge current, you need to find out what the cells you are going to use can each handle. Then multiply times the number of parallel cells. That would be your MAX current you could set it to.

I'd recommend setting lower than that, because it's been speculated (possibly tested) that charging at lower rates does the same thing discharging at lower rates does--allows longer lifespan.

For voltage, it's like you say...number of cells times voltage you want to charge them to. Since you're going to use Methods' monitoring system, it can cut off charge if you set it up to do so, when any cell goes too high.

BTW, some of these Sorensons can be programmed to shut down from an external signal, so you wouldn't have to have the Methods charging cut off board. You just have to get the manual for this series of Sorenson off of that link I posted above, then look up how to connect a signal from the HVC output on those monitoring boards to teh Sorenson's shutdown pins.

Some don't have shutdown pins but can have their voltage externally controlled, so you could make a circuit that cut the voltage to zero when the HVC triggers. I'm not sure if there is an output diode on that sorenson or not (schematics are often included in the Sorenson manuals, so you can check), but if there isnt' you'd need to put one on it to block current flow from battery back into Sorenson to be able to use either the shutdown or zero-voltage feature.


You can also have the Sorenson sense voltage at the actual load if you like, rather than at it's output terminals. This will let it monitor what's actually happening at the pack rather than at it's end of the wires, in case there's a lot of voltage drop on them.

 

[ryan]

Whoa, sounds like I'll need an EE to visit to get this up and running! This is above my pay grade for sure.

ZIPPY 8Ah 6S1P LiPo is rated 5C for Max Charge. If I remember correctly that's 5 x 8Ah for 40A? And multiply that times 4P for 160A? Even if it's just 8A x 4P that's more than I'd have available on the 18A PSU. I guess I was asking if it was likely possible to charge at 74.7v 18A or if the Amps go down as the voltage goes up.

I'm hoping once I set it up at 74.7v I can plug it in, turn it on and let it ride until the packs are charged, balanced or whatever else I need. And if I can charge at 18A, it would certainly go a lot faster than the other chargers I was looking at.

 

[amberwolf]

No degree needed, you can just set the voltage and then raise the current up until it reaches the limit you want.

All the other stuff is just gravy if you want it.


That PSU is rated to handle it's max voltage at it's max current. But the batteries will drop in current draw as they near their full voltage; it's just the nature of them.

As for the current you'll need to charge with, the cells *can* take 5x8A each, but lifespan will be shortened by doing that. How much shorter? I dunno. You'd have to do it to find out.

Let's say you set the charger/PSU to 18A. Your pack is 32Ah. So you're charging at a bit more than 0.5C, which I'm sure will be a healthy enough rate for the cells. It will take a bit more than two hours to fully charge (becuase charge current will slow down near the end, making it take longer than it would seem just by the straight numbers).

 

[MadRhino]

You can charge as fast as this PSU can. The high C-rate Zippy can be charged 5C without any damage, and probably could take 10C before starting to heat. I bulk charge my 24s 1p 5.8 Ah packs at 20 A and they don't even get warm.

I'm now thinking about either setting multiple charging stations and/or charging at a higher rate. I use genuine Mean Well PSUs, and this lab PSU that you're buying is better, yet cheaper. Its only downside is the size and weight, but in a garage this is not a problem.

Sorry to hijack your thread, but I have a question while we are on the subject: I have many MW PSU that are of different type and power, mostly of the PSP and SE series. I now charge with two PSP-1000-48 and I know they can be paralleled together with another pair, but can I parallel them with the SE series?

 

[muchalucha]

I have ann interest in electric vehicles, and browse here often. But my main hobby is radio control aircraft. And reading this, I coudlnt take it. Someones about to killm themsleves here.

the answer is NO NO NO NO NO NO you absolutely cna not charge lipos froma cosntnat current pwoer supply. If not immediately, they will eventually puff up and explode. They are being irreversibly damaged. Let me explain.

Yes a constant current powersupply is a great asset for charging batteries, and thats a lovely one. But not directly. Even if you where charging nicad batteries, id say, go get a propper charger, the batteries will live longer. But for lipos, its absotleuyl neccesary, all the time, every time. Lipos are charged in 2 stages, constant current (CC) and cosntant voltage (CV). For the first 80 % of the charge, the lipo is charged at cosntnat current (here your power supply would do great) but after that, to avoid damage to the battery, it needs to be charged at a constant voltage. which requires a micro controller to carefully monitor the voltage and current going to the battery and slowly taper it down to 0 amps when the cells reach 4.2 V each. PLEASE for the love of god, buy a propper charger.

This brings me on to my next complaint about your suicidal attitude to charging batteries.

C ratings. Yes sure, those cheap batteries from hobby king say 5C on them, but its common practice on rc plane application to never ever push 2C. I only ever do 1C unless im in a hurry, and even then i charge them in an emtpy metal biscuit box. Just becasue the sticker says 5C, does not mean it can safely take 5C, and 10C is just ridiculous.

The high C-rate Zippy can be charged 5C without any damage, and probably could take 10C before starting to heat

I understand that your only charging them part way. But even charging them to 4 volts is too much like this. And as high quality as it may be, a pwoer supply designed to run server equipment just dosent have the tolerances or failsafes that are accpetable to charge potnetial;ly explosive batteries in your home.

Please, have some common sense. Id consider this bad practice for a tiny battery, but for a huge pack as you see on e bikes, its just plain idiotic. Those batteries easily have the potential to burn down your house. And once youve spent 600 on an ebike, its not worth teh extra 100 for a decent charger.

There should probbaly be a sticky on safe lipo charging, as a lot fo people here seem to be very misinformed. Im not trying to pick holes, but its just plain dangerous.

edit:

Another thing i forgot to mention is that only charging the cells partially makes them more suseptible to going out fo balance, not less. The packs you have quoted are all fairly cheap low quality packs. And at least for the zippy packs, I know for a fact the cells arent resistance matched. This means, sooner or later, its very likely the weaker cells will go out of balance. Normally thats ok, a balancing charger will fix that no problem. But when a pack has likely arrived unbalanced, and is then only charged partially with a power supply, and never balanced. Its the perfect recipe for a dangerously unbalanced battery. Think about it, the weaker cells will keep disharging first, so lets sya in ideal conditiions you start with a 10 cell pack all at 4 volts exactly out of the factory. After a discharge they could all be at 3.5 apart from one cell thats at 3.3 (a very plausible scenario. If you charge it using the mthod you mention it only takes a few charges before the other cells in the pack start getting higher and higher in voltage, before long your overall apck voltage averages out at a nice 4 volts per cell. But one cell could be gettign charged to 4.3 volts (bad enough on tis own), having being charged at constnat current all the way ( :x ). Please buy a charger, pretty please.

nother edit:

One maybe 2 of these would charge your packs in an acceptable amount of time.

http://www.hobbyking.com/hobbyking/store/__6609__iCharger_1010B_300W_10s_Balance_Charger.html

 

[MadRhino]

@muchalucha

I understand your concern, but I believe that you hadn't read enough here, to know about our use of Lipo.
We know about the danger of the way we use and charge them.

First, I don't spend 600$ on a bike, it is more the like of 7000$.
I have proper Lipo chargers, many of them. I use 4 I-charger 208b to individually balance charge my lipo bricks, occasionally.
The purpose of bulk-charging is to charge fast, I don't really care about how long my Lipo will last.
I can't burn my house with them, for I am charging in an all concrete environment in a garage.

My 100v Lipo packs are all made of 5.8 Ah Zippy cells. They have been bulk charged for over a year, powered my bikes for over 10,000Km, yet never puffed, overheated, or lost capacity. I figure that they have well rendered the expected service already, and if I had to buy another 20 bricks to ride my bikes this year, I'd find it OK

This may seem pushing it a lot, for someone who is not familiar to performance Ebikes, but I assure you that it is quite conservative when compared to what some are used to on this forum. My Lipo packs could very well last another year or more, and that would make them a very cheap mean of powering performance. I don't care if they end in a fire or just plain die losing capacity, riding and charging fast is their only purpose.

 

[ZOMGVTEK]

muchalucha, You are aware a supply like this is CCCV, correct? There is no 'microcontroller'. The cells have an internal resistance, this dictates the voltage drop under load, as well as the voltage rise while charging. The higher the current, the more noticeable this is, but this power supply will be current limited, as well as voltage limited, and it will maintain its voltage as long as the current is under the limit. That generally means CCCV, just like you indicated. These batteries are not going to puff from being charged with a power supply, that has absolutely nothing to do with it.

Regardless, you can charge these batteries with things vastly more dangerous than a decent power supply, and its perfectly fine. I don't understand why you would feel the need to step in on the conversation. RC Lipo could be charged at full current the entire way if you wanted it to charge faster. All this would mean, is manually adjusting the voltage above the HVC to allow the cell to take more current. If charging to 4.10V/cell, this won't even take the cell voltages above 4.20. Some people charge to 4.30 or higher if they want to get everything out of the battery.

Like I mentioned in my previous post, 1C is a perfect fine rate to charge your battery at daily. Don't charge at 5C unless you really want to juice up quick, 2C is about as high as you should go. Also, be aware you can NOT turn the power supply off while the battery is connected. The result will be bad. A big diode across the outputs wouldn't be a bad idea if you plan to use this as a charger alone.

 

[ryan]

Also, be aware you can NOT turn the power supply off while the battery is connected. The result will be bad. A big diode across the outputs wouldn't be a bad idea if you plan to use this as a charger alone.

Can you break this down for me? What happens if the power goes out while I'm charging? PSU is turned off, but battery is near 74.7v. What happens, and what do I do to prevent negative or disastrous results?

 

[ryan]

Let's say you set the charger/PSU to 18A. Your pack is 32Ah. So you're charging at a bit more than 0.5C, which I'm sure will be a healthy enough rate for the cells. It will take a bit more than two hours to fully charge (because charge current will slow down near the end, making it take longer than it would seem just by the straight numbers).

It's been a while since I've calculated charging time, but isn't it somewhere close to 1.5x the total AH divided by the Amps of the charger?

So: (32*1.5)/18 = 2.67 hours? (2hr 40min)
or @10A: (32*1.5)/10 = 4.8 hours? (4hr 48min)

 

[Ypedal]

Well.... :lol:

I do have proper RC chargers ( including a turnigy 4x6 crap unit ) a few Thunder Power 1010's , a EOS 1410, and an Icharger 3010B ( my prefered unit ) ...

however, 99 % of the time i charge straight up using MeanWell supplies.

That being said, i always cycle new packs individually using the RC chargers and stress test them a number of times before assembling into a pack, once cycled and proven " ok " .. they get bulk charged and manually monitored via ballance plugs with gizmo's like battery medics and similar.... I NEVER completely discharge my packs and also bulk charge to an average of 4.17v per cell x N cells.

2 years and going strong with lipo.

3 more years prior doing this with Lifepo4 cells of various brands and types..

no explosions yet.

but i do have 2 fire extinguishers in the house !! :wink:

 

[amberwolf]

Let's say you set the charger/PSU to 18A. Your pack is 32Ah. So you're charging at a bit more than 0.5C, which I'm sure will be a healthy enough rate for the cells. It will take a bit more than two hours to fully charge (because charge current will slow down near the end, making it take longer than it would seem just by the straight numbers).

It's been a while since I've calculated charging time, but isn't it somewhere close to 1.5x the total AH divided by the Amps of the charger?

So: (32*1.5)/18 = 2.67 hours? (2hr 40min)
or @10A: (32*1.5)/10 = 4.8 hours? (4hr 48min)

Unless you are balancing them, too, then a straight bulk charge on RC LiPo and similar chemistries is only going to take just a tiny bit (1-2%?) more energy in than you get out of them, in my limited experiences so far.

ON CrazyBike2, for instance, I am charging backwards thru the CA, so it will run the Ah count backwards down to zero from however much I used up on a ride. Typically it will be at some very low negative number when done. Like say, 0.09Ah when recharging 2.5-3Ah total. So I guess that's 3% more, assuming 3Ah used, or 3.6% assuming 2.5Ah used. AFAICR this also includes the current the CA itself uses, because I'm running the other way thru the shunt, so it's really even less by some small amount.

As for the time it takes, I haven't actually timed it, but presently I am charging using two old small (but heavy) Sorensons only capable of a max of 0.91A at a time, so it takes about 3-4 hours, maybe a bit more, to recharge the typical usage I have from a work commute. That's just with me checking every so often, and fidning it done I shut it off and disconnect it.




Power failures: If power failed while pack is connected to a straight up lab power supply, assuming it is like the Sorensons with no output diode protection, then you would be draining the pack thru the PSU's circuitry. I haven't blown my big Sorenson up yet but I have accidentally turned it off before disconnecting the pack, and it doesn't like it much. Trips the OVP protection and I have to actually unplug it from the wall.

 

[ryan]

Bad news... I brought it home after 2+ hours of travel time to get it here. Replaced the AC power cord. No big deal. Plugged it in. Multimeter read 120v to the rear AC input. Then turned it on (after making sure the current and volts were set to zero). Nothing. No LED light up, no meter moving. Turned up the volts to 10, current to 3. Nothing. Multimeter on the outputs, nothing.

Frustrated.

I suppose this is why non-EE folks like me typically order ready-made chargers from China.

Any help on debugging this?

 

[amberwolf]

If it's reading ok at it's internal meters, it's working, just not setup to output to the normal terminals for some reason. If there's no hookups at the back terminal, nothing will work. IIRC on those types, it's all done from the screw terminals on the back.

Can you take a pic of the terminal strip on the back, so I can see which ones are hooked up where? The pics on the ebay auction are not clear enough and they may not all be the same (at least one has it's terminal strip cover isntalled so I cant see it at all).

Those terminals have to be hooked up to each other (or external controls) in a certain way to have it work as expected. Normally you want it to be hooked up with it's voltage sense wires to the output terminals, and the current sense just hooked up across itself.

There are different terminal strips for different models, but the manual for each one specifies how to hook them up. I'm pretty sure yours is the same series as this:
http://www.programmablepower.com/products/Discontinued/Downloads/DCR%20A%20800%20Watt%20Series%20Rev%20September%201968.pdf
even though it's a newer bigger version that isn't listed in that manual.

So it is likely that yours uses the same hookups listed in Section III. If your terminal strip has 12 terminals, numbered 1 on left and 12 on right, then probably you can make sure they're hooked up like this:

Make sure you disconnect input AC power to the unit first!

Short:
1-2
4-5
6-7

8-9
10-11

Then 8-9 should be one side (positive) of the output voltage, and 10-11 should be the other side (negative).



If yours isn't like this one, we'll have to find the right manual.

 

[ryan]

Thanks AW.

I'll run through the troubleshooting section today.

In the meantime, here are some pics:





Earlier when I said that the multimeter read 120v at the rear AC input, this is what I mean... My handheld multimeter confirmed that the AC input taps (#1 and #3 above), read 120v when plugged in. Hopefully the troubleshooting section of the doc you linked to will help me track down where the power stops.