Cooling Dummy Load for Power Converters

That could be a good idea.

However, I do not know how we would prevent burglars from stealing not only the heaters, but also the cables. As far as I know, fenced windows, the security guard, the alarms and the security cameras are barely effective for preventing people from getting inside the building.

Put them on the roof where they are hard to see.

Haha, boy you do live out in the sticks....I once had a 20 ton package unit stolen off the roof of a 5 story building....nobody saw or heard anything....we know they had to have a crane, and a big one at that...

Yeah, happened to our neighbour here also, and that was two stories straight up to get to them.

Sounds like you either have a crime problem, or a suspicion of crime problem. Hard to tell from where I sit.

Why not do both? How about putting the dummy loads up high and out of reach?

Have one or two dummy loads (resistive), and at least one A/C unit that only runs when the test is active on the dummy loads? Heat does affect the value of the dummy resistors, so that might be a ready justification for getting the A/C...

instead of the resistors, use plain old water, with either salt or baking soda added to draw the needed amount of current.

Simple plain old metal electrodes ( Threaded rod) work just fine.

20 kW electrolyzer - kind of a serious hydrogen-oxygen explosion hazard unless the bath is outdoors and fenced off?

Then if that is the case, then these are awful dangerous,

There is a difference between an actual heating element, and an electric current through salt solutions, where the potential is high enough to make hydrogen and oxygen. It appeared to me the suggestion was to electrolyze water as a test load.

One other weakness of that idea, the current-voltage characteristic of said electrolysis is never a constant load, but increases power draw (same potential more current) as it heats up, so it wants to heat up more.

Yes the water in the tank serving as the load may even reach the boiling point after certain period time elapsed... and the higher the current drawn the hotter the water temperature will be!

Typically In such setup, output voltage is set then current drawn is regulated by either;

1. adjusting the water level in the tank

2. adjusting salt concentration

3. raising /lowering the height of the submerge electrodes..

Maybe with 1/4"-20 all-thread electrodes, the thing will tend to self-limit, since electrode area is pretty small. I presume this set up is with AC applied, correct?

It was an auto adjusting /reversing 3-phase motor coupled to the water bath load...

We used 3 pieces of rebars 1/2" X 60", separately mounted equidistant with one another on a 1/2" thick plastic slab. The plastic slab/ plate being hanged via a rope thru a pulley driven 3-phase motor for dipping level water bath adjustments that also adjust the amperage of current drawn by the water load..

How much inductive overload can your converters deliver? The LRC of a compressor can easily be 10x the FLA, and air conditioners can be sensitive to off-nominal voltages.

Better to follow TMC's advice and relocate the fans outside, or if you have a need for hot water, get some electric water heaters and use the elements as your dummy loads. Surely there's an engineer there that can figure out how to wire them up to match the converters' output.

I "forgot" to mention one of our clients is an Air Conditioning manufacturer, and we do have IGBT inverters installed in Light Rail Trains powering their air conditioners. Therefore, my idea is very likely to work, but there must be something wrong with it since none is doing it nowadays in the industry. 10x times inrush current does not sound very nice, but if it works it should not be an issue.

BTW, I have not called them because they usually are not very eager to help their distributors. And I do not think I have a good deal to offer them (yet).

You could buy kilns and fire ceramics and sell in the gift shop, or bake bread and have loaves for employees to take home...Really you could use any purely resistive load....you could power a boiler to steam clean stuff.....or you could have an absorption system....

1. Air conditioners start/stop according to air temperature - not when you want to have a load.
2. They are notorious for being difficult to start 6 - 8 times running current.
3. Running current will actually depend on entry/exit temperatures.

It seems to me far more sensible to put inverter output into the mains supply - only the rectifier + inverter losses will heat the room & be drawn as net amount from mains & you will reduce power demand from mains considerably - GREEN POINTS with knobs on.

Obviously, the units must be synchronised to mains & if you need a different output voltage, you will have to have a variable auto-transformer [VARIAC] between inverter & mains. Maybe rectifying output to a "DC bridge" and a "solar power" type inverter to mains would be more flexible - your inverter could then be 60 Hz if necessary. I suppose you could have some solar panels hooked-in as well.

What I am thinking is that whenever I need heat, I would use resistive loads, and whenever I need to extract the heat, I would use the air conditioner (at fixed power, regardless of the temperature of the room).

Grid connected Power Converters: Using the AC mains as a load is very dangerous due to stability issues. Voltages go crazy, and so do currents, tripping the MCCBs. Back to back connections take no less than a month and it is very easy something to go wrong. For instance, bidirectional Electric Vehicle converters have been tested for a while but as far as I know, currently research papers is the best that has been achieved. However, since our electric bill would decrease, it is a possibility to take into account.

VARIAC performance: Regarding VARIACs, I have seen 2 types. The small ones, up to about 20kW, which are manual, can barely handle the ZERO LOAD currents of our converters. On the other hand, the big ones (hundreds of kW), use an induction motor to operate so they have wear and tear susceptibility, apart of being bulky, heavy and expensive. Autotransformers are helpful for preliminary tests, but when it comes to routine tests they lack productivity.

Product range: We have rectifiers, hooked to the 3ph AC mains, and 3ph inverters, hooked to DC power supplies. Inverters could be connected directly to the 3 phase load. Rectifiers cannot, so I would need to design an auxiliary "(2 stage) wide input 3 phase inverter".

As others have said, ac's are not a good load, with high starting current and inductive pf. Can you use resistive heaters to heat a large tank of water? Water has a high heat capacity and can absorb a lot of energy. Just a thought...

Keep talking.

Let's say I heat up tap water. What comes next? I pour it into the drain? There must be some legal issue preventing me from doing that in large quantities.

What do you define as large quantities and why do you feel you would need such a flow rate?

If you are at an engineering firm I would assume that at least one person there is competent enough at their profession to do basic engineering calculations to know exactly what flow rate of water would be required to carry away 20 KW worth of thermal energy for a given temperature differential.

And then use a fan to cool that water off (dry cooling) enough to send back around for another go, then there is no wastewater to dispose of.

It could be antifreeze or liquefied gold suspension, and the cost is about the same, if you never throw it away.

That would be me?

Let's assume it's 100kW, average of what we have. At 4.184J/ºC/g, 70ºC (boiling point), we would have a flow of 0.3 L/s.

Not bad at all, but I still wonder for how long I could keep doing this without someone complaining. After all, it would not be the first project that does not work due to the poor capacity of the soil absorbing heat (WaterSeer debunk).

Since when does water boil at 70 °C? Do you live in outer space or near it? Just kidding, I see now that you meant to say 70 °C rise in water temperature.

OK, 4.2 kJ/Kg-°C, so 100 kJ/s/(4.2 x 70) = 0.34 L/s (20.4 L/m, 1224 L/hr.

That is a considerable amount of boiling water to contend with, so you will need a substantial fan applied against a large coil, with a substantially larger pump than 20 L/m in order to keep the thermal rise of the water within bounds, and maintain a steady cool side temperature. When you are attempting to cool boiling water by fan, you will need a lot of air flow.

Other option: Heck just go ahead and make steam, and vent away the steam to the air. You will find the latency of converting to vapor a large heat absorber.

I'm not seeing how any of that constitutes "substantial".

A small hydronic heating system circulator pump can easily circulate 5+ GPM (20+ LPM) through 50 feet of 1" PEX pipe. Double that at less pipe distance or a larger diameter.

Plus most any full size pickup truck radiator can easily handle 300+K BTU of heat transfer with any fair quality fan systems on it. Step up to a large commercial truck or heavy equipment radiator and you can do 2 -3X that easily.

But as I have said before, why bother when a substantial and passive electric load bank can be made from cheap common items like repurposed baseboard heaters and their parts solving the whole issue with minimal engineering and work to set up. Which for that matter they could be easily mounted on a cart that could be wheeled out to wherever it would work then put away when done thus eliminating the theft issue.

My point is, that to me all of this seems like what should be some very basic exercises in logic and basic engineering principles which makes me think the OP is not actually an engineer nor works with any real engineers or mechanically compliant people in general let alone at a legitimate business.

Too many basic details and the lack of justifiable reasons behind them don't add up in my books.

Hot tub for the employees...?

Instead of heating "AIR" make hot water for building use, and if the water gets too hot simply run the water some to cool the tanks down.

Here is a similar situation to yours, my cousin used to work as a glass blower in a house converted to a multi-studio. There were several high temperature kilns and each station had their own torch setup.

Yet despite all these heat sources and scorching summer heat from time to time, the artist loft stayed down near winter temperatures and I had to wear warm clothing while visiting.

The difference here is they had sealed every gap in the converted house and mounted 2 foot diameter, high speed, cowled fans, all pointing outside.

There were at least a dozen of these that created enough vacuum to slam the front door shut every time we entered and left.

However, the fans did produce around 60db of sound.

I know that feel.

We already have extractors (fans) making about 80dB of noise. The worst part is that certified Personal Protective Equipment are not as effective to 50Hz (fans) as to frequencies above 1kHz (SMPS). I wonder if that would have something to do with WAVELENGTH of the sound.

and if you are worried about running the hot water down the drain, then get a cooling tower. and then no problems.

For the scope of OP's project, they only need a loop with a fan blowing air over a set of coils of copper, in parallel flow. One of those floor type industrial fans on wheels should about do it.

I think the reason for that testing approach is just to verify the functionality of the product without breaking/destroying anything!

Any unexpected, unscheduled power surges, intermittent shutdowns will be harmful to both the test load(s) and the DUT...

In the early 70s, I used tanks of brine solution with 3 electrodes in it for testing UPS and power inverters similar to yours! The tank was located outside and fenced for convenience and safety..

At a navy reactor prototype they ran the power to a salt water tank and boiled water. If this is DC, you could run the power to Peltier junctions and use the output as a solid state heater or cooler depending on the season.

This is a potential investment decision that only the <...we...> can make, really.

At ~20 KW that could easily be dissipated through a window using a simple 'U' shaped ductwork or dedicated externally located wall mounted heater grid made from the guts of multiple smaller baseboard heater units.

For an engineering firm I have to wonder how good of engineers they are to not be able to figure out how to set up a simple 20 KW heater system in some fashion that is unobtrusive plus of so little concernable value and ina location of such to where no one would ever put forth the effort to try and steal it.

In terms of theft, things that are of low value located in hard to get to locations plus that are hard to remove tend to not be messed with simply due to the perceived ROI being not worth the effort and risk of taking them.

Besides that, if the test loads are hot while being pilfered, they will let go in the morning, when they realize they just got through braising their finger tips, and that ain't no beef.

I have doubts to the actual claim of needing to be theft proof to any great degree given at best his claim in post 2 says they have security countermeasures in place that are apparently working being 'barely working security' that does keep stuff from being stolen is still working security that is keeping things from being stolen.

So unless his security guards are in nightly gun battles to keep the properly secure I have doubts that theft is as much of an issue as he claims or want to imply it is. Especially so for low value items that could be placed in hard to access locations.

Ok let me explain. All the measures are there, as I said. However, a few months ago someone tried to break in trough the air ducts. They have been found half broken. Luckily, the one responsible got stuck with something or just thought it is not worth it because they did not get in.

I do not think we have security issues. I am just saying that any proposal I make to the Board of Directors can be rejected if it contains clear security risks, as might be leaving equipment outside an industrial ship.

Call me crazy, maybe you are the one who is right and I am just talking nonsense.

Now you are telling the forum this is aboard an industrial ship?

Really? Just drop the conducting lines over the side of the ship, and shock a couple of fish, and be done with it.

The other suggestion: place the dummy load outside, but always have some power active on the load. You will find out who the thief is - - posthumously.

Industrial manufacturing building, no water nearby. I used Google Translate with the equivalent in Spanish (nave industrial) and I did not get a proper translation. My apologies, I did not have the luck of studying engineering in English.

I studies Spanish two years in High School, and it was supposedly Castillian. Meaning that when I failed to speak it properly and swore, Mrs. Lattimore would wash out my mouth with Castille soap. LOL

Now my Spanish is more mutt-style version that comes out of Mexico. Not only that, I am more of less clueless how to engineer anything in Spanish.

No problemo my amigo, vamos a ver hoy a hoy!

Here ya go,, Take the power being generated, and use a solar grid tie in to use the grid to dump all the generated power into, Infinite capacity, and the "Load" never changes, Plus most elec companies will pay you for the power you put into the system. a win win for everyone.!

Yes, that was the most beneficial answer to all concerned, unless they need room or district heating. 100 kW could heat a considerable office space, really.