Author Topic: Heat pump reverse engineering and ground reference problems  (Read 2274 times)

0 Members and 1 Guest are viewing this topic.

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Heat pump reverse engineering and ground reference problems
« on: November 25, 2017, 12:43:04 pm »
Hello Dave and the EEV community!

I've got a problem relating to mains and reference ground in my electrical project.  Thing is, I don't understand how I can get 240VAC potential between two grounds.  Any ideas??  Here are the details of the problem.

https://youtu.be/fpdHFLBdKP0
 

Offline jaycee

  • Regular Contributor
  • *
  • Posts: 206
  • Country: gb
Re: Heat pump reverse engineering and ground reference problems
« Reply #1 on: November 25, 2017, 12:58:41 pm »
The most plausible explanation is that you dont have 240VAC to ground, but rather 240VAC between two phases of a 120-0-120VAC "pole pig" or such like.
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #2 on: November 25, 2017, 01:55:00 pm »
Ah, video is finally uploaded and you can see my issue now!

Also, I did go and verify that the furnace frame is connected to earth ground by comparing voltages with a nearby outlet on the meter.  As expected, ground lead showed zero, a neutral (white) wire showed zero, and the hot wire (black) showed about 120V.

k.
 

Offline jaycee

  • Regular Contributor
  • *
  • Posts: 206
  • Country: gb
Re: Heat pump reverse engineering and ground reference problems
« Reply #3 on: November 25, 2017, 02:16:15 pm »
The wire to ground connected to the frame is NOT a return path for current, it is a safety ground designed to blow a fuse/breaker if there is a short. Whenever you see the safety earth symbol it is only intended for use as a path for fault current, or for suppression capacitor ground.

So, the unit is designed for 240VAC power. In your 120VAC country, it is quite common for high loads like this to be run by using 2 phases using a Scott T transformer or so. Each phase will be 120VAC with respect to ground, but they are 180 degrees out of phase between them, so you will measure 240VAC across them.

In a 240VAC country like mine, L1 would be the "live" and L2 would be connected to neutral. This is why you can see L and N used in some of the wiring diagrams - if it was wired to a 240VAC system, this is how it would be wired. It's also why you see the Brown/Blue or Red/Black wiring in places - this is what we use for Live/Neutral in UK & Europe.
 

Offline jaycee

  • Regular Contributor
  • *
  • Posts: 206
  • Country: gb
Re: Heat pump reverse engineering and ground reference problems
« Reply #4 on: November 25, 2017, 02:28:30 pm »
I did some further reading into what sort of power outlets are available in Canada. You may be familiar with "Electric Range" outlet points which have 4 terminals - two are hot phases, one neutral, one safety ground. The two hot phases on these outlets are 120V AC with respect to neutral, but they are also 180 degrees out of phase with each other, so you get 240V AC between them. This is used to more efficiently run the electric heating elements, as they are connected across the two phases to get 240V AC. The neutral is then used with one of the phases to produce a 120V AC power for other things such as oven timers.

I suspect that your heating furnace is using the same principle, and that HVAC (air conditioning) systems in your country can also use this power scheme, as it is more efficient than using 120VAC (which means double the current draw, and much larger cable thicknesses needed as a result to handle the current and prevent excessive voltage drop)
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #5 on: November 26, 2017, 12:11:35 pm »
thanks for insight. i think i understand. electronics of unit dont use neutral return at all (neutral/ground is only connected to chassis for safe return path in case of shorts). makes same circuit good for 240/220VAC supplies.  I did see optio isolation between motor circuit and controller  circuit so eventual cnxn to RS485 bus should be fine. MAX485 adapters also have optio isolation anyhow. my only issue is with this motor circuit, must be careful when measuring! If i switched L1 and L2 at supply to unit would my meter read -240VAC? Or, would it be closer to zero?
« Last Edit: November 26, 2017, 12:19:48 pm by kevinateev »
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #6 on: November 26, 2017, 12:26:49 pm »
FYI, the furnace is natural gas fired so it only uses one 120vac circuit to power control system and blower motor. If it were electric heat, it would definately use 240vac as the heat pump does.

Also, where we are right now, economically in this time, it is cheaper to run gas furnace for heat. Back in 2000, electric heat pump was cheaper. Tomorrow? who knows!
« Last Edit: November 26, 2017, 12:31:10 pm by kevinateev »
 

Offline jaycee

  • Regular Contributor
  • *
  • Posts: 206
  • Country: gb
Re: Heat pump reverse engineering and ground reference problems
« Reply #7 on: November 28, 2017, 08:39:48 pm »
thanks for insight. i think i understand. electronics of unit dont use neutral return at all (neutral/ground is only connected to chassis for safe return path in case of shorts). makes same circuit good for 240/220VAC supplies.  I did see optio isolation between motor circuit and controller  circuit so eventual cnxn to RS485 bus should be fine. MAX485 adapters also have optio isolation anyhow. my only issue is with this motor circuit, must be careful when measuring! If i switched L1 and L2 at supply to unit would my meter read -240VAC? Or, would it be closer to zero?

It's important not to think of neutral and ground as the same thing. Even though they are often returned to a common ground rod/neutral terminal on distribution transformer, on your property you always regard them as seperate. Neutral is for return path of current from the load(s). Ground is return path of FAULT current only.

Switching L1 and L2 will make no difference, and you will still read 240V because it is AC voltage, which goes negative every half cycle anyway.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Heat pump reverse engineering and ground reference problems
« Reply #8 on: November 28, 2017, 09:01:49 pm »
In my experience, central AC/heat pump outdoor units are universally 240V, as are electric resistance furnaces. Gas furnaces are typically 120V and air handlers for heat pumps are 240V as they also contain several kW of electric backup heat.

This is referring to residential equipment in the US, commercial stuff is often 3 phase 208 or 480V. Red and black are typically used as the two live sides of a 240V circuit. White is always neutral and ground is either bare copper or green. IIRC any color other than white or green (or bare obviously) is acceptable for use as live but anything other than black and red are rare in residences.
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #9 on: November 29, 2017, 01:43:49 am »
Well, in that case, I'm extra keen to get this motor off this circuit.  To this point, it has just been sitting on the metal case of the furnace which is ground.  The base has rubber so it has been OK but I'd bet there would be a big POP if the shaft touched the tin of the duct-work.

That Hantek DSO has two-channel so I imagine I can use the two of them without ground leads to measure between the Vgnd and Vsp.  I was hoping to use one channel for Vsp and one for PG so that it would be easy to see the phase between them.

The MAC has isolated power-supply so I could theoretically use the ground wire of the probe.  But really do I want to?!  No, I'm afraid of destroying my new 800$ MAC.

Anybody know what the phase typically is between a PCM signal and a feedback signal?  Perhaps it is common enough I could just presume what it is?  The PG signal I'm expecting to be 2 or 4 pulses per rotation.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Heat pump reverse engineering and ground reference problems
« Reply #10 on: November 29, 2017, 11:59:40 pm »
I must have missed something, why would the motor shaft touching something cause a pop? The motor frame (and indirectly the shaft) should be earth grounded, as should the duct, so no potential difference between them. If one is floating you may read a significant voltage due to capacitive coupling but no exposed metal should ever be live. If it is, the first thing I'd do is open the terminal box and look for chafed wires.
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #11 on: December 01, 2017, 03:07:52 am »
This motor is ment to be in a plastic inside unit case of a ductles heat pump.  i would have expected Vgnd to be ground too! this is why i made the video.

I'm using the electronics board for its microcontroller, thermocouples etc, but not motor.
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #12 on: December 03, 2017, 12:02:33 pm »
So back to my goal here, i'm still nit sure how to proceed. The circuit i want to measure is DC of only 15 V - 35 V. But, there is a 240VAC offset. My DSO only has 8bits of resolution. So, if i used 100x probes i loose much fodelity.

can i remove this bias somehow?
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #13 on: December 03, 2017, 01:09:13 pm »
For instance, this guy figured out how to get good measurement with a reference ground for measuring 3-phases of a motor.

https://youtu.be/yY3VHfxVnJA

I'm thinking i can put a 1MOhm resistor and 10microF capacitor to my Vgnd lead and connect probe ground leads to that.

Note: with hvac unit powered off, there is no continuity from Vgnd to Earth so the 240VAC potental i'd measured is at least switched.
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #14 on: December 06, 2017, 12:53:27 pm »
problem solved.  dont use Hantek DSO and bought a BK-2542 100Mhz DSO. it can handle 400V p-p and has AC coupling. overkill, i know but u take what u can get at auctions.

That and take motor out of circuit for further testing with bench power and signal generator after some basic measurements in circuit.
 

Offline kevinateevTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: ca
Re: Heat pump reverse engineering and ground reference problems
« Reply #15 on: December 30, 2017, 03:47:23 pm »
Hey Y'all, so I've solved the immediate question.  The reading on the volt meter.  The switching power supply is quite noisy and I'm measuring on AC setting.  This meter is too dumb to tell me when I'm using it incorrectly so it is simply giving a false reading.  When I switch the volt meter to DC measurement, I get a more expected -120VDC.?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf