AUX2 Buzz

[keyturbocars]

When we were discussing AUX2 and using the word "buzz" to refer to the PWM output, man were we not kidding!!  It literally buzzes! 

I took the video (link below) yesterday when there was a light wind (10-13mph) and the Classic was diverting power to my hot water heater.  The buzz is coming from a Supco snap disc thermostat that I have wired in series with the AUX2 output.  I have it wired this way so if the water gets too hot, then the snap disc switch opens the circuit and shuts off the diversion.  This will probably rarely ever happen in my household.  In any case, the buzz does not appear to be the contacts bouncing in the snap disc switch.  The switch appears to just be ON.  The PWM signal just appears to be resonating the whole case of the switch.  Since the switch is pressing against the water heater tank, then it sort of gets transferred and amplified on the water heater.  In this video, it doesn't sound too loud, but tonight it sounds louder.  Tonight the winds are averaging around 15mph.  I actually don't mind an audio indicator to know when the Classic is diverting.  My utility room is my office, so I can sit at my computer and hear what is going on with the water heater diversion.  I could just deal with a little lower volume.  It's a little annoying at times (almost makes my eardrums resonate at times).   

http://www.key-ideas.com/AUX2-PWM-Buzz-1.mpg

Sharing the buzz...

Edward

[Westbranch]

sounds more like there is a telegrapher inside it, sending Morse code!

[keyturbocars]

You are right Eric!   At lower wind speeds, it sounds more like morse code.  When the winds hit about 18mph, then it turns from morse code into a high pitch screeching sound that is very annoying.  It's like fingers on a blackboard.  I'm getting over a flu and my head (and ears) still has some congestion and somehow my ears are more sensitive to noises like this.  It almost feels like my eardrums are resonating when the winds pick up and the screeching pitch increases.   

boB (or others), Any ideas of how to electrically dampen this buzzing/screeching sound (apart from removing the snap disc switch from the circuit all together)?

I have a sound insulating blanket that I plan to try to wrap around the front of the water heater.  That should help and might just be good enough.  Just curious if anyone has any electrical ideas.

Edward

[boB]

Edward, can you tell exactly where the sound, buzz, noise is emanating from ?  Load resistors ??

Is this fairly recent software ?

Maybe I can make it a bit better.

boB

[Westbranch]

boB, Edwards mpg tells it all...

Eric

[boB]

Thanks ! I forgot to watch that...

If that sounds like Morse code, there are no dashes !  Just dits !
Sounds more like a Geiger counter to me....

I can't tell where the sound is actually coming from though.
I think I am looking at a snap dome thermal switch in the MPG video ??

So, Edward, let me get this straight...   You are using Aux 2 PWM to drive
the 3 channel AC TRIAC on the AC side of the turbine ??   With or without
resistors ??  I can't remember...

The way things are working right now, I would not run PWM diversion from Aux 2
into an AC side of the rectifier because PWM as is coming out of the Aux 2 terminal
for diversion does not know anything about where the turbines' AC cycle is at.
That PWM output should really be used for a DC SSR on the DC side and a big
load resistor.

In addition to the PWM on the DC side with a DC SSR and load resistor, it really
needs a good Cache diode to stop the high voltage spikes that come from the
inductance of the load resistor.  I think I forgot to mention this.

For now, on the AC side and 3 triacs and resistors, Aux 1 should be used in a
slower   Bang-Bang method and that should stop the pulsing like we hear in that video.
i.e.  Attack or Delay Time can be short, but have at least a few hundredths of a second
Hold time so it doesn't tend to ping and pulse.  Now that the Aux diversion software
uses peak voltage sensing, it has the capability to pulse and buzz more if the time
is too fast.  Shouldn't be too much of a problem with a DC SSR and loads, but
there may be some imbalances on the AC side diversion because the Aux 2 PWM
output will sometimes tell the Triacs to fire and odd times in the turbines' ac cycle,
which I ~think~ may cause some of this noise.

If the AC triacs are triggered slowly and just stay ON for many cycles as with Aux 1
output, then the complete AC cycle will be loaded and only (maybe) the first AC
cycle from the turbine might  trigger somewhere in the middle of a cycle but
should be much more tolerable to the ear.

Are you sleeping next to your Classic ??   I know that some people tend to
sleep with their charge controllers they love them so much....  And when they
make noises during the night, they can't get back to sleep.

boB

[keyturbocars]

I can't tell where the sound is actually coming from though.
I think I am looking at a snap dome thermal switch in the MPG video ??

Yes, it's coming from the area right at and above that snap disc switch.

So, Edward, let me get this straight...   You are using Aux 2 PWM to drive
the 3 channel AC TRIAC on the AC side of the turbine ??   With or without
resistors ??  I can't remember...

No, I'm using AUX 2 to drive a 100A DC SSR which diverts 75A to hot water heating elements.  I have that snap disc switch wired in series with the control signal going to the DC SSR.  The PWM signal somehow seems to "excite" the snap disc switch body and transmits the vibrations to the water heater.  It's really not a big deal.  It's just that I work at my computer in my "office" (utility room) and the buzzing is literally right behind my head a few feet to the water heater.

The way things are working right now, I would not run PWM diversion from Aux 2
into an AC side of the rectifier because PWM as is coming out of the Aux 2 terminal
for diversion does not know anything about where the turbines' AC cycle is at.
That PWM output should really be used for a DC SSR on the DC side and a big
load resistor.

OK, good.  That's the way I have it set up now.

In addition to the PWM on the DC side with a DC SSR and load resistor, it really
needs a good Cache diode to stop the high voltage spikes that come from the
inductance of the load resistor.  I think I forgot to mention this.

Thanks for mentioning about the Cache diode!  Could you please point me to a good diode that I could purchase for this purpose?  I don't want to fry another DC SSR!

Are you sleeping next to your Classic ??   I know that some people tend to
sleep with their charge controllers they love them so much....  And when they
make noises during the night, they can't get back to sleep.

Hahaha.  Yes, I could run a set of long 4 guage extension cords up to my bedroom, so I could snuggle up with my Classic at night.  The warmth generated from the FETs could keep my cozy and warm at night.  Just so long as I don't drool into the vent and SHOCK get an abrupt wake up call!

I do have the latest software, and everything is working great with my Classic.  AUX 1 AC Clipper diverts to AC heating elements to slow the turbine and keep input voltage under control.  The AUX 2 diversion using the DC SSR to water heating elements working very well too (now that I replaced my first burned out DC SSR).  I honestly think I can take care of the noise by wrapping the front of the water heater with a sound absorbing blanket.  I use them over my air handler from my heat pump and they drastically cut down noise on that.  Again, it's because my utility room is my "office".  Between the sound of the air handler blower motor, the occasional loud noises from my water softener, and now my "singing" water heater.... it can get quite noisy in this tiny space.  Combined with the constant ringing in my ears (tinnitus), and the the noise from 7 kids in the house, it all makes for quite the orchestra of sounds in my head!!     

Edward

[boB]

OK Edward....

  I will get you info on the Cache diode and how I hook it up with a separate large-ish capacitor (electrolytic) wired very close to the DC SSR to snub those spikes.  It may very well be what is helping to make the noise louder than it should AND more importantly, what may be contributing to the demise of the SSR itself.  I measured about 800 Volt spikes whenever the DC SSR turned off.

The capacitor is necessary and has to be close because of the inductance in the wires themselves that hook up the diode and capacitor.
I used a 315 Volt  800 microFarad or so cap from a Classic.  You may want to use a couple of capacitors in parallel.  I don't think it has to
be quite that large of capacitance most likely.

I am in Florida at the moment and will look at the diode as soon as I can.  I get back on Thursday night.

boB

[keyturbocars]

Sounds good boB.  I definitely want to add the cache diode and capacitor to my circuit.  At $130 a pop, replacing the DC SSR's can get expensive!

Enjoy Florida.  Good time of year to be there (not too hot yet).  I lived in Orlando for a few years. 

Edward

OK Edward....

  I will get you info on the Cache diode and how I hook it up with a separate large-ish capacitor (electrolytic) wired very close to the DC SSR to snub those spikes.  It may very well be what is helping to make the noise louder than it should AND more importantly, what may be contributing to the demise of the SSR itself.  I measured about 800 Volt spikes whenever the DC SSR turned off.

The capacitor is necessary and has to be close because of the inductance in the wires themselves that hook up the diode and capacitor.
I used a 315 Volt  800 microFarad or so cap from a Classic.  You may want to use a couple of capacitors in parallel.  I don't think it has to
be quite that large of capacitance most likely.

I am in Florida at the moment and will look at the diode as soon as I can.  I get back on Thursday night.

boB

[Volvo Farmer]

This is just an uneducated comment from the peanut gallery, but are you putting DC through one of those snap switches?  I fix appliances for a living and those things are commonly used in clothes dryers and ovens, but I have only seen them used in AC applications.  I just wonder if like switches and breakers, there would be some sort of concern for welding the contacts with a DC arc, should the thing open up under load and not be rated for DC.

[Halfcrazy]

Good catch Volvo. I do think that he has it in the low voltage/current control wire from the Classic though not the High current load line?

[keyturbocars]

You are right guys.  That snap disc switch is rated for AC.  I think if memory serves me right, it's rated for around 15A max AC current. 

I'm only using it in the control circuit and so the current is very low (probably on the order of 250mA).  In my previous setup, I was using one like this for around 1 year with low current DC and it worked perfectly.  I think the key is the very low current.  I agree that at higher currents, the DC would me much harder to "break" and could fry the switch contacts.

Edward

[Volvo Farmer]

Ahhh, Okay, I misunderstood, I thought he had the current for the elements going through the thing.

[keyturbocars]

boB,

I was thinking about my AUX2 controlled DC SSR, and I wonder if my situation is different than what you tested.  I am using my DC SSR to divert directly off the battery bank, so the voltages are clamped to my 48V battery bank.

When you did your testing and found voltage spikes of 800V when the DC SSR shut off, were you diverting off the DC rectified output before it went into the Classic?

If so, then I wonder if the unregulated DC voltage was much higher coming off the turbine and if that is partly why you had such large voltage spikes when the DC SSR turned off. 

If you think that a cache diode and capacitors would be best for my AUX2 DC SSR diversion, then that's what I want to do.  You're the electronics expert! 

I know just enough (not much) to get myself in trouble! 

Edward

Edward