Diode for Battery Bank Merge Solenoid

[chris4905]

David,

I know your system is 12V and mine is 24V, another option check out:

http://www.wranglernw.com/pc-6420-1008-isolation-relay.aspx

[chris4905]

Dave,

I know my system is 24V, and your's is 12V............so another option check out:

http://www.wranglernw.com/pc-6420-1008-isolation-relay.aspx

[David Anderson]

Kd5kfL,

I perused the link you sent and thanks.  Excellent illustrations.  I will bookmark that one.  I think a diode would be very simple and work well as the link explained.  Now, how are they sized?  What do I ask for at the electronics store?  Radio shack has them, and I can find them there without asking, but give me something to go on to look at.  Are there AC and DC diodes or all the same?   What would be the physical size of the diode?  Excuse my ignorance, but this is new territory for me.

Also, just a comment about Jerry's post about current flow, the alternator will always be at higher voltage than the house bank, ie. 14.2 vs 13.5~.  Therefore, I'm making an assumption the house bank will always be the "downstream" the current flow.  The voltage on the house bank side of the solenoid will always be lower than the alternator side.   Is that a valid assumption?

Thanks,
David

[Jerry Liebler]

David,
    Not if one charges the coach battery with the inverter/charger when plugged in.  I keep my bus plugged in and have  the solenoid wired to close if shore power is present or if no shore but the bus alternator is charging..
Regards
Jerry 4107 1120

[Don4107]

David,

Current also flows the other direction if you find your start batteries low and use the solenoid to boost the start batteries with the house bank.  Even if the start bank is charged and you start the engine with the solenoid engaged it with flow current to the starter because of the heavy draw.  Bottom line is the solenoid will flow current both direction.

Good luck

Don 4107

[Tim Strommen]

Mr. Anderson,

If you think of tanks of water with a hose at the bottom between them - they will equalize at the rate allowed by the "flow" through the hose.  If you correlate the battery banks to the water-tanks - and the tie-together-solenoid as the "hose", you can kind of see where I'm going with this.


If you tie together two banks - the bank with the lower charge will immediately take charge from the bank with the higher charge at the rate allowed by the tie-link (solenoid, fuses, wire) - even with the alternator not adding charge to the overall load.


The alternator's voltage regulator is designed to maintain a voltage.  When the alternator is spinning, and the voltage regulator is "on", you will get a voltage at the set-point (assuming the alternator is spinning fast enough to support the current load of the charging bank{s}...).

The alternator is like a tap filling the tanks (if you want a water analogy for arcing - think of the water hammer effect in plumbing...).


What this means is there is a higher voltage source with the alternator running in conjunction with the battery bank(s).  The surge when the solenoid is connected/disconnected with the alternator outputting power will be higher than if the solenoid was connected/disconnected with the alternator "off".

If anything - this only indicates a probable "best practice" (i.e. not connecting the tie together during high load/charge conditions in the battery bank{s}).

If you intend to tie together the banks for something like a "start assist", I recommend the same thing people recommend when jumping a car: tie together the banks and let the charges equalize a bit before trying to start the prime-mover again.  Once you've started the prime-mover and the alternator is running, wait until the current through the tie-together link is down under the rating of the contactor before opening it (or just leave it hooked up until the charge current is at a "float" level).

Remember that if you remove the load from an alternator quickly, the alternator will try to continue outputting the same current as the voltage regulator is trying to sustain - but will come back down as the voltage regulator changes its contol of the field in the alternator (in the order of milliseconds).  This short transient can end up being 5x the rated voltage of the charging system at the current that was being supplied by the alternator before the event (14.4 * 5 = 72 ... 72volts * 120Amps = 8,640watts even if this is a short transient, it can exceed the rating of the contactor quickly eroding the contact plating).  The previous comment about the air (as a dielectric insulator) is spot on.  Over about 50Volts DC, arcing becomes a problem (and as you can see from the above numbers, this gets exceeded during an alternator load dump event) the air will conduct enough current to jump the gap between the two contact points while the points are close enough to have a low resistance in the air.


If you want to pick a diode for the transient supression role in your charging system - take the 5x system voltage number and the full output current of the charging system and use that as your "rule of thumb".  FYI - the diodes will get pretty big if the alternator puts out a lot of current at a high voltage.

The electric car contactor is a good place to start - but most electric vehicles have either a solid-state controller which does the final connection (so there is little arcing at the master contactor) or a resistive (old-school) pot-box which just limits the current/voltage to the motor (and wastes the rest as heat).  The Tayor-Dunn or Cushman (or any other industrial cart variant) contactors are not really intended to continuously switch the full rated load of a motor drive system (rather as a low leakage cut-out for the electronics, or a failsafe system in the event of a semiconductor punch-through failure).  I'd recommend looking at switches that are desiged for cutting a load to a large UPS battery bank at full rated load called "load break rated" (600VDC @ 800amps or more) if you are using it frequently.  These will have better alloys for open arcs (withstanding high temperatures which would normally vaporize or pit other metals)- and will often have arc fingers that spread the wear out over a larger area (instead of localizing the arcs at the center of the contact points).

Like this one: http://www.boltswitch.com/125.pdf

Cheers!

-Tim

P.S. I wrote this out with a few tall glasses of wine in me - sorry if this is a little wandering  -Tim

[David Anderson]

I've spent two hours learning about diodes and it appears the best fit is a silicone transient suppression diode.  However, I haven't been able to match the available numbers with the appropriate size. 

Using the formula 14.2 x5 =71  My alternator is 350amp peak output  71 x350 =24850 watts.  Based on the formula I need a diode of at least 71 volts and 350 amps.  So far, all the searches haven't led me to the right one.

I'm not sure, but in my reading it seems I could use a small diode with a resistor to snub the voltage to a managable level.  I feel like I've been in an online electrical engineering class for 2 hours.  My head hurts from all the reading.

David

[Jerry Liebler]

David,
    Tim is far to conservative.  All you need is a voltage rating slightly above normal system voltage and a peak current rating above what the alternator can ever deliver.  This is because we are NOT EVER disconnecting the alternator from the start battery.   A suitable device for 12 volt systems is the 'Littelfuse' 15KP17CA which is rated at 17 volts and has a peak current rating of 512 amps.  The best answer for a 24 volt system is to use 2 of the above device in series.  These are from the Digi-Key catalog.
Regards
Jerry 4107 1120

[Tim Strommen]

Thanks Jerry, and sorry about the drawn out explanation David.

Jerry, the part you listed also comes in a 75Volt rating at 15,000Watts - a few of these in parallel right next to the contactor would be a good start (say two or three) as they are only about $9 a piece.  Make sure they are on the alternator-side of the contactor and ground as this is where the transient will be.

The "standard" which the automotive industry plans around for load dump scenarios for 12-Volt systems is about 40Volts (keep in mind this is with alternators that are usually design-limited to a 60-90Amp output).

While the 5x number may be a bit higher than one would think is necessary while the batteries are still connected to the alternator, it is prudent to consider a battery bank that may not be able to absorb that pulse quickly enough (and some of that spike will end up at all of the attached devices).  Some devices like GPS systems, or household LCD TV's with the wall power supply removed (drawing 12-14 volts directly from the battery), are more sesitive to spikes than batteries .  These parts are avilable, and aren't that expensive (granted they cost more than your average 1N4001), and this meathod produces a setup that is likely to not need service ever.

Cheers!

-Tim

[Jerry Liebler]

Tim,
    The objective is/was to protect the merge solenoid.  This can be done with 1 for 12 volt systems or 2 in series for 24 wired across the merge solenoid's contacts.  If one wanted to limit the system voltage during a load dump event, which,I think, you are suggesting, then connecting from the engine side of the contactor to ground would be appropriate.  But why chose such a high limiting voltage?  With this type of device parallel connections should be avoided as, due to tolerances, current will not divide evenly.
Regards
Jerry 4107 1120   

[kd5kfl]

How I do this:

These go on the load side of almost everything; headlights, heater fan, AC clutch. Not so much the horn, I have a finger that performs that function:

http://www.allelectronics.com/cgi-bin/item/D660/205/RECT_DIODE_6AMP_600PIV_.html

RatShack has something close to this, about the size of a pea.

This protects the switches for each individual item. I don't tear apart existing vehicles and install them, but if I'm working on a problem or starting from scratch, I use them.

Big coils and big loads; Starter solenoid, starter, battery banks:

http://www.allelectronics.com/cgi-bin/item/1N1190AR/205/40_AMP_600_VOLT_RECTIFIER_.html

Just get the biggest thing the surplus store has.

Where you need this is at any load which can be disconnected, as close to the switch as possible. If you have multiple battery banks, one for entertainment electronics and one for the coach and one for security electronics, you need one diode per battery bank. Assuming that each bank can be switched to the charging system. One load, one diode, connected to the load side of the switch..

[David Anderson]

Ok, it is getting clearer but not quite there yet. 

Jerry says to connect the rectifier/diode to the engine side of the merge solenoid, and Tim says the alternator side, both I assume mean the same.  Kd5kfl says to connect to the load side of the solenoid which would be the house bank lug.   Which one?


The diode in the link:  http://www.allelectronics.com/cgi-bin/item/1N1190AR/205/40_AMP_600_VOLT_RECTIFIER_.html shows the picture with the arrow pointing to the circle lug which I assume is the cathode.  So, this is wired to the post of the solenoid load (house bank side)  and the screw post is bonded to the bus frame.   Am I describing this correctly? 

Thanks,

David

[Chaz]

Boy...............
  I hope you are getting all this David!        I'm doing - I think - the same thing you are and I am wore out trying to decipher it all. (My brain hasn't been working up to 100% capacity lately anyway.  )
  Hang in there buddy, I'll try and put this all together when YOU finally get it all figured out. I know we have some REALLY GOOD guys here who understand this sort of thing very well.........................................................And I'm THANKFUL!!!!!!!!!!!!!!!!!

Chaz

[Jerry Liebler]

David
    Some how you are missunderstanding what I'm advocating.   The merge solenoid's contacts connect the bus system positive side to the house system's positive side.   I'm advocating connecting a suitable device in parallel with the merge solenoid's contacts, it also will have one end on house positive and the other on bus positive.  The device I'm advocating is a double anode zener transient suppression diode, rated with a voltage rating just above the peak system voltage and a peak current rating greater than the alternator's rated output.
Regards
Jerry 4107 1120

[Tim Strommen]

Wow - I think I may know where all of the confusion came from (i.e. ME).  Some times it's a really bad idea to chime in to a topic when I'm not really reading the thread  - or even putting forth the effort to logically proof the concept I'm attempting to explain before hitting "post".  Allow me to redeme myself while actually adding to the clarity of this topic (wouldn't that be nice...)


But first off, my most sincere appologies to the board... Jerry L here is correct and his instruction is going to saving you a bit of headache.


I realize I have above instructed you to connect the diode between the Alternator side of the contactor to the ground.  This is correct for clamping the colapsing field of the contactor coil - almost... actually, not even really close .

Below is a schematic diagram for the contactor an suggested (and correct this time - I promise) TVS diode installation.

Cheers!

-Tim