RVelectricity: Hughes Autoformer testing

By Mike Sokol

Page Contents

My Hughes Autoformer testing has begun. Here are details on how much additional current an air conditioner draws as the voltage goes below 100 volts. Hint: Everything You Know Is Wrong!

Since this is a rather complicated answer to what appears to be a simple question about Hughes Autoformers (and autotransformers in general), the results of my study will be presented in pieces over the next several weeks as I gather more information that’s been peer reviewed.

Note that while Hughes has sent me a demo unit at my request, they neither offered nor have I asked for any compensation for this study. That way it can be 100% unbiased. Hughes has also agreed to accept the results of my study, which will be sent to the NEC board for review.

Here’s Part 1, where I show how much extra current an air conditioner draws as the voltage is reduced.

But before I show you the data I’ve gathered so far we all need an understanding of the theory of operation. So here’s your primer on the how this is supposed to work.

While resistive loads behave predictably as the voltage goes up and down, air conditioners (and all inductive motors) are not that simple. To understand what’s happening we need a quick look at an ohm’s and watt’s law chart. For a resistive load like an electric water heater element, there’s a predicable relationship between voltage and power.

Since watts of power is equal to the voltage squared divided by resistance, as the voltage drops so will the power. So an electric heater element rated for, say, 1,200 watts at 120 volts will be reduced to 1/4 power at 1/2 voltage. So at 60 volts it will only produce 300 watts (1/4) of power. There’s nothing to cause it to try to draw more amperage as the voltage goes down. We’ll revisit this concept below.

Not so with inductive motors. You’re probably aware that they have something called a start winding and capacitor, along with a rating for LRA (Locked Rotor Amperage). What this means is that when the motor starts up it tries to draw a lot more amperage (maybe 400% to 800%) of running amperage, which we call Inrush Current. The reduction in amperage as the motor comes up to speed is due to something called Back EMF (Back Electromotive Force).

Basically, the motor itself acts like its own generator that’s 180 degrees out of phase with the incoming current. That reduces the amperage draw when the motor is up to speed with no load. When you load the motor, the slip angle increases, which draws more current. Reduce the voltage to below its design rating and it draws even more current.

Air conditioners and refrigerators

So air conditioner (and refrigerator) compressor motors act differently than our basic water heating element (which is just a big resistor). This is what leads to the infamous extra current (and overheating) of an air conditioner’s compressor as voltage is reduced. The question is just how much and how fast does the current increase as the campground voltage decreases?

Enter the Autoformer

The Hughes Autoformer is basically an automatic low-voltage/high-current transformer connected in autotransformer mode via a relay and sensing circuit. So when the campground voltage gets below the set threshold, it performs a 10% voltage boost.

But what’s the cost?

Now, this 10% voltage boost comes at a cost of a 10% increase in current from the campground pedestal. But the real question should be: Is the extra 10% voltage provided by the Hughes Autoformer enough to offset the increased current draw from the pedestal due to the air conditioner’s extra current draw from running it at a low voltage? Hughes makes the claim that it does indeed reduce the air conditioner current enough to make up for its own 10% input amperage increase. That’s the subject of this entire test.

What about other loads?

Well, the electric element in your water heater will indeed draw more current from the pedestal when the Hughes Autoformer kicks into boost mode. That’s a given, which follows Ohm’s and Watt’s laws exactly. It’s basically just a big resistor that responds to voltage changes without any crazy Back EMF generator action.

And you can’t damage a water heater element because there are no moving parts. As I noted in my explanation above, even if you drop the incoming voltage to 60 volts, it will happily keep producing 1/4 of it’s rated wattage forever.

However, there are some weird power factor effects from PWM (Pulse Width Modulated) power supplies in RV converters and battery chargers which are largely self-regulating as the campground voltage goes up and down. when the incoming AC voltage is higher the Pulse Width Modulator produces a pulse with less duty cycle. But when the AC voltage is lower, it automatically increases the duty cycle of the current pulse. That’s why PWM power supplies will tend to draw just enough extra current to make up for any reduced AC voltage. And they don’t need an Autoformer in the circuit at all to output the correct voltage. But they are indeed part of the campground current balance equation.

The big picture

So what does this mix of load types (resistive, inductive and PWM) mean to campsite amperage with or without voltage correction from the Hughes Autoformer? To know for sure, I’m building a test system one piece at a time and adding them together to see how the entire system works. Maybe Hughes is correct and their Autoformer is beneficial in reducing overall current in campgrounds with low voltage due to increased air conditioner current. And maybe the NEC is correct that any autotransformer (or Autoformer) creates an extra amperage load on the campground’s electrical system, increasing the stress on an already overstressed electrical grid. Who’s right and who’s wrong? I honestly don’t know for sure yet.

Here are the numbers….

Here are the numbers I’ve found so far showing how much extra current an RV air conditioner draws as the voltage is reduced. Next week I’ll put the Autoformer in the circuit and do a comparative voltage sweep. That’s going to set the direction of the next experiment where I add in a PWM (Pulse Width Modulated) battery charger to see how it reacts with and without the Hughes Autoformer connected.

The setup

This is the preliminary data showing my proof of concept test bed with what I used for the experiment. Note that the Hughes Autoformer is NOT in the circuit for this test.

3kVA (3,000 Watt) VARIAC
Dometic Penguin II 15kBTU air conditioner
SoftStartRV™ starter
Southwire 14090T Tru-RMS digital voltmeter
Southwire 25015T Tru-RMS clamp ammeter

These are the amperage measurements with only the fan running at the following voltage steps:

120 volts = 3.78 amps
115 volts = 3.71 amps
110 volts = 3.84 amps
105 volts = 3.97 amps
100 volts = 4.12 amps
95 volts = 4.22 amps
90 volts = 4.53 amps

This works just as I predicted, with a 10% increase in amperage when the voltage was reduced from 120 volts down to 100 volts. But this next set of measurements blew my mind. These are volt/amp readings with the compressor running. Note that this is the total combined current of the air conditioner fan and compressor, not just the compressor amperage.

Here’s the amperage measurements with the compressor running at the following voltage steps:

120 volts = 12.06 amps
115 volts = 11.15 amps
110 volts = 10.86 amps
105 volts = 10.90 amps
100 volts = 10.93 amps
95 volts = compressor shut down

What the Heck!!! Even though all my EE books and urban myths say that air conditioner compressors will draw more amperage as the voltage goes down to 100 volts, that’s the opposite of what I’m seeing with this first test. While it’s drawing just over 12 amperes at 120 volts, when I run the VARIAC down to 100 volts, the current DECREASES to 10.93 amperes instead of INCREASING! That’s a 10% DECREASE in current when the voltage is reduced from 120 volts down to 100 volts.

Note that the compressor would drop out and refuse to run when I sent the air conditioner voltage down to 95 volts. Now that’s understandable, as Dometic may have some sort of low-voltage shutoff to protect itself.

Am I right or am I wrong?

So did I measure this incorrectly? I don’t think I made any mistakes, but I’ll double-check my work. Heck, I’m gonna triple-check my work with entirely different meters this week. I used high-end Southwire meters for this, but I have access to some top-shelf Fluke meters. Time to step up my game.

Is there some magic infused by the SoftStartRV™ controller? Definitely Not! Maybe it will help with starting the compressor at lower voltages, but it’s out of the circuit once the compressor is running. So there’s no magic there…

Has Dometic installed some sort of automatic current reduction circuit in the compressor? Nothing in their literature hints at that, but I’ve already sent these numbers to my engineering contact at Dometic for peer review.

Are all the text books wrong about induction motors having increased current draw at low voltage? I’m pretty sure not, since this engineering knowledge goes back 100 years; but perhaps it’s not as much of an effect as we’ve all assumed. Or maybe there’s another variable at play due to this being a compressor load. I just don’t know.

Much more on this drama next week, as I create alternate tests and try to qualify what’s happening. I LOVE it when measurements are unexpected. That’s true science.

So hold your breath until next week. But in the meantime, let’s play safe out there….

Mike Sokol is an electrical and professional sound expert with 50+ years in the industry. His excellent book RV Electrical Safety is available at Amazon.com. For more info on Mike’s qualifications as an electrical expert, click here.

For information on how to support RVelectricity and No~Shock~Zone articles, seminars and videos, please click the I Like Mike Campaign.

##RVe35

Mitch October 28, 2020 At 7:24 am

Mike
Fascinating article. While your test methodology is sound I can’t help feeling that consumers would benefit from a big picture test. If you tested with a RV you would see the impact that different loads have on the electrical performance at low voltages.

sally October 10, 2020 At 10:40 am

We bought our current rig in 2015. One of the first mods we did was to power. Due to the fact that our 1978 Prowler it had power issues. the cord caught fire . we didnt know yet about rv power. when we bought our current rig we modded power first. Chucked the 35$ 25 foot 50 amp wally world cord and replaced with Marinco 50 amp 25 ft with 90 degree elbow. replaced the receptical with Made in USA Marinco, and bought a 50 amp Hughs autoformer. One winter we were staying where the power was continuing to go through brown outs where the incoming was measuring 55volts. for hours. The autoformer was keeping us at about 95v. we werent running AC at this time. eventually the brown out was over. We have meters in 4 plugs in the rv .We have not suffeered any electrical problems we did in our Prowler or our Kit Kamper, we had previous to the Prowler. Our AC we put in New 5 years ago and have never had an issue and it runs 6 months out of the year. Not an expert but this is our experience.

Tony October 6, 2020 At 7:02 pm

HUH! The current did what? Started studying electronics as a hobby 60 years ago and if I heard your results from anyone but you, i would have quit reading when I got to the combined amperage chart. Can’t wait to hear the rest of the story.

- Mike Sokol October 7, 2020 At 8:03 am
  
  It will be in the mid-80’s this afternoon, so I’m going to reinstall the stock starting capacitor and run another set of tests today. Watch for my results in this Saturday’s newsletter.
  
- Mike Sokol October 7, 2020 At 8:06 am
  
  Crazy, huh? Yes, I about fell off my seat as I watched the running current go down instead of up while I was reducing the voltage using a big VARIAC. Stand by for more intel this Saturday in my RVelectricity column in RVtravel.
  
Sean McDougall October 5, 2020 At 3:57 pm

Hello Mike.
You are failing to consider an important element in your tests.

In setup #1) The Fan, which is a type of motor, does behave as expected. By pulling a similar amount of power, regardless of the available supply input voltage. i.e. more current draw occurs when voltage is reduced.

In setup #2) The Air Conditioner, which is a significantly more complex circuit, has a compressor which is also reactive. However, modern air conditioners are designed with energy efficiency taken into consideration. For this reason, that piece of equipment will have a RUN Capacitor installed into it. More then likely the run capacitor will be oversized by at least 25% as a safety consideration for low voltage situations which is common hazard. A Reactive Load with Capacitators compensating will behave like a resistive load. Hence a decrease in Voltage will result in a decrease in current as well. It will use less total power, but also do less work (Less Cooling).

- Mike Sokol October 5, 2020 At 5:02 pm
  
  That’s an interesting theory. More to study.
  
  - California Travel Videos December 6, 2020 At 1:30 pm
    
    Good points and if a given air conditioner or device is changing the phase (angle) between the voltage and current that results in a significant power factor reduction, this can be tested with a more comprehensive meter such as the Extech PQ2071… assuming you’re ready to spend $300 for an early Holiday Gift for yourself 🙂
    
Richard J Whitaker October 5, 2020 At 1:43 pm

I would be interested after running with low voltage how much damage is being done to the windings of the motors. We use a meggaohmer for all of our motors to check the insulation break down. I know on large HVAC units low voltage causes higher amperage which in turn causes higher temperatures in the windings. This breaks down the insulation on them and shortens their life.

Roy Brody October 5, 2020 At 10:00 am
- Mike, why are you using a SoftStartRV™ starter in your tests? 99% of the RV air conditioners do not have this. Shouldn’t this be a real world test?
- Mike Sokol October 5, 2020 At 11:03 am
  
  As I noted in another post, that’s simply because it was already wired in and this is just a preliminary test. I’ll put it back to the stock capacitor today and run another test tomorrow. I also plan to try this with a Supco Hard Start Capacitor at various low-voltages. So not to worry, I’ll have scads of data to interpret before making any formal announcements. It ain’t over ’til it’s over…
  
Leon October 5, 2020 At 7:17 am

Mike,
What are the readings with both the fan and compressor running at the same time? Have I missed something? Maybe the combination will be a surprise.

- Mike Sokol October 5, 2020 At 8:02 am
  
  That was the combined amperage of both the fan and compressor. The plot thickens….
  
- Ranger Carvel October 5, 2020 At 8:31 am
  
  Note that this is the total combined current of the air conditioner fan and compressor, not just the compressor amperage.
  
friz October 5, 2020 At 5:57 am

Thanks for the Ohm’s Law chart. I was not familiar with a Autotransformer. I had to read at least halfway through the article to learn it’s function! The “But what’s the cost” paragraph left me wondering “But what’s the cost”?

Dave Pellegrino October 5, 2020 At 5:28 am

Mike, why even have the Softstart in the mix? It’s just adding a layer to the equation. Try the tests without the softstart.

- Mike Sokol October 5, 2020 At 8:05 am
  
  I left the SoftStartRV connected since it was already in circuit. But I’m going back to the stock starting capacitor today and will rerun the experiment. I’ll probably run this test a few dozen times before I have a definitive idea of what’s happening. What this first test suggests is that I have to test EVERYTHING since the textbook answers don’t seem to apply.
  
Tommy Molnar October 5, 2020 At 4:48 am

I read all your newsletters and answers to other’s questions. Mostly I just scratch my head in wonderment at all the technical stuff. But, when you’re all done with an experiment, you put it layman’s terms that I can understand, and I appreciate that. Truck drivers never spent much time studying electrical engineering . . .

- Mike Sokol October 5, 2020 At 11:06 am
  
  While I didn’t drive as much as an OTR trucker, I did used to travel around 40 to 50 thousand miles a year doing seminars. So I spent a lot of windshield time writing articles in my head and doing little mind experiments like Einstein was famous for. Now I mostly drive my desk, but I still miss a good 1,500 mile trip.
  
Dick and Sandy near Buffalo, NY October 4, 2020 At 9:08 pm

As they used to say on SNL years ago VERY INTERESTING!

- Drew October 6, 2020 At 9:59 am
  
  I think it was actually Rowan and Martin’s “Laugh in.”
  
  - Mike Sokol October 7, 2020 At 8:09 am
    
    Yes, I believe you’re correct. We must be accurate with our pop culture references. https://www.youtube.com/watch?v=_SVSak1oBCw
    
    - California Travel Videos December 6, 2020 At 1:36 pm
      
      Or was Laugh In an homage to Hogan’s Heros dating back 3 years earlier with Sgt. Schultz famous chiche, “I See Nothing!” (yeah, I’m dating myself!)
      
      https://www.youtube.com/watch?v=OsXrpxo4uC0
      
Bob Wohlrab October 4, 2020 At 5:30 pm

Mike,
I’m not sure about your numbers but I will tell you the we stay at a campground that has power go low in peek times of the day. (Understandably hot days busy park)
I have had to replace 2 airconditioners and a Sony TV on my motorhome over the last couple years of workcamping for the summer at this park. A volt meter showed fluctuations from 100 to 130 volts almost every day. I bought a Autoformer and have not had any problems at all. Even if lower voltage causes a decrease in amps that decrease causes problems and damage to equipment.

- Mike Sokol October 4, 2020 At 7:07 pm
  
  Low voltage generally won’t damage electronics. But high voltage certainly can.
  
J richard October 4, 2020 At 4:42 pm

You have me scratching my head! That defies everything I learned in engineering classes.

- Mike Sokol October 4, 2020 At 7:05 pm
  
  Me too, but I have a theory that could explain my initial observations. Please stand by…
  
Dave Pechan October 4, 2020 At 4:39 pm

Is the compressor slowing as voltage is reduced, if so, power output would be reduced and therefore less input power would be needed?

- Mike Sokol October 4, 2020 At 7:08 pm
  
  That’s my SWAG…
  
  - RV Staff October 4, 2020 At 8:38 pm
    
    Well, I just learned something new — “SWAG.” For those of you who don’t know and don’t want to look it up, it’s “Scientific Wild A**ed Guess” — i.e., Mike has put in a disclaimer. 😆 —Diane at RVtravel.com
    
    - Mike Sokol October 5, 2020 At 8:06 am
      
      And a SWAG is better than a WAG since I’ve thrown in some Science…
      
  - California Travel Videos December 6, 2020 At 1:42 pm
    
    I’m with you Mike – I’d surmise the RPMs are going down, which in turn leads to a reduced Electro Magnetic Field, which makes the device less inductive, reducing the Power Factor… or something like that 😉
    
- Steve October 5, 2020 At 9:13 am
  
  It will be interesting to see if the compressor was turning slower – but you did not change the frequency so this is interesting. Is there a way you could track the motor temperature while you reduce the voltage? WOW too many variables!!
  
Reid October 4, 2020 At 4:14 pm

That can’t be !!!!
Did hell freeze over ?
Did the earths north and south poles
Change places ??

- Mike Sokol October 4, 2020 At 4:39 pm
  
  Hey, I make mistakes all the time. That’s how I discover new things. 😀
  
Drew October 4, 2020 At 4:03 pm

I bought an Autoformer anyway. It should help- no matter what. I’m not aware of what my Coleman Mach a/c’s draw at various voltages…maybe the reverse is true from what Mike’s Dometic unit shows. I wait patiently anyway to see these experiments and will enjoy my voltage increase while I read.

Beastman October 4, 2020 At 1:12 pm

Hi Mike, when the fan is running, why the dip in current between 120v and115v before it rises at 110v, 105, 100 etc.? Much Thankage

- Mike Sokol October 4, 2020 At 2:33 pm
  
  IDK, as they say. But it’s interesting. Later this week I’ll rerun at 1 volt intervals and load it into an excel spreadsheet. Something very strange is going on here…
  
Michael October 4, 2020 At 1:04 pm

It may depend upon compressor loading which is related to ambient temperature and humidity. High temperature High humidity puts a large load on the compressor motor. If your test setup is barely loading the compressor it may not produce the expected results. A more reliable test would be to use a similar motor and load it using a variable viscosity fluid brake or generator. That way you know exactly how loaded the motor is. We use this to test diesel engines.

- Mike Sokol October 4, 2020 At 4:32 pm
  
  Yes I realize that variable compressor loading is likely causing the drop in amperage instead of the expected rise in amperage as the pedestal voltage decreases. But the real question is does a standard RV air conditioner draw more current as the pedestal voltage goes down, and will any damage to the compressor motor result.
  
  - Michael October 4, 2020 At 9:34 pm
    
    If your intention is to verify the operation of a rv air conditioner then it may be a good idea to setup a controlled chamber where humidity and temperature can be controlled for testing. To insure maximum compressor loading you can set the temperature in the chamber at 100F and 99% RH. This will create a large load on the compressor. This should cause phase angle shift.
    
    - Mike Sokol October 4, 2020 At 9:51 pm
      
      Sadly, I have zero budget for this test. All of this is out of pocket for me. That’s exactly why I need a research grant.
      
      - Rob Sorensen October 5, 2020 At 8:45 am
        
        From a very non technical person, when I would need to adjust the refrigerant charge in my home split unit A/C, I would simply partially block the air flow to the condenser causing a greater heat load on the compressor. just a thought….
        GREAT information, as always
        My monthly subscription donation is the best money ever spent.
        Thanks again Mike.
      - Mike Sokol October 5, 2020 At 8:58 am
        
        You’re most welcome. And thanks for you support!
      - Michael October 6, 2020 At 9:30 pm
        
        Yes I understand that funding is a concern. The problem is that if we can’t come up with a proper test setup the results may not be useful. If the test setup has inconsistent compressor loading due to variable temperature and humidity we won’t learn much. A possible test setup could be to install a vent free natural gas or propane heater into an area that is reasonably sealed and boil water to further increase the humidity. Obviously this won’t work if the air conditioner is installed into a RV.
      - Mike Sokol October 7, 2020 At 8:04 am
        
        Yes, I’m well aware of that. However, I’m not calibrating nuclear guidance systems here (Yes, I used to do that in the 80’s), so general trends with this test could point to a more detailed study. If that is indeed the case, then what you suggest would be indicated.
WILLIAM H CARVER October 4, 2020 At 12:42 pm

The Power company uses voltage regulators on the primary of distribution transformers which are nothing more than a voltage control auto transformer.

- Mike Sokol October 4, 2020 At 2:36 pm
  
  Yes, autotransformers have been used for many years to adjust power company voltage. I used to install buck/boost transformers all the time for industrial power. Very simple things, but vital to delivering the correct voltage.