How to fix motor starter circuitry?

Is this part a


  • Total voters
    5

Thread Starter

bobcart

Joined Jul 7, 2011
55
The heater broke and I diagnosed that it was the blower motor. It's a 120V single phase brushless motor and replacement cost is about $1400! I took it out and got it to run by turning it by hand while power was supplied. This led me to think it might be a centrifugal switch or starter cap. I took it apart and found that there was a burned area behind what looked like a ceramic disk capacitor or perhaps a varistor of some sort. See the image of it attached.

It shows about 14 ohms resistance at room temp. I get an "out of range" on the DMM when testing capacitance. After poking around, I'm thinking it may be a Surge Guard Thermistor by RTI. Here's the link: http://www.rtie.com/ntc/surggard.htm

Also attached is a photo from their site. The logo on the two parts looks kind of similar, but the resolution is low. Also, I can't find the same product code in their listings, though this one could be obsolete. The SG might be Surge Guard and the 421 might be 420V RMS if one were using capacitor style coding. I called RTI and couldn't get anyone on the line or to call back.

So, what do you think this thing is? The circuit has one end from a big power diode. If it's just some over voltage protection, could it be related to my start problem? How can I test it to know? I can't afford $1400 and am hoping to get anyone to help send me in the right direction.

Also, attached are two photos of the motor driver. It looks like a 6 transistor H-bridge of some sort. The three big electrolytic caps are wired in parallel. A couple of ICs including a LS7262 brushless DC motor controller (the board is fed AC at 120V so maybe there are rectifiers in there and not mosfets) and a second IC that is unmarked. The rest are diodes (a ton of them) and a bunch of resistors, three big caps and few smaller ones, a speed control pot, the motor coils and power jack. The motor is variable speed, but has no outside controls, so I'm wondering how it senses the need to speed up or slow down. Might be a thermistor? If hot, turn up fan, if cold, slow down? If that is it, I don't think that would solve my dilemma.

Any advice on what to do appreciated. I'm about stumped.

Thank you in advance!!!

Bob
 

Attachments

strantor

Joined Oct 3, 2010
6,782
The heater broke and I diagnosed that it was the blower motor. It's a 120V single phase brushless motor and replacement cost is about $1400! I took it out and got it to run by turning it by hand while power was supplied. This led me to think it might be a centrifugal switch or starter cap. I took it apart and found that there was a burned area behind what looked like a ceramic disk capacitor or perhaps a varistor of some sort. See the image of it attached.

It shows about 14 ohms resistance at room temp. I get an "out of range" on the DMM when testing capacitance. After poking around, I'm thinking it may be a Surge Guard Thermistor by RTI. Here's the link: http://www.rtie.com/ntc/surggard.htm

Also attached is a photo from their site. The logo on the two parts looks kind of similar, but the resolution is low. Also, I can't find the same product code in their listings, though this one could be obsolete. The SG might be Surge Guard and the 421 might be 420V RMS if one were using capacitor style coding. I called RTI and couldn't get anyone on the line or to call back.

So, what do you think this thing is? The circuit has one end from a big power diode. If it's just some over voltage protection, could it be related to my start problem? How can I test it to know? I can't afford $1400 and am hoping to get anyone to help send me in the right direction.

Also, attached are two photos of the motor driver. It looks like a 6 transistor H-bridge of some sort. The three big electrolytic caps are wired in parallel. A couple of ICs including a LS7262 brushless DC motor controller (the board is fed AC at 120V so maybe there are rectifiers in there and not mosfets) and a second IC that is unmarked. The rest are diodes (a ton of them) and a bunch of resistors, three big caps and few smaller ones, a speed control pot, the motor coils and power jack. The motor is variable speed, but has no outside controls, so I'm wondering how it senses the need to speed up or slow down. Might be a thermistor? If hot, turn up fan, if cold, slow down? If that is it, I don't think that would solve my dilemma.

Any advice on what to do appreciated. I'm about stumped.

Thank you in advance!!!

Bob
So all it has going into it is 2 wires for 120V power? no inputs from the thermostat? If so, then All this circuitry seems overkill; I would guess it is to make the DC motor spin at a constant speed regardless of positive or negative pressure. so why not just replace it with an AC motor? you can probably get one <200$
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
There is also an equipment ground, but only the two leads from the supply source and no other inputs at all. The manufacturer calls it a variable speed motor. Any such sensors are directly on the board. I'd prefer to actually fix this board rather than try to replace it. I am interested in motor controllers in general and this is a great opportunity to learn more. any clues about the passive device I reference or how to troubleshoot appreciated. Also, any clues about where else I might look for info would be appreciated.
 

strantor

Joined Oct 3, 2010
6,782
I'd prefer to actually fix this board rather than try to replace it. I am interested in motor controllers in general and this is a great opportunity to learn more. any clues about the passive device I reference or how to troubleshoot appreciated. Also, any clues about where else I might look for info would be appreciated.
ok then, lets see what we can see

I took it out and got it to run by turning it by hand while power was supplied. This led me to think it might be a centrifugal switch or starter cap.
Shouldn't be any starter cap if its actually a DC motor.

I took it apart and found that there was a burned area behind what looked like a ceramic disk capacitor or perhaps a varistor of some sort. See the image of it attached.

It shows about 14 ohms resistance at room temp.
If it's inded a varistor and it shows 14ohms then I would say it's blown. varistors should read very high resistance if they are good. You should remove it to test it, as you might be reading 14Ω across something else in the circuit. where is it in the circuit? I would expect a varistor to be found across L & N pretty close to where the the incoming 120V terminates.

If it is a varisor and it is 14Ω then it is likely shorting out your 120V before it can get to the rest of the circuit to power the motor. you will know if it is super hot.
 

debe

Joined Sep 21, 2010
1,389
Hi Bob, If you lookup the data sheet for that chip there is a lot of in fo on how it works. That chip drives transistors for coil switching & prabably uses Hall devices to sense where the rotor position is. The disc is probably an NTC resistor for surge protection if its reading 14ohms. First thing i would be looking at is solder joints through a strong magnifying glass for poor joints, pictures not quite good enough to see properly.
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
Great idea to check the data sheet on the IC. There is a device that looks like a hall chip I've used, though it could be a transistor. I'll study it. There is also a line painted on the motor bell suggesting a mark for a pole for hall testing. If it is a NTC resistor, does 14ohms seem low? you can see the burn mark on the back side of the PCB to the right where I removed it. I assume that if it is for surge and burnt that it is defective, but can't understand why that would cause motor to not start. maybe the data sheet on the controller will help. by the way, the soldering is really well done and heavy with potting compound on top of that. I've looked pretty closely and not seen any solder issues. only thing I saw was the burn marks... thanks for your help!
 

debe

Joined Sep 21, 2010
1,389
I wouldnt be to worried about the device as they run fairly warm & that is probably why it looks burnt, doubt it is the start problem. Most brushless DC motors have at least 2 hall devices to sense the poles of the rotor magnet for switching.( Bit like a normal motors comutator)
 

strantor

Joined Oct 3, 2010
6,782
I assume that if it is for surge and burnt that it is defective, but can't understand why that would cause motor to not start.
... it is likely shorting out your 120V before it can get to the rest of the circuit to power the motor.
did you by chance try running with it removed? I'm not going to advise you do that because I still don't know what it is, but theoretically if it is for surge protection and if it were removed, the circuit should still function, just without protection. If it is for surge protection and it is across your incoming 120V and it's 14Ω, then how much power do you think is left over to go to the circuit?
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
it looks like the max voltage for the controller chip is 28, but source is 120AC. There are plenty of diodes for a diode bridge to convert the DC to AC and the caps might be for ripple. some of the six TO220 style FETs may be voltage regulators too. That could explain a lot of the circuitry and also the heat. the blower is only for exhaust. if the heater thermostat says heat up, and if the pressure differential is good between the gas supply and the exhaust, it closes the switch to the motor. It thus only gets on off signaling and there is never a reason the reverse the direction. The variable speed control is not dynamic. You can set a pot to adjust speed to make the gas pressures work out right (long story there with differential manometers, etc.). So, I basically think this is a simple one direction DC brushless motor controller board with a 120VAC to 24VDC rectifier/voltage converter built in along with a variable speed control. It is either off or running at the full speed it is set to run. I agree my burnt passive device may be a perfectly good NTC resistor and unlikely the culprit. So, I suppose now I need to learn more about brushless motor starting circuits. there doesn't seem to be any secondary winding or inductor of any kind so there isn't any centrifugal switch. The IC and hall sensors can certainly tell motion and speed. Somehow it isn't starting. maybe windings are out of balance? That seems unlikely. Hmmm. More clues???
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
did you by chance try running with it removed? I'm not going to advise you do that because I still don't know what it is, but theoretically if it is for surge protection and if it were removed, the circuit should still function, just without protection. If it is for surge protection and it is across your incoming 120V and it's 14Ω, then how much power do you think is left over to go to the circuit?
I haven't tried that yet. It's quite a job to assemble. it took me a week to get it apart without damaging it in the first place! I had to order a special wheel puller for the bell and make a special set of cir clip pliers for goodness sakes. I'm not quite ready to go through that again just to see if it motor smokes or not. I did consider it though.
 

SgtWookie

Joined Jul 17, 2007
22,230
In the 3rd image, it looks like the solder was cracked away from the component leads. That would cause it to act as if it were removed from the circuit.

I would remove the old solder, clean up the leads of the SG421, and reinstall it using Sn63/Pb37 solder to see if the motor works. I don't think 14 Ohms is unreasonable to limit inrush current during start-up; you'd get a max of ~8.6A RMS through it - say, to charge up those capacitors.

There are similar-type part numbers on this page:
http://www.rtie.com/ntc/sgspec.htm
however, your specific part number, SG421, is not listed. Further, I can't discern a pattern to how those SG part numbers were assigned; they seem rather arbitrary. If you can't get ahold of RTI, then try the manufacturers' reps, listed on this page:
http://www.rtie.com/salesrep.htm
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
The photo is after removing the SG421, but I stuck it back in there to show position. Good idea on trying a rep. However, I have a feeling the SG421 isn't the problem in the first place. it's solder was good before I removed it. If I put it all back together and it worked, I won't know why. If If doesn't, I'll only know it wasn't the solder joint of the SG421. I feel like something else is amiss. If the SG421 is an NTC thermistor, I should be able to see if resistance drops with increased temp. I'll try it. I'll also study the circuit to see if I can tell if it is in the right place for surge suppression or controlling speed. If the idea is for the motor to speed up if it gets hot, that might make sense for an exhaust blower. I'm thinking the variable speed capability of the blower assembly may not be used dynamically, but rather at each furnace install due to different pressures of different ducting, etc. In that case, this may be start surge device. Otherwise it may be the NTC thermistor. I cannot imagine why you'd use a NTK thermistor to suppress start surge. We're making progress! Thank you!
 

shortbus

Joined Sep 30, 2009
10,045
Your motor is a three phase motor. They take the 120V AC, rectify it to DC, and invert it back to three phase. It is basically a motor with a built in VFD(variable frequency drive). A lot of the newer energy saving appliances use this. I'll bet some where on your furnace there is a "Energy Star" sticker or rating label.

More and more of the companies making fractional HP motors are going to this. The motors adjust them selves to the load and control speed internally.

Fisher and Paykel was one of the first to start doing this.

If it was me, I'd start by checking the diodes for the rectifier.
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
Okay, I've determined a bit more. I tested the SG421 with temp. Increasing temp dramatically reduces resistance. It went from 16.2ohms to 10.4 ohms. In reading more about the "surge guard" products, this all makes sense. When the motor controller is first switched on, the bridge diodes need to charge the big caps and that requires a lot of energy that can toast the diodes. Once the motor is hot, the caps are certainly charged so the resistance is no longer needed to protect the surge. Now in looking at the specific label on the motor cover, it says it is "thermally protected L fuse protected". I now get the thermal protection, but haven't found a fuse anywhere. Such a fuse could be for start, I suppose, but where is it???

The motor cover also says "120V 50/60Hz 2.0A Max 5000 RPM". The 6 TO220 devices are three pairs of N and P channel MOSFETs IRF730 and IRF9640s. Perhaps the three phases? A single phase AC supply is bridge rectified and used with the motor controller IC to control the motor's speed and direction? Each mosfet pair has a rectifying diode. The mosfets are on a shared heat sink that has what must be another little thermistor attached - more thermal protection. I followed the lead

I tested the rectifying diodes on the mosfets with my multimeter and they are each fine. I found what must be the diode bridge and they use the same rectifying diodes as the mosfets and all tested fine.

There is a big axial resistor labeled 4.7k. I tested it and it's fine too. There is also a devide that includes what appear to be three not two HALL sensors, but looking up the ID, I can't find anything. They are positioned right under the stator edge. I can understand the need for two for quadrature direction, but not sure why three. There are 6 other small transistor looking devices scattered around and not looking like hey are in the right spots to be the hall sensors.

Not sure what to look at next, but I'm going to solder the SG421 back on. I may also reassemble and test the whole thing to make sure it remained broken in the same way.
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
I put it all back together powered it up and it started like a champ. I'm not fooled, because the first time I took it out and put it back in the same thing happened. Yes, I've checked the 120AC supply lines. The thing seems intermittent at best. I have tested it where it wouldn't start in my shop too. Giving it a bump seemed to have worked once, but not again. I know, you are thinking bad solder somewhere, but I sure can't find it. I'll put it back in the main heater again to see if I learn anything new there. Thanks for your persistence in the face of mine!
 

Thread Starter

bobcart

Joined Jul 7, 2011
55
I suppose I have good news to report. I hooked up the motor/blower assembly back into the heater and it worked. First time it didn't work, but then I did a lot of on off testing and it worked every time. Perhaps I need to let it cool down and try it - or warm up. I'm skeptical that my work actually resulted in any long term fix. You'll hear back from me if I'm right. If not, cheers and thank you for following this!
 

#12

Joined Nov 30, 2010
18,224
This is one of my pet peeves. A $1400 fractional horsepower motor to save energy. How long will it need to run to save $1400 in electricity cost? I'd redesign that heater in a New York minute, but I can't recommend that for you for legal reasons and the fact that you said you want to explore the motor until you understand it.

calculating...10.24 years at 24 hours a day if it saves one amp of 120 volt power. You can estimate how many hours a year it actually runs.
 

strantor

Joined Oct 3, 2010
6,782
Your motor is a three phase motor. They take the 120V AC, rectify it to DC, and invert it back to three phase. It is basically a motor with a built in VFD(variable frequency drive). A lot of the newer energy saving appliances use this. I'll bet some where on your furnace there is a "Energy Star" sticker or rating label.

More and more of the companies making fractional HP motors are going to this. The motors adjust them selves to the load and control speed internally.

Fisher and Paykel was one of the first to start doing this.

If it was me, I'd start by checking the diodes for the rectifier.
This is one of my pet peeves. A $1400 fractional horsepower motor to save energy. How long will it need to run to save $1400 in electricity cost? I'd redesign that heater in a New York minute, but I can't recommend that for you for legal reasons and the fact that you said you want to explore the motor until you understand it.

calculating...10.24 years at 24 hours a day if it saves one amp of 120 volt power. You can estimate how many hours a year it actually runs.
So, what is the point of making the motor variable speed? how does that save energy?
 

shortbus

Joined Sep 30, 2009
10,045
Until the energy star motor, the motors for furnace blowers had three to as many as six windings in them for different speeds. A lot of copper and a bigger motor. At installation the tech would set the speed to suit the size of the duct work. Longer runs faster speed to keep flow correct at all registers. Smaller house lower speed. Air conditioning use instead of heat and higher speed to 'cool' with a higher actual evaporator temp.

In washers and driers (where this type motor was used first). A small load takes less HP to turn the machine so it adjusts to a lower actual amperage use .

The energy savings is just not in the electrical use during its life time. Its also about the amount of copper in the motor, silicon chips are more energy conserving to make than a motor :)

Plus, it makes you buy a new machine when you see the estimate to fix the old one. Except for idiots like us. Think about your car, how many "shade tree mechanics" will try to fix one of the new electronic rich cars? Beside me that is.
 

shortbus

Joined Sep 30, 2009
10,045
@ Bobcart- it may act up again, the probing around you did checking every thing may have moved a bad trace on the board so it made contact again.

I fixed a friends window air conditioner last year that had a copper trace that came loose from the surface of the board. He could get it to work some times by hitting it on the side of the case. The trace would stop conducting when it moved apart. Soldered a short piece of wire between both end of the loose trace and works fine now.

So if you have to take it apart again, look real close at all the traces. Not just the solder joints.

By the way a loose trace is not to be confused with our AAC court jester Loosewire:)
 
Top