Forum Discussion

jrnymn7's avatar
jrnymn7
Explorer
Oct 05, 2014

Paralleling Power Supply Units Made Easy ???

I am considering paralleling a couple of 20a variable adjustable psu's for 40a charging. According to what I have read on the world wide inter-web, it can be done quite easily, provided the following conditions are met:

... units must be the same brand and same internal parts, (resistors, etc.), be current limiting, and have c.v. as well as c.c settings.

The idea is to parallel them by first connecting the + and - leads from both psu's together, to a single + and - lead, which is, in turn, connected to the load/batts. The belief is the voltage drop over the leads will compensate for any minor fluctuations in voltage output of the two psu's. This will, allegedly eliminate the need for a more sophisticated arrangement, including diodes and such.

In due course, one must also use one of the psu's as the voltage controller, by setting it at say .01v higher than the other unit, and putting it in c.v. mode, while using the second unit as the amperage controller, by cranking it to max (limited) amps, and placing it in c.c. mode. (both units set at max amps)

As I'm sure some of you know, the idea is to cause the two units to properly share the load, by causing the unit in c.v. mode to lower its voltage to match the other's lower voltage setting.

My understanding is the unit in c.v. mode will take over on it's own, once charge current tapers to =/< the max (limited) amp setting of the psu's.

The units I'm considering paralleling are the gophert cps1620.

??? Any thoughts on the feasibility of such a simple set up?

(Thinking back, this seems to be essentially what BFL does with his various chargers.)

Thanks!!

160 Replies

  • MEXICOWANDERER wrote:
    Check with the designer first - some can be paralleled others no.


    good idea, thanks mex.
  • Thanks BFL. I was hoping you'd add your input.

    I think the "big deal" has to do with overworking one of the psu's, compared to the other(s), and because of excessive heat, etc. it's life span could be greatly decreased compared to the others? That seemed to be the main concern in the articles and posts I read on the subject.

    I mean, some of the extravagant efforts gone to, to "perfect" paralleling, are way beyond my level of understanding. But yes, it seems that absolute precision is not that necessary, thus the method of simply connecting the leads as stated above would suffice, in the event of subtle variations in output voltages during the charging process.

    The other issue, apparently, is input currents perhaps fighting with each other, and apparently pfc can add to the problem? Don't know. I just wanted to confirm the set-up is at least feasible, if not the grand daddy of perfection. And from the responses so far, it appears it is.

    Of course, I would also like to spare the psu's (and my wallet) of any undo stress, if at all possible.

    thanks!
  • "When in voltage mode, power supplies do not share current well. The higher voltage supply will back bias the other supply. Depending on the cable resistance, you may get something from the lower voltage supply."

    Yes, this is the inherent issue with trying to parallel psu's, and this is why it was recommended to use one in c.v. and the other in c.c.
  • "Or, mess with the innards and get an automatic alternative to rubbing two sticks together."

    Thanks! I'll try to remember that if I'm ever stranded and in need of making a fire! :)

    As for #2 taking over, I think with the set-up I described, both psu's share the load until amps taper to at or below the 20a setting, so well into the 40a charge; not when reaching the preset volt setting. Perhaps it has to do with the psu's being c.v., c.c., and current limiting?
  • I don't understand the OP why it has to be so complicated. I use multiple chargers all the time and the chargers are not exactly the same for voltage.

    It does work best when all the chargers have the same voltage but "perfection is the enemy of good enough."

    They all add their full amps until the battery voltage rises enough that the lowest voltage one doesn't have enough "spread" between its voltage and battery voltage to do its rated max current anymore, so that one's amps start to taper. But it still adds those tapering amps until battery voltage reaches that charger's voltage.

    If the chargers are all set to voltages near where you want Vabs to be, then the main bulk charge is mostly with all doing near full amps.
    Good enough. Amps will taper when Vabs is reached anyway, so who cares about a few minutes in there when you get fewer amps from some chargers before reaching Vabs? Big deal.

    You wouldn't try to run a 13.6 with a 14.4 in hopes they will add their amps more than a minute, but you can run a 14.4 with a 14.6 and a 14.8 just fine.

    If all the chargers have adjustable set points, then just twiddle to get them near the same voltage and that will get it done.

    Once the battery acceptance rate has fallen so that you have too many chargers on the job, you can leave them all on and they will each do some of the amps or you can yank a charger and put it away if you want when the total max amps of the remaining chargers is enough to meet that falling acceptance rate.
  • The two supplies will share current when they are in current limit. That means setting both supplies to 15.5V or what ever when battery is discharged. You will get 40A.

    If you set the supplies to 14.4V and when the battery reaches that voltage the supplies will switch from current mode to voltage mode. When in voltage mode, power supplies do not share current well. The higher voltage supply will back bias the other supply. Depending on the cable resistance, you may get something from the lower voltage supply.

    I think it's a lot easier to get a 40A supply than to constantly baby sit the voltage adjust to get them to share.
  • I agree with this. With a depleted battery bank set one psu at 14.2 and the other at 14.4. Both will chug away nicely as the bank voltage S-L-O-W-L-Y rises to 14.2, then #1 Sun will drop off, and #2 will continue and begin to taper amps as it approaches 14.4.

    Set both at 14.4 and I think you'll just get tapered amps, only more steeply tapered.

    Or, mess with the innards and get an automatic alternative to rubbing two sticks together.
  • While I've never done what your proposing I have built many many systems involving charging with solar. In these systems there are often multiple charge controllers providing charging to the same battery bank. They work well in parallel without any "tricks" in the programming. All are programmed to the same battery charging voltage and current limits appropriate for the battery bank. All do their job accordingly.
    Power supplies may act differently but I really don't think it's even as difficult as you describe.