How do I use every last electron?

Discussion about the FM100, FM80, and FM60 Charge Controllers

Moderators: OutBack, OutBack Moderator Team

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Tue Aug 05, 2008 7:08 am

We're still working on this diversion project... 8-[ [-(

Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Sat Aug 09, 2008 6:49 am

I installed a software update into my FM80 yesterday. I'll be conducting more tests this weekend.

More later,
Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Sun Aug 10, 2008 11:02 am

Initial results from yesterday's tests are quite encouraging. :grin: I need to discuss the results with my engineering colleagues, and I'll then post a more detailed report.

Regards,
Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Tue Aug 12, 2008 5:47 am

Note: I moved this thread from the MX60 forum to the FM80 forum because the "AUX Diversion SSR" feature discussed below applies to the FM60 and FM80. The MX60 controller does not include this new feature. -- Jim / crewzer

OK After a difficult start, Ive got some really good news to report. \:D/ But, let me first thank my esteemed Outback colleague, Darren, for his interest, dedication, and hard work in refining The FLEXmax charge controllers (FM CC) unique AUX Diversion Solid State Relay (SSR) feature. =D>

ItÔÇÖs generally well understood that charge controllers leave power ÔÇ£on the tableÔÇØ when operating in the current-limiting mode of the absorption and float stages. Specifically, the controllers only supply enough charge current to maintain target voltage and sustain any loads. The original question (challenge?) was to instead find a way to put ÔÇ£every last electronÔÇØ to use. For example, would there be a way to use this available but essentially unregulated power to heat water and reduce oneÔÇÖs propane consumption?

The new OutBack FLEXmax charge controllers (both the FM80 and the just introduced FM60) include a new ÔÇ£Diversion SSRÔÇØ feature in the AUX output menu. I believe the original intent of this feature was to allow the FM CC to be used as a PWM diversion-type controller with an external power source, a battery bank, and a dump load.

However, it soon became apparent that another application would be to use the ÔÇ£Diversion SSRÔÇØ feature to control a diversion-type dump load in parallel with the controller's battery bank. The idea is elegantly simple: Use a PWM-controlled switch (the SSR) to rapidly connect and disconnect a load across the battery bank to use up some / most / all of the residual PV power otherwise left behind when the controller is operating in a current-limit (absorb or float) mode without significantly perturbing the bankÔÇÖs target voltage.

IÔÇÖve been keenly interested in this new application for a while, so I gathered up the parts to test it out. Here are the test configurations I used:

A) A series circuit consisting of a DC circuit breaker, the SSR, a shunt, and a 2 Ohm @ 1,200 W resistor array (4 x 0.5 Ohm @ 300 W wired in series). This circuit is wired across the 24 V nominal battery bank terminals, and the control wires for the SSR are connected to the FM80ÔÇÖs 12 V AUX connector. Relative V and Hys V were both set at 0 V, Hold time was set at 7.5 secs, and Delay was set at 5 secs.

B) A series circuit consisting of a DC circuit breaker, the SSR, a shunt, and a 1 Ohm @ 600 W resistor array (2 x 0.5 Ohm @ 300 W wired in series). This circuit is wired across the 24 V nominal battery bank terminals, and the control wires for the SSR are connected to the FM80ÔÇÖs 12 V AUX connector. Relative V and Hys V were both set at 0 V, Hold time was set at 7.5 secs, and Delay was set at 5 secs.

IÔÇÖll save you the results of my initial tests. LetÔÇÖs just say that surface charge builds/decays differently in AGM batteries than it does in their flooded-cell cousins. Darren reviewed my reports and we discussed the behavior, and he figured out and tested a software rev. I installed the new software in my FM80 late last week, and I tested the updated Diversion SSR feature this past Saturday.

So far, so good: It works! :cool:

My initial test used Configuration ÔÇ£AÔÇØ above. However, the combination of the power required (~250 W) to maintain the absorption target voltage (28.7 V) plus the diversion load (~412 W) was too small (diversion load resistance too high) to fully load the PV array, so the AUX PWM % indicator just sat at 100%.

My next test used Configuration B above. At 28.7 V, this 1 Ohm resistor could dissipate 824 W, which the array canÔÇÖt supply this time of year (array too hot), so this turned out to be a good regulation test for the controller.

With the AUX Diversion SSR feature OFF, my array was generating an indicated ~250 W to maintain the battery bank at its 28.7 V absorption target voltage. With the AUX feature turned ON, the indicated power drifted between 590 W and 730 W initially, and then rose and varied between 630 W and 760 W. The PWM % indicator fluctuated between ~45% and ~76%, and indicated battery voltage slowly drifted between 28.7 V (absorption target) and 28.6 V.

And, the resistors got hot! I estimate they were dissipating ~450 W ((630 W + 730 W)/2) ÔÇô 245 W). Over an hour, 450 Wh would equal 1,536 BTUÔÇÖs. Assuming 90% system efficiency, three gallons of water (25 lbs) could be heated from 55 F to 110 F.

Cool! :cool: :cool:

And, the controller was able to handle my ÔÇ£dynamic environment testÔÇØ. When I turned off ┬¢ of my array, the PWM % dropped to the 7% to 35% range, and the controller continued to maintain absorption target voltage. When I restored the ┬¢ array, PWM % increased back to the original range.

OK time for a break. But, we now need to consider how to size and spec this type of diversion load for individual applications. Ive got the basics sorted out, but I need to work on a clear, useful, and safe discussion. Related questions, comments and suggestions are welcome!

More later,
Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

kitestrings
Forum Member
Posts: 7
Joined: Tue Apr 01, 2008 12:45 pm
My RE system: Off-grid since 1984: Sencenbaugh 1.2 kW wind turbine on 100' Rohn 25G, 16- 180W Mitsubishi panels, OB VFX3648 w/mate, MX-60, PS2DC175, 16-L16HC's, Onan 4.5 LPG gen, Solar Converter 12/48-30 equalizer, Rutland 913 "functional" weathervane.
Location: 'Kingdom', VT

Re: How do I use every last electron?

Post by kitestrings » Wed Aug 13, 2008 2:36 pm

Crewzer (& Darren),

First off, thanks for taking the time to delve into and report your progress on this issue. IÔÇÖm sure that many will benefit from your efforts. It sounds like you are well on the way to having answered the initial question posed. This is great stuff.

We have been steadily moving toward the same end game for some time now. I recently installed a new pre-heat DHW tank; having replaced the stock elements with low voltage (48) DC elements. I still have to snake a couple wires to have everything wired properly.

On a typical good sun day weÔÇÖre absorbing by late morning to mid-day, and two hours later, depending of course on loads, happily floating the batteries. I was anxious to get a started, so my KISS interim solution was to hot-wire the tank (via a breaker) and wire the relay to a simple spring-wound timer. With a bit of weather watching weÔÇÖve been able to provide most of our DHW needs in addition to keeping the batteries fully charged. And, weÔÇÖve logged some of the highest kWh/ day rates recorded (with less than ideal weather).

The elements are wired in the standard, 2-element, interlocked configuration. The specÔÇÖs are for 30A (48V) with the sub-elements in series (1.93 ohms), so I was expecting approximately 1,600 watts. They are actually drawing closer to 22-23A, or about 1,200 watts. Still, with a delta T of (170-55=) 115 degrees F, this ÔÇÿthermal batteryÔÇÖ as I like to think of it can store 28,000 BtuÔÇÖs, or about 8 kWh. A tempering valve keeps the outlet (or inlet to the gas unit) in a safe range.

Last night while my ÔÇÿhome-from-collegeÔÇÖ son took his ÔÇÿIÔÇÖm-not-paying-for-propane-showerÔÇÖ [-X I watched the inlet water temperature to the Takagi holding steady at ~100-105 degree F. Now IÔÇÖm firmly in the ÔÇÿDadÔÇÖs-not-paying-for-propane-eitherÔÇÖ mode \:D/ . I see no reason why this won't work very nicely through the better sun months anyway.

HereÔÇÖs my basic question:

The firmware of the FM80 & now FM60 appears to have some nifty new features, but is there no hope for the lowly masses still using the tried-and-true FX60ÔÇÖs?

WeÔÇÖre had seriously been considering investing in a second unit in lieu putting up the money for the extended warranty, since without the CC youÔÇÖre basically system down even with timely turn-around(s). My hope, however, has been to find workable settings, using the Aux-Diversion mode, with the FX60 to semi-automate the capture of this otherwise lost resource.

I look forward to hearing more.

Kindly, ~kitestrings

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Thu Aug 14, 2008 6:01 am

The firmware of the FM80 & now FM60 appears to have some nifty new features, but is there no hope for the lowly masses still using the tried-and-true FX60ÔÇÖs?
Kitestrings,

The MX60ÔÇÖs manual suggests there is a way (to control a water heater via an SSR), but I havenÔÇÖt tried it (so may buttons to push, so little time... :cry: ). Specifically, there are four references to ÔÇ£solid state relayÔÇØ applications in the manual (pages 22 and 29). Or, download the manual and do a search on ÔÇ£solidÔÇØ.

Link: http://www.outbackpower.com/pdfs/manual ... v_mppt.pdf

Thanks for the kudos!

HTH, and more later...
Jim / crewzer

P.S.
ÔÇÿhome-from-collegeÔÇÖ son took his ÔÇÿIÔÇÖm-not-paying-for-propane-showerÔÇÖ
I have two of those!! :wink:
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Mon Aug 18, 2008 8:35 am

DRAFT Application Note:

FLEXmax Charge Controller AUX Diversion Solid State Relay (SSR) Feature ÔÇô ÔÇ£Opportunity LoadÔÇØ Control

OK. Time to wrap this up. \:D/ Comments, suggestions, and constructive criticisms are always welcome.

Background

ItÔÇÖs generally well understood that solar charge controllers leave power ÔÇ£on the tableÔÇØ when operating in the current-limiting mode of the absorption and float stages. Specifically, the controllers only supply enough charge current to maintain target voltage and sustain any loads. ÔÇ£ExcessÔÇØ power is ÔÇ£left on the tableÔÇØ when the controllerÔÇÖs input switches are closed, which limits current flow from the array. This condition manifests itself has a relatively high array voltage and a correspondingly low array current.

The original question (challenge?) was to instead find a way to put ÔÇ£every last electronÔÇØ to use. For example, would there be a way to use this available but essentially unregulated power to heat water and reduce oneÔÇÖs propane consumption?

Past solutions have included connecting fixed-power ÔÇ£opportunity loadsÔÇØ (i.e., a water heater or a water pump) that were irregularly powered when controller was in Float mode and a useful amount of solar power was available from the array. Once the array could no longer sustain the battery bank and the loads, the battery voltage would fall and the loads would be disconnected. The FM60 and FM80 controllersÔÇÖ AUX connection includes a Float setting just for this purpose, as does the legacy MX60.

But, what about the ÔÇ£excessÔÇØ power -- often considerable in the early afternoon -- available when the controller is operating in Absorb mode?

The new OutBack FLEXmax charge controllers (both the FM80 and the just introduced FM60) include a new ÔÇ£Diversion SSRÔÇØ feature in the AUX output menu. I believe the original intent of this feature was to allow the FM CC to be used as a PWM diversion-type controller with an external power source, a battery bank, and a dump load.

However, it soon became apparent that another application would be to use the ÔÇ£Diversion SSRÔÇØ feature to control a diversion-type dump load in parallel with the controller's battery bank while operating from a PV array. The idea is elegantly simple: Use a PWM-controlled switch (the SSR) to rapidly connect and disconnect a load across the battery bank to use up some / most / all of the residual PV power otherwise left behind when the controller is operating in a current-limit (absorb or float) mode without significantly perturbing the bankÔÇÖs target voltage.

Load Classification: Diversion Load or Opportunity Load?

Applying the FM CCÔÇÖs AUX Diversion SSR feature requires a bit of new thinking (for me, anyway), as traditional load diversion style battery charging and variable duty-cycle ÔÇ£opportunity loadÔÇØ control are two fundamentally different applications.

Diversion-load type battery charging is a well known technique. Those interested in learning more about this charging method will find Section 6.0 in the Morningstar TriStar controller manual to informative reading. One important application consideration is that the diversion load must be carefully specified, and backup regulation is required.

See: http://www.morningstarcorp.com/en/suppo ... .03.EN.pdf

The opportunity load application does not use a ÔÇ£traditionalÔÇØ diversion load to prevent overcharging the batteries. The controller still does its regular job of regulating battery voltage by managing charge current. Instead, the controller connects the opportunity load to the battery bank and sustains it if it can by increasing array current. If operational parameters arenÔÇÖt met, then the load is disconnected.

Accordingly, itÔÇÖs my view that the ÔÇ£AUX Diversion SSRÔÇØ function is actually an ÔÇ£AUX PWM Opportunity Load ControllerÔÇØ for the purpose of this application.

Load Specifications

Load type appears to be limited. Resistive loads such as water heater elements should be fairly straightforward applications. Simple motorized loads (such as large DC fans, perhaps for attic ventilation) should work as well, with the controller varying fan speed (and therefore load) as conditions warrant. The DC motor would probably "sing" when operated at less than 100% PWM duty cycle. The load must operate at the nominal battery voltage.

Load sizing appears to be fairly flexible from a technical perspective. A relatively small load may well operate as a full-time load (100% PWM) much of the time. A large load would operate at a low PWM%. However, a small load would likely not use ÔÇ£every last electronÔÇØ, and an excessively large load might hit the batteries too hard when switched on and causing the battery voltage to dip briefly (the coup de fouet, or "crack of the whip"), thereby creating a regulation problem.

The ÔÇ£idealÔÇØ load is probably one that would cause the controller to operate the AUXÔÇÖ Diversion SSR output at just under a 100% PWM duty cycle, even under favorable environmental conditions. Each individual system configuration will therefore determine ÔÇ£idealÔÇØ load size.

For example, my PV array is rated at 966 W STC, and the battery bank is 24 V nominal. Assuming very good conditions (high insolation, cold ambient temperature, little/no load on batteries, etc.) a load rated at 966 W might be ÔÇ£idealÔÇØ. At a maximum absorb target voltage of 29.5 V (cold batteries), the loadÔÇÖs resistance value would be (29.5 V x 29.5 V) / 966 W = 0.9 Ohm.

I used a 1 Ohm resistor for my test. The slightly higher resistance means that it's a slightly smaller load. Ideally, the resistorÔÇÖs power rating should be ~870 Watts (typo correction; thx, ks). Thus, the resistorÔÇÖs power spec of 600 W, while OK for the test in the summer (reduced PV array output, normal battery voltage, other loads on the system), is probably too low for a permanent installation.

Test Circuit Reference
B) A series circuit consisting of a DC circuit breaker, the SSR, a shunt, and a 1 Ohm @ 600 W resistor array (2 x 0.5 Ohm @ 300 W wired in series). This circuit is wired across the 24 V nominal battery bank terminals, and the control wires for the SSR are connected to the FM80ÔÇÖs 12 V AUX connector. Relative V and Hys V were both set at 0 V, Hold time was set at 7.5 secs, and Delay was set at 5 secs.
So, whatÔÇÖll happen if the opportunity load is other than ÔÇ£idealÔÇØ? LetÔÇÖs consider the 24 V / 48 V water heater element available from http://www.solarseller.com as an example.

See: http://www.solarseller.com/diversion_lo ... _loads.htm

For my 24 V system, I could configure the heater element in parallel as a 0.62 Ohm load capable of dissipating 1,265 W at 28 V. I believe the power handling spec is OK for my array, but the load is too large (resistance too low).

That shouldnÔÇÖt be a problem. The FM CC AUX output would just reduce the PWM duty cycle to the opportunity load. But, IÔÇÖd be concerned the controller keeping up with very quick changes in my AGM battery bank voltage as the large load in connected/disconnected.

Alternately, I could configure the element (in series) as a 1.24 Ohm load rated at 632 W. This too should be OK, as the controller isnÔÇÖt going to supply more power to the from the array than OhmÔÇÖs Law says the load will dissipate (28 V x 28 V / 1.24 Ohms = 632 W). I suspect that the AUX output would spend a lot of time operating at a 65% to 85% duty cycle. For my system, this may well be an ÔÇ£idealÔÇØ load as far as readily available options are concerned.

This same water heater element can be used for 48 V system applications. Operating at ÔÇ£full tilt boogieÔÇØ, itÔÇÖll dissipate 1,265 W. Over an hour, thatÔÇÖll heat 60 lbs of water (~7 gallons) from ~55 F to ~120 F.

Solid State Relay (SSR) Specs

The SSR much be selected to handle the maximum DC voltage, the highest possible load current, and it must be heat sunk.

Returning to my test setup, highest possible voltage is 29.5 V, and highest load current is 29.5 V / 1 Ohm = 29.5 A. I used a DC-type SSR rated for 60 V and 75 A mounted on a large heat sink. SSRÔÇÖs can be paralleled to handle very high current loads.

See: http://www.power-io.com/products/hdd.htm

Safety

The opportunity load needs its own circuit breaker or fuse between the battery (+) and the SSR and load. I should have used a 30 A continuous duty DC breaker for my test, but I used a 40 A model, because that was readily available from my super well-organized parts bin. :wink: :roll: Circuit wire needs to be correctly spec'd for conditions of use.

A water heater opportunity application should also include a DC thermostat mounted on the element or separately mounted to the heater tank wall. If the water temperature should rise enough to open the thermostat, the load will be disconnected from the battery and the controller will reduce array current to compensate for the reduced load

See: http://www.survivalunlimited.com/diversionloads.htm

I suppose it might be possible for some sort of failure to cause the SSR to fail (ON or ÔÇ£closedÔÇØ) and seriously load the battery bank. A low-voltage disconnect (another SSR or just a regular olÔÇÖ DC relay) operated by another AUX output in the system might be the fail-safe solution to this unlikely predicament.

AUX Diversion SSR Settings

This function includes settings to allow operation within certain voltage and time parameters. I used the following settings for the test of my system with AGM batteries:

ÔÇó Relative Volts: 0 V
ÔÇó Hysteresis: 0 V (these two values were selected in an effort to keep the battery voltage at the target charge voltages)
ÔÇó Hold: 7.5 seconds (the battery voltage would have to drop below target for 7.5 seconds before the controller would turn off the AUX Diversion SSR function)
ÔÇó Delay: 5 seconds (the battery voltage would have to meet or exceed target for 5 seconds before the controller would turn on the AUX Diversion SSR function)

These settings worked for my initial tests; I expect other settings would work as well. I also expect that many systems would require individually unique settings.

Conclusion

I like the idea of finding a way to use ÔÇ£every last electronÔÇØ, or something close to it. I also like the new AUX Diversion SSR feature, as it appears to do the job in a satisfyingly elegant technical manner. IÔÇÖve found it to work with my AGM battery bank, and Darren and his posse have tested the application with flooded-cell battery banks and immersed water heater elements.

Again, IÔÇÖd like to thank Darren and the rest of the Engineering Team for developing and offering this feature as we continue to evolve our line of MPPT charge controllers to provide a broader range of application solutions to our customers. =D>

I hope this report is useful, and I have to admit that it may not be as clear as it might be. Comments, suggestions, and constructive criticisms are welcome, but I now need to set this aside for a while and tend to other priorities.

Until later, caveat emptor!

Regards,
Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

solarstudent
Forum Guru
Posts: 136
Joined: Thu Feb 23, 2006 8:56 pm
Location: CA

Re: How do I use every last electron?

Post by solarstudent » Mon Aug 18, 2008 7:04 pm

Great write-up and solution! Thanks to Crewzer and staff at OB for prioritizing this "challenge".

How many other companies interact with their customers on this level? Again, kudos to OB! =D>

I'm placing my order for a FM-60 and putting my electrons to work!

Question- Do I need a later software version than may be on the distributors shelf?

More later. :grin:
Trace/Xantrex/Schneider Onduleur=SW5548/D. Shell SM110 x 32, MX60, Surrette 1100Ah@48v, TM500a

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Tue Aug 19, 2008 5:51 am

SS,

Thanks for the kudos! :cool: It's my understanding that the new FM60 includes the "latest" firmware. :wink:

I hope you'll keep us posted on the results of your installation and operation.

Regards,
Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

kitestrings
Forum Member
Posts: 7
Joined: Tue Apr 01, 2008 12:45 pm
My RE system: Off-grid since 1984: Sencenbaugh 1.2 kW wind turbine on 100' Rohn 25G, 16- 180W Mitsubishi panels, OB VFX3648 w/mate, MX-60, PS2DC175, 16-L16HC's, Onan 4.5 LPG gen, Solar Converter 12/48-30 equalizer, Rutland 913 "functional" weathervane.
Location: 'Kingdom', VT

Re: How do I use every last electron?

Post by kitestrings » Wed Aug 20, 2008 8:04 am

Crewser, Solarstudent, OB Folks, contributors,

Again my thanks for this thorough and informative post.

I have two minor comments:

1) I think you meant to say ((29.56^2)/1ohm = ) 870 watts in your last paragraph under Load Specifications.

2) I suppose there are potentially a lot of different water heater configurations, but my plan has been to use two SS relays and the exsisting snap-action, stock thermostats not to break the element load, but to break the coil (or low-voltage signal) to the relays. This way you still have the temperature adjustment afforded by the stock thermostats, and don't need to figure out where to mount something that is not.

I'd attach a skematic if I knew how, but it was genarally an approach I gathered here (though slightly different wind/dump context):

http://www.fieldlines.com/story/2007/1/27/145044/045
My plan was to use SPST SS relays, I've found this source to be pretty inexpensive (no affiliation, or back-ground checks), but my first order was very prompt:

http://www.futurlec.com/RelSS.shtml

~kitestrings

tonyrabbit
Forum Member
Posts: 8
Joined: Thu Dec 04, 2008 2:39 pm
My RE system: 1st system - Off Grid:
2500W PV array at 48V
Flexmax80
Outback FX Inverter
1400 Ah Staab L-16 Battery bank at 48V
Trimetric
--------
2nd system Off Grid:
2600W PV array at 48V
Flexmax80
Trace 4048 inverter
1600 Ah Rolls-Surrette battery bank at 48V
Trimetric

Re: How do I use every last electron?

Post by tonyrabbit » Wed Mar 11, 2009 2:50 pm

I have been attempting to implement an opportunity load system as described here and am having a problem with my SSRs burning out and I'm wondering if anyone has any suggestions.

Here's the specs:

PV array: 2600 W at 48 V
Charge Controller: FM80
Diversion load: 2 water heating elements, each with 3.1 ohm resistance, wired and switched separately
SSRs: two SSRs each rated for 40 A DC load, one for each of the resistance heaters.

My FM80 is set with the Aux mode in Solid State with the Relative V and Hys V both set at 0 V, Hold time is set at 7.5 secs, and Delay was set at 5 secs.

Each water heater draws 18-20 amps when switched on (measurement taken when they were hooked up to bypass the SSRs).

Each SSR is mounted on the heat sink provided by the manufacturer which is mounted on plywood near my breaker box.

After some brief experiments with setting the Aux manually to "On", I turned the Aux on to "Auto" as the FM80 was going into Float. The display brought the Aux up to 100%. I then switched on the two breakers for the water heaters and the Aux went down to between 3 and 15%. The LEDs on the SSRs blinked on and off showing they were getting pulsed by the FM80.

After a few minutes the SSRs and their heatsinks got warm to the touch. I heard a "sizzling" sound and one of the SSR's light went out. I immediately switched off the breakers and turned the FM80 Aux to "off". The other one soon sizzled and died as well. Once they had cooled I noted that each had come off of its backing and was no longer connected to the heat sink. Basically whatever adhesive they were held together with let go when it got hot.

Here's the SSRs I bought:

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 0169795453 (scroll down to get details)

and here is their user manual

http://www.sure-electronics.net/DC,IC%2 ... manual.pdf

Their manual says:

"If you want to use those SSRs with over 20A current, you should be careful that they should be installed on a very good heatsink, it is strongly suggested that add an additional fan to enhance the air flow. Heat resistance must be less than 2.5C/W. Over load and over heat will destroy the SSR."

The current was just under 20A and it was installed on the heatsink they provided (not sure if its "very good" or not), and there is full airflow over the device and heatsink with an ambient temperature of 60 degrees F.

So, do you think the problem is that I bought cheap SSRs and I need to buy something of higher quality? If so can anyone suggest a brand or source?

Is there something about the rapid switching that could have caused this problem?

Thanks,
Tony

Vic
Forum Czar
Posts: 995
Joined: Sun Jul 09, 2006 12:56 pm
Location: San Joajuin County, CA

Re: How do I use every last electron?

Post by Vic » Wed Mar 11, 2009 3:57 pm

Hi Tonyrabbit,

Interesting. First, in looking @ the listing, it notes, "If SSR (DC-RR010 and DC-RR011) works with DC-RR012 heat sink, the maximum continuous load current could be 5~10A ".

Think that 18-20 Amps is TOO MUCH for the sink provided. A fan could help/solve the problem, but when the fan fails, U may be back here again. If there is a note in the manual about the maximum voltage drop across the SSR vs current, you could calcualte the required sink at some temp rise.

For convection cooling, sink fins should vertical and unobstructed. Gluing SS devices to skinks is uncommon. Perhaps what you are seeing as adhesive is really Thermal conductive grease. Normally, fasteners or some form of clip is used to nail down the SS device to its sink.

Suggest a large sink, with numerous fins an inch or so in height. Use separate sink for each SSR, and each sink insulated electriacally from the other.

Perhaps find a USA name-brand supplier, with a real spec sheet/application note available.

Am sure that you realize that the sink temp rises a specific number of degrees above ambient with at a specific heat output from the SSRs. A very hot environment means that the SSRs will run that much hotter. It is just temp rise above the ambient temp.

Crewzer shoud chime in with more on what he was using for the experimental setup at OB. Good Luck Vic
4/20/08: 18 Shell SQ 175-106 Vmpp, Stacked 5548 SW+, 1350 AH Surrette 4KS25's, MX-60, Kubota SQ-3250 25 KVA Polyphase Diesel genset. Thanks OutBack for this Forum + the great Support and Service.

tonyrabbit
Forum Member
Posts: 8
Joined: Thu Dec 04, 2008 2:39 pm
My RE system: 1st system - Off Grid:
2500W PV array at 48V
Flexmax80
Outback FX Inverter
1400 Ah Staab L-16 Battery bank at 48V
Trimetric
--------
2nd system Off Grid:
2600W PV array at 48V
Flexmax80
Trace 4048 inverter
1600 Ah Rolls-Surrette battery bank at 48V
Trimetric

Re: How do I use every last electron?

Post by tonyrabbit » Wed Mar 11, 2009 4:27 pm

Hi Vic,

Thanks for your input.
Vic wrote: Interesting. First, in looking @ the listing, it notes, "If SSR (DC-RR010 and DC-RR011) works with DC-RR012 heat sink, the maximum continuous load current could be 5~10A ".

Think that 18-20 Amps is TOO MUCH for the sink provided. A fan could help/solve the problem, but when the fan fails, U may be back here again. If there is a note in the manual about the maximum voltage drop across the SSR vs current, you could calcualte the required sink at some temp rise.
I guess I need to read the details more fully. I guess the heat sink provided was indeed NOT "very good"
Vic wrote: For convection cooling, sink fins should vertical and unobstructed. Gluing SS devices to skinks is uncommon. Perhaps what you are seeing as adhesive is really Thermal conductive grease. Normally, fasteners or some form of clip is used to nail down the SS device to its sink.
I should have been more clear. The SSR was attached to the heatsink with screws. The SSR body came loose from its metal back (which remained attached to the heatsink). The substance holding them together looks and feels like silicone caulk. Inside the SSR I can see a small metal plate surrounded by a black substance that looks like it was once liquid. I imagine all the magic parts are covered by the black stuff. :smile:
Vic wrote:Suggest a large sink, with numerous fins an inch or so in height. Use separate sink for each SSR, and each sink insulated electriacally from the other.
The heatsink provided is 2 by 3 by 2 inches and has five fins. Makes me wonder how big of one I will need.

Tony

Vic
Forum Czar
Posts: 995
Joined: Sun Jul 09, 2006 12:56 pm
Location: San Joajuin County, CA

Re: How do I use every last electron?

Post by Vic » Wed Mar 11, 2009 4:42 pm

Hi again tonyrabbit,

OK, well, if you could find USA SSRs you would be far ahead IMHO. And sorry, my comments may have assumed too much.

Have not reread this thread, or any supplemental info. Think that Crewzer might have speced the SSR he was using. In the old days, International Rectifier and Crydom (sp?) made very good AC SSRs. Do not know who is a leading high quality supplier now

It is possible that I am being too critical on these things from China. Sometimes, many corners are cut. Sometimes, reject parts are sold into secondary markets and so on.

SSRs are a very good approach for this application as opposed to mechanical relays. For experiments, a smaller sink and a fan should be fine, but in time, all fans will fail, and ideally, a fan should be temp-controlled, which adds some complexity, and its own reliability issues.

Good luck, and Crewzer is probably lurking, and he has good info on this. Good Luck, Vic
4/20/08: 18 Shell SQ 175-106 Vmpp, Stacked 5548 SW+, 1350 AH Surrette 4KS25's, MX-60, Kubota SQ-3250 25 KVA Polyphase Diesel genset. Thanks OutBack for this Forum + the great Support and Service.

User avatar
tjdadj
Forum Guru
Posts: 228
Joined: Sun May 27, 2007 4:00 pm
My RE system: FW500 Gridtie w/ [2]GVFX3648, MX60, 2400W PV [12-Sharp ND-200W (3x4)]
(3) 220Ah battery banks, [(2) are stationary, (1) is in my 48V EZGO Golf Cart]
Location: PacWest
Contact:

Re: How do I use every last electron?

Post by tjdadj » Wed Mar 11, 2009 5:21 pm

tonyrabbit

After reading the heat sink.pdf from the ebay link you posted, which stated that that sink was good for 5A w/o fan, you're gonna have to find some big heatsinks to keep those SSR's cool enough to survive.. I recommend you look for some surplus heatsinks from the electronics surplus websites..
or this unit: HERE:

hth
Terry In Oregon
---------------
PC Support (my other life!)

User avatar
crewzer
Forum Emperor
Posts: 1927
Joined: Mon Sep 27, 2004 10:50 am
Location: Vienna, VA

Re: How do I use every last electron?

Post by crewzer » Thu Mar 12, 2009 4:12 am

Chief lurker/chimer here... :wink:

The heatsink-mounted SSR assembly I used is generally described in this post above: http://www.outbackpower.com/forum/viewt ... =25#p17545 I ordered the following pre-assembled unit:

HDD-06V75-HS1.0 HDD-06V75 on a HEATSK-DIN-1.0 heat sink

Pre-assembled units: SSR, mounted on a properly sized heat sink, including a safety cover, thermal pad and properly torqued installation screws. The 1.0 C/W heatsink is rather large... quite a bit larger then the 1.6. Vertical mounting to promote convection cooling is preferred.

This particular SSR is rated for 75 A at 60 VDC (resistive loads) at 40 C when properly heatsinked (heatsunk?). The 60 V limit is fine for my 24 V system, but it's probably too low for a 48 V system. The HDD-1V40-HS1.0 (HDD-1V40 on a HEATSK-DIN-1.0 heat sink) is rated at 100 VDC (resistive loads) and is probably appropriate for a 48 V system with up to 40 A of charge current. I'd consider two of these bruisers in parallel for larger systems.

Note that the operating voltage is reduced by ~50% when used with motors and other inductive loads.

See this link on paralleling SSR's: http://www.power-io.com/library/appnote ... tching.htm

Here are links to additional info:

SSR family info: http://www.power-io.com/library/databul ... family.pdf
"Parts" page": http://www.power-io.com/products/heatsinks.htm
Heatsink dimensions: http://www.power-io.com/library/drawing ... nsions.htm
PWM: http://www.power-io.com/products/pwm-description.htm

Power I-O was very easy to work with. I placed my order for a single SSR assembly (and a few other misc small parts for another project) on line, paid by CC, and the gizmo arrived a few days later.

HTH,
Jim / crewzer
090805 System Configuration: 966 W STC (849 W CEC PTC) 48V PV array, FM80, 24V x 400 Ah AGM battery bank, FX2524T w/ BTS, Hub-4 & Mate; Link-10 w/ BTS, & E-Panel.

tonyrabbit
Forum Member
Posts: 8
Joined: Thu Dec 04, 2008 2:39 pm
My RE system: 1st system - Off Grid:
2500W PV array at 48V
Flexmax80
Outback FX Inverter
1400 Ah Staab L-16 Battery bank at 48V
Trimetric
--------
2nd system Off Grid:
2600W PV array at 48V
Flexmax80
Trace 4048 inverter
1600 Ah Rolls-Surrette battery bank at 48V
Trimetric

Re: How do I use every last electron?

Post by tonyrabbit » Thu Mar 12, 2009 4:35 pm

crewzer wrote:
This particular SSR is rated for 75 A at 60 VDC (resistive loads) at 40 C when properly heatsinked (heatsunk?). The 60 V limit is fine for my 24 V system, but it's probably too low for a 48 V system. The HDD-1V40-HS1.0 (HDD-1V40 on a HEATSK-DIN-1.0 heat sink) is rated at 100 VDC (resistive loads) and is probably appropriate for a 48 V system with up to 40 A of charge current. I'd consider two of these bruisers in parallel for larger systems.
Thanks for the info, Crewzer. I have gone ahead and purchased two of these 40 amp SSRs with heatsinks. I'll keep you posted on how things turn out.

For anyone else who is looking for SSRs and heatsinks I also found some good options from Crydom

http://www.crydom.com/en/Products/Catal ... 0500%20VDC

They were a bit cheaper (from the right reseller) but the heatsinks they sold were larger and space requirements were a factor for me.

Thanks for everyone's help.

Tony

tonyrabbit
Forum Member
Posts: 8
Joined: Thu Dec 04, 2008 2:39 pm
My RE system: 1st system - Off Grid:
2500W PV array at 48V
Flexmax80
Outback FX Inverter
1400 Ah Staab L-16 Battery bank at 48V
Trimetric
--------
2nd system Off Grid:
2600W PV array at 48V
Flexmax80
Trace 4048 inverter
1600 Ah Rolls-Surrette battery bank at 48V
Trimetric

Re: How do I use every last electron?

Post by tonyrabbit » Mon Mar 16, 2009 10:53 am

So I got my SSR's from PowerIO (amazingly they arrived within 36 hours of ordering). The Heatsinks are serious!

I've got one of them wired up and am testing the system and am seeing some odd behavior on the FM80.

When the FM80 goes into absorb mode the Aux turns on and starts showing a percentage on the main screen. The percentage fluctuates as the FM80 keeps the voltage at the absorb voltage. Soon enough the FM80 starts reading MPPT Bulk but the Aux stays on and continues to fluctuate. At first the fluctuation was between 25% and 45% but now that we are getting full sun its between 90% and 100%. The voltage stays close to the target voltage for Absorb, fluctuating 0.1 V above or below the target. During this time the Absorb time does not increase nor does the ChgT on the Misc screen.

My settings are as Crewzer suggested:

ÔÇó Relative Volts: 0 V
ÔÇó Hysteresis: 0 V
ÔÇó Hold: 7.5 seconds
ÔÇó Delay: 5 seconds

I also tried fiddling with the settings and found the following:

If I changed relative volts to .1 the Aux never goes on. This is not surprising as the FM80 will cut the solar input to keep the voltage from going over the target.

If I changed Hysteresis to .1 or .2 it changed the speed of the fluctuations on the Aux percent and the voltage but I still get essentially the same behavior

Changing the Hold or Delay did not seem to make much of a difference.

It seems to me that when the FM80 goes into MPPT Bulk the Aux should go off. In addition, it seems like it should keep the Aux percent lower so that it stays in absorb mode.

I also adjusted the Absorb voltage lower as part of testing. With it lower, the Aux % would get to 100% and stay there, and then the FM80 would go into absorb mode and the counters count up.

Anyone have any suggestions?

Crewzer, did you experience anything like this?

Thanks,
Tony

kitestrings
Forum Member
Posts: 7
Joined: Tue Apr 01, 2008 12:45 pm
My RE system: Off-grid since 1984: Sencenbaugh 1.2 kW wind turbine on 100' Rohn 25G, 16- 180W Mitsubishi panels, OB VFX3648 w/mate, MX-60, PS2DC175, 16-L16HC's, Onan 4.5 LPG gen, Solar Converter 12/48-30 equalizer, Rutland 913 "functional" weathervane.
Location: 'Kingdom', VT

Re: How do I use every last electron?

Post by kitestrings » Tue Apr 21, 2009 1:33 pm

Greetings,

I thought I would post a short update to this post. We have been pre-heating water using the aux-diversion function for about a week and a half now. It seems to be working very well for us (in our case using an MX60). Here's a link to a more detailed discription (with photos):

http://www.fieldlines.com/story/2009/4/16/20127/6123

Tony, Crewzer is probably a much better source to answer your question, but I'm not at all convinced what you've described is 'odd behavior'. It sounds like the CC is attempting to maintain target voltage (absorb and float), and there are times - perhaps when loads are applied - when the you have a MPPT Bulk mode re-initiated.

Regards, kitestrings

Keyosz
Forum Member
Posts: 13
Joined: Fri Apr 03, 2009 3:21 am
My RE system: 8Kwp PV Sharp modules, 2 FlexMax80 charge controller, 1 Mate, 1 FNDC, 2000Ah 48V tubolar stationary flooded lead-acid (2V elements), Studer HP-6000VA inverter

Re: How do I use every last electron?

Post by Keyosz » Wed Apr 29, 2009 6:39 am

Hello everyone,
i was very interested about this feature of the FlexMax80, that's one of the reasons i have bought it, and finally i have been able to build it me too.
I have 4 water heaters elements each with 48V 600W resistor, all connected in parallel and driven by an SSR connected to the aux port, everything is configured as described in this thread and it is now working since about 3 weeks, i can say that it is working very well and i am satisfied of this fine diversion control, the battery are always 100% charged and i have water pre-heating working very well reducing the gass bill, congratulations for this feature Outback! :grin:

Justme
Forum Guru
Posts: 106
Joined: Sun Apr 12, 2009 2:55 pm
My RE system: 15 x 80 watt panels (1200w)
1 x FX80 controller
6 x 2v cells 1550ah @ 5c
1 x 200watt wind turbine
3000w 120amp 12v inverter charger

Re: How do I use every last electron?

Post by Justme » Wed Apr 29, 2009 9:54 am

Would I be right in thinking that this feature will allow people to use the FX range as wind turbine controllers on any (within its power & voltage limits) wind turbine not just the controller & turbine package that is sold?

Not that I am planning to just that reading about this its screaming out at me as a possible solution.


Justme

kitestrings
Forum Member
Posts: 7
Joined: Tue Apr 01, 2008 12:45 pm
My RE system: Off-grid since 1984: Sencenbaugh 1.2 kW wind turbine on 100' Rohn 25G, 16- 180W Mitsubishi panels, OB VFX3648 w/mate, MX-60, PS2DC175, 16-L16HC's, Onan 4.5 LPG gen, Solar Converter 12/48-30 equalizer, Rutland 913 "functional" weathervane.
Location: 'Kingdom', VT

Re: How do I use every last electron?

Post by kitestrings » Thu Apr 30, 2009 9:36 am

Justme,

A couple thoughts...

'Crewzer and Darren did a pretty good job of outlining this appoach for PV diversion, and expaining how an opportunity load differs from diverison regulation.

My feeling is the MX is pretty crude as a diversion regulation device/controller. It seems like a Morning Star or Trace/Xantrax cntroller would be a lot faster and more precise.

I would also be concerned, assuming you could match the controller to the turbine, that once the diversion load was satisfied that you'd no longer have a valid dump. There are ways around this (a alternate dump load for example), but it is one more thing to consider.

The Outback manual, at least for the MX, warns against using the unit with wind to the point of voiding it's warrantee. Far be it from me to say "don't color outside the lines", but you should be aware of it. There's been some recent related discussions, like these:

http://www.outbackpower.com/forum/viewt ... f=9&t=3419
http://rockriver.us/7.htm

In our particular case we have a small (1kW) wind turbine, with it's own regulator/controller, connected to the same battery bank. I've noticed conditions where the the array output is below that needed to sustain the diversion load, but where the wind is contributing enough to cycle the aux diversion. I've had to reduce the delay time to account for this, but it does effectively capture some otherwise lost power.

With the Aux Diversion enabled our PV kWh average for April is already over last July's.

kind regards,

~kitestrings

Justme
Forum Guru
Posts: 106
Joined: Sun Apr 12, 2009 2:55 pm
My RE system: 15 x 80 watt panels (1200w)
1 x FX80 controller
6 x 2v cells 1550ah @ 5c
1 x 200watt wind turbine
3000w 120amp 12v inverter charger

Re: How do I use every last electron?

Post by Justme » Thu Apr 30, 2009 10:12 am

I think what you say could have some validity apart from


1, Outback do market (via a third party I think) an FM80 with a wind turbine with minor software setting changes (relating to power tracking I think but dont quote me)
2, The type of load could easily be changed from an opportunity load to a dump load (and lots of people are using water heaters as dump loads)


Not sure why any one would want to do it when other controllers are available for much less money. Just seemed like it "could" be done.

Justme

Guy
Forum Junior Member
Posts: 3
Joined: Sat Jan 12, 2013 3:53 pm
My RE system: 1150 solar, Magnum MS4024PAE inverter, Outback FM60 charge controller, 4 strings of 2 -135ah 12V AGS batteries, 24v system. Weekend cabin. Biggest load is smaller electric fridge. added a HL-100 air diversion heater.

FM60 Diversion Relay Settings

Post by Guy » Sun Jan 13, 2013 9:09 am

I have recently added a HL-100 air diversion heater in an effort to reduce propane space heating when there is excess solar power available. We left the heater at the 4 ohm setting and tied it with a mechanical relay.

My current set points are 0.2V for both relative and hyst. Delay is 5 sec and hold is 25 sec.

When there is lots of power (say 500W) the relay came on in absorb and stayed on in float until power dropped to 300W. Then it began to cycle until the power dropped off in the late afternoon and the voltage stayed below the float 27.2 plus the 0.2.

Any suggestions on better settings that would minimize cycling through this transition period where the cabin load plus diversion heater load and battery state of charge causes the voltage to cycle tripping the relay on and off for the hold 25 sec?

Any experience / suggestions would be appreciated.

Guy

Guy
Forum Junior Member
Posts: 3
Joined: Sat Jan 12, 2013 3:53 pm
My RE system: 1150 solar, Magnum MS4024PAE inverter, Outback FM60 charge controller, 4 strings of 2 -135ah 12V AGS batteries, 24v system. Weekend cabin. Biggest load is smaller electric fridge. added a HL-100 air diversion heater.

Re: How do I use every last electron?

Post by Guy » Sat Jan 19, 2013 6:28 am

It has occurred to me that the mechanical relay I installed to control the diversion heater is a bit of a clumsey device for this purpose. As this is simply on and off, I will always get the cycling when I am on the edge of having enough power to operate it. Adjusting the settings may reduce cycling at the cost of some production.

Am I correct in thinking that I would not get on / off cycling had I installed a solid state relay? The relay would prorate the power between the batteries and the relay to hold their charge at the float coltage. A smoother diversion control process?

Guy

Post Reply