Author Topic: Desk Top PS Charge Deep cycle ?  (Read 10234 times)

0 Members and 1 Guest are viewing this topic.

Offline seajinn

  • Newbie
  • *
  • Posts: 16
  • Karma: +1/-0
  • A 2 bit Jinn
    • Hearbal Equine K9 Health care
Desk Top PS Charge Deep cycle ?
« on: June 20, 2013, 08:42:24 am »
I was wonder if any one here knows off the top of the head if it is possible to use 1 or 2 stranded computer power supply maintain a pair of 109ah battery, I want to change out my trailer lighting to low volt led's and do some of the charging with portal wind and solar. Have on hand a few 18 and 24 amp desk top power supply's on hand and want to get the inefficient monster out from under the trailers sink !
Thank for any ideas or thoughts
There's more to the eye then the cat thats nimble

Offline tomw

  • Not as bad as you might think
  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 739
  • Karma: +35/-0
  • hoplophobic people will fear my lifestyle
    • Zubbly's photos!
Re: Desk Top PS Charge Deep cycle ?
« Reply #1 on: June 20, 2013, 09:25:55 am »
seajinn;

Once again, I cannot track what you are getting at?

What is a "1 or 2 stranded computer power supply "?

I know what a computer power supply is but not a 1 or 2 stranded...

Bit of a terminology barrier I think?

Can you be a bit more specific what you mean?

I think a computer power supply may have limitations for keeping a battery at the proper voltage but fairly sure it can do it with some kind of outside regulation or control? A "12 volt" battery is not really kept charged by 12 volts it may need anything up from  about 14 15 volts to do it properly depending on the battery in use.

Probably doable but not just hook it up and go.

Tom
Do NOT mistake me for any kind of "expert".

( ?° ?? ?°)


24 Trina 310 watt modules, SMA SunnyBoy 7.7 KW Grid Tie inverter.

I thought that they were angels, but much to my surprise, We climbed aboard their starship and headed for the skies

Offline oztules

  • Forum Advisors
  • Hero Member
  • ****
  • Posts: 1177
  • Karma: +105/-8
  • Village idiot
Re: Desk Top PS Charge Deep cycle ?
« Reply #2 on: June 20, 2013, 03:17:59 pm »
Yes you can try this for start.

http://www.thebackshed.org/forum/forum_posts.asp?TID=1693&PN=2&TPN=3

Dinges has also posted complete material on doing the tl494 conversions on that site somewhere.

oztules
Flinders Island...... Australia

Offline seajinn

  • Newbie
  • *
  • Posts: 16
  • Karma: +1/-0
  • A 2 bit Jinn
    • Hearbal Equine K9 Health care
Re: Desk Top PS Charge Deep cycle ?
« Reply #3 on: June 21, 2013, 10:44:49 am »
Sorry Tomw  I meant a more basic power supply for the average home or office desk top. A unit around 12 to 20 amp output on the 12 volt side. The one im experimenting seams to hold at 13.4v and climb to 14.5 under load and I know i need 14.8 or better. My math is a real challenge to me these days . My best calculations seam to show if I was to use desk top PS I would need 2 if I am to try top off 2 109 ah marine battery's, Worse is I tossed several larger server power supply's away that one would of been more then enough. I wanted to list them for postage but I had a brief absence on the board and though me trying to long in at the other board got be booted because I was logging in from around the world. Then i got to south central China were there internet gave me shear he__ ware i thought i was found some threat. Just a few months ago realized the board split in two. Aggravating how these days the littlest a thing can be a major over  sight on my behalf, But a few fellow here got me back on track.
 So basically I will be on different locations that all ready are equipped for travail trailers on Horse training centers or farms. My biggest load to the battery's will be in transit keeping the 120volt refrigerator running , Then when set up for the days I run my demo's the load is just a few lights and the water pump. The truck keeps the charge of coarse while in transit. I think I will find 90 percent my stops will allow me grid access, But this is not certain. Starting this venture I can not afford a newer trailer power pack to start , and I hope along the many roads I find a repair shop to get my Marine Excide 2000EX back in service since it handles my required system needs. Or afford a newer up dated power pack for this travail trailer in time.
  I am working on a small wind charger building small and light so it can be portable. I have the advantage i will be set up in wide open air typical of a race tracks so I am building for easy up and quick down and packing. Small prop and self feathered blades I came to admire all over China ! NO KIDDING! Home builds every ware! Its a hobby bigger then kite's ?
 At that point I i might become even more independent on the grid power and the power pack just strictly back up.
  I hate i be a nuisance and come off to frugal but life as disabled has been a life i create even my income . Or is it my Dutch Blood.
  So I am trying to do the best I can and draw from the vast amount of experience collected here.
  I also expect, the moderators might see the membership numbers rise once i begin flying my mill across the country. I don't need a wind vain but it is free advertising and promotion and most people not knowing better will be expecting to see one. See I am selling wind for speed to there horses !  Gotta fly a wind mill ! !
  Hay Oztules. Were near enough neighbors now. Home is Cebu and I will soon be treading your turf and stone tracks to, doing the same nonsense in your back yard ! One tough country to get a  Business Visa in , Ay ? @ least I got a good connection in the race Commission.  One of your heads Runs the State of Pennsylvania racing here to.
  The link is quite handy ,
Thanks folks
There's more to the eye then the cat thats nimble

Offline MadScientist267

  • Impossible Condition Curator
  • Hero Member
  • *****
  • Posts: 1514
  • Karma: +44/-4
  • Rules? What rules?
Re: Desk Top PS Charge Deep cycle ?
« Reply #4 on: June 22, 2013, 11:00:25 am »
 In short, yes, a PC power supply can be converted into a battery charger, but be prepared to make a whole bunch of modifications to do it properly. A PC supply has a series of protection devices built in that need to be bypassed and/or dealt with to prevent crowbar, and current limiting needs to be addressed.

A PC power supply does not generally monitor the +12V rail for voltage regulation purposes, as the 12V is primarily for driving the various motors found in the PC, which are more tolerant of voltage variances, although too high or too low may trigger the crowbar and shut the supply down.

The primary voltage  regulation takes place on the 5V rail, as this is the most critical. The +12 essentially just "follows" the 5V regulator (eg, an increase in load on the 5V rail that causes the voltage there to dip will result in the regulator pushing harder on the 5V, and the 12V rail typically will rise slightly as a result).

To get past all of this, you have to eliminate (or heavily modify) the crowbar circuitry, and change the regulation feedback source from the 5V rail to the 12V rail, then make it variable so you can adjust it to the proper voltage for charging the battery.

Then, because you've essentially eliminated the protection system (crowbar), you need to devise a current limiter so that you don't exceed the supply's capabilities during the heavy load times (eg, bulk charging).

And, then as if that weren't enough, there really should be a temperature compensation mechanism that adjusts the voltage regulation based on the core battery temperature. This involves relatively precision voltage correction and can get tricky in a situation like that.

The other thing is, to do it right, you should have a mode switch mechanism in place to switch to a lower voltage (for float) at the end of the absorption stage.

So, can it be done? Sure.

Is it simple? No.

Are there much better ways? Absolutely.

But if you're looking for a serious challenge and have plenty of time, spare parts and/or money to throw at it, as well as the patience and skill to play with it until it works, by all means, go for it.

Personally, I've been playing with electronics for approaching 30 years now, and wouldn't mess with it myself, but that's just me. YMMV. ;)

Steve
Wanted: Schrödinger's cat, dead and alive.

Offline oztules

  • Forum Advisors
  • Hero Member
  • ****
  • Posts: 1177
  • Karma: +105/-8
  • Village idiot
Re: Desk Top PS Charge Deep cycle ?
« Reply #5 on: June 22, 2013, 03:03:00 pm »
Actually they are really very simple to change over... but you must first try to get TL494 AX supplies to work on... if not then ATX will do but 494 systems seem the easiest to understand.

The 12v and 5v rails are from the same single winding. The 5v is just a tap on the 12v winding, so control one, you control both... they do have different "in hand" strings... ie the 0-5v may be 4 in hand then tap .... then the 5-12v may be only 2 in hand... but the same winding none the less... hence different current for 12v and 5v circuits.

You get rid of everything on the LV side of the 494 except the timing cap ( pin5) and the timing resistor (pin 6) and get rid of  anything to do with the dead time control ( pin 4)

Strip the board of everything else on the LV side, but for the totem pole transistors for driving the driver transformer, and any support resistors and diodes.... ie the driving side is left intact.

At this point you need only provide a voltage divider from the 5 or 12v rectifier for your voltage level ( I use pins 1 and 2 for this), and another divider for the current (I use pins 15 and 16),  some feedback to pin3 to rate the gain..

Thats about 10 resistors and maybe a cap or two to reuse the soft start functions.

For this  we end up with the ultimate battery charger. It can be used to dial up between 0 and 18A, can run from 5-28v or so.
It can drive dead loads ( DC motors under heavy start up loads )

It will be the cheapest most useful device you will ever make. It is a laboratory supply with complete short and current limit  protection, .01v control.

If you experiment with power electronics, you really really  NEED this so that you can screw anything you like up, and the current protection will protect your project from shorts or anything else, and provide variable voltages to test the unit under different conditions......

"Are there much better ways? Absolutely."..... I have never seen them. No other charger I have seen can do what a PC supply can be made to do.

It requires no money, no outlay, all the bits you need are there... but you will need to study the tl494 in detail and understand how it works to be confident of results.

"
Personally, I've been playing with electronics for approaching 30 years now, and wouldn't mess with it myself, but that's just me. YMMV. ;)"... geez Steve, you really need to do one before you depart the planet,... just to see what your missing out on.

 I must have done 30 or more for different application for different people... takes an hour or so once you get the hang of it.... ATX needs a little more understanding to get the feel for the AUX supply built into it. Put a current meter on the front, with voltage and current control knobs, and this thing is indispensable on a electronics work bench.

They are a good bed ready for exploitation. If you want different voltages like 50 or 100, then wind another transformer. Getting the original ones apart requires heat from a gun or boiling water.... easier to get another bigger core, you provide the dividers, protect the 494 from over 34v, and you can have anything you want.

...oztules

Flinders Island...... Australia

Offline MadScientist267

  • Impossible Condition Curator
  • Hero Member
  • *****
  • Posts: 1514
  • Karma: +44/-4
  • Rules? What rules?
Re: Desk Top PS Charge Deep cycle ?
« Reply #6 on: June 22, 2013, 07:37:07 pm »
 That's just it, I have twisted PC power supplies into all kinds of odd configurations...

I never said I *haven't* done anything like it, I said I wouldn't bother for this.

Is everything you need basically there? Yeah, of course it is. Its a switching power supply, that's what it does.

But I wouldn't call it a novice project. Not that I'm calling anyone a novice, just that its a lot more involved than you make it sound.

It all depends on the supply's design out of the box. A 494 based unit (typical) most certainly can be made into a very suitable bench power supply, and yes, I've made them.

I would NOT however just hand this off as something as trivial to do, it simply isn't. Particularly as a full time, unattended automatic battery charger.

It's easy for those of us who have bent them into other usage purposes to leave out all of the little details that need attention to do something like what the OP is after.

 There's a big difference between a bench power supply and a *proper* battery charger.  I've already covered everything you mentioned,  just didn't get as in detail about it. What, you think I make this stuff up? How else would I know what's involved?

Lets keep this real, shall we?

Steve
Wanted: Schrödinger's cat, dead and alive.

Offline oztules

  • Forum Advisors
  • Hero Member
  • ****
  • Posts: 1177
  • Karma: +105/-8
  • Village idiot
Re: Desk Top PS Charge Deep cycle ?
« Reply #7 on: June 22, 2013, 08:52:15 pm »
Steve, I am not going to comment further on this, other than.
I tire of hearing how hard it is to work on PC supplies to make them useful devices. I don't know why folks do that, but I assume they have tried to incorporate the original control circuit into what they want it to do ... this is a losers game.

Look at this and tell me how hard it is to adapt to a psu if you CUT out all the original crap but for the driver.... thats is not difficult is it?
2395-0
Now  the difference is ignore pins 8,9,10,11, as that is done for you already, ignore pin 13, as it will be currently run to pin14 for push pull.

Now look at it, a handful of resistors, and a cap or two... thats it.... oh and the current sense need only be lifting the center tap out of the board, placing a few inches of thinner wire and that will be the  current load resistor..

I suspect you spent hours trying to make the original circuit do what you wanted, and thats not easy
The OP may not be familiar with a soldering iron, but plenty of folks can do this simple adaption and have a better device than you are likely to have developed from the original circuitry.

Don't let your difficulties in modification be a blind to others who may be put off doing what is really simple stuff.. I can't see the hard part. If you can solder, and read a circuit, your home..... and the full Texas Instrument PDF is here: http://www.ti.com/lit/an/slva001e/slva001e.pdf.. page 24 go and look at it and tell me how hard it is to do.

If they take the time to read the pdf, it will make perfect sense....

I am no electronic engineer, but  I managed it easily, and shown others how to do it.... just tear off all the junk from the board that does not drive the driver, leave the rectifier and some filter if you want, but don't try to tell me how hard it is to control the PWM, coz it is not if you read the TI pdf.

To not blow transistors you need voltage AND current control... then it is virtually indestructible... the original circuits don't have current control, but rather scr shutdown on voltage faults.... ripe for destruction when you pull the scr out and try to drive low impedance loads....

Think as you will, but I know it is easy... coz I did it... so it must be if the village idiot can do it.

That is all I'll say on it. Others can judge their skills and move accordingly.... but it is not scary difficult.


..............oztules
Flinders Island...... Australia

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: Desk Top PS Charge Deep cycle ?
« Reply #8 on: June 22, 2013, 09:40:48 pm »

That is all I'll say on it. Others can judge their skills and move accordingly.... but it is not scary difficult.

The only thing I will add to this conversation - and it's directed at those who may not have as much experience as Oz and many others on the forum...

Playing with a complete PC SMPS and running it from 120/240V AC, with the original capacitors in place can and will develop enough energy to be LETHAL to the unwary.

Running it off 12, 24 or 36V I'd have no hesitation. But if someone takes this as "any idiot can grab a fully functioning PC power supply and do this" - just be very wary of the high voltage side because it ISN'T for the village idiot :)

Offline oztules

  • Forum Advisors
  • Hero Member
  • ****
  • Posts: 1177
  • Karma: +105/-8
  • Village idiot
Re: Desk Top PS Charge Deep cycle ?
« Reply #9 on: June 22, 2013, 10:05:45 pm »
Very fair comment Ross.

 Any modern appliance falls into this category now days, they all use PWM.... transformers have gone to heaven now.

So Ross's beware is very relevant for all mains appliances in todays world. HV caps of 2-400 uf are killers... and so is the 240v mains.

If your not familiar with working with mains equipment this is not for you..... even if it is easy.

Also, using this sort of thing can get you  up to 28vdc..... this means the original 16v caps WILL explode if your going to drive it beyond about 15v... needs higher voltage caps in that case too.

They are very good at emulating solar panels as you can lower the current until pretending to be dark, and see how your test circuit handles the transition...doing that right now.


..............oztules
Flinders Island...... Australia

Offline kurt

  • Senior Moderator
  • Jr. Member
  • *******
  • Posts: 37
  • Karma: +6/-0
    • my ham radio blog
Re: Desk Top PS Charge Deep cycle ?
« Reply #10 on: June 23, 2013, 01:27:57 pm »
it is clear to me here that oztules and MadScientist267 are talking two different things.  seajinn asked for a battery charger to make a battery charger that does not require constant babysitting you need allot more than a constant current/voltage power supply unless you just talking a float charger.
Climate is what you expect. Weather is what you get  Mark Twain

my small engine page