Adapt 120V switching between 4 different wires to 0-10VDC signals?

[ddbear]

Circuit ideas?  I would like to figure out a way to retain use of the fan speed, on-timer, fan filter reminder button features on the vent hood while installing an inline duct fan in the attic to replace the built-in fan (Miele DA5100D).  The new inline Fantech EC motor (FKD 14 XL EC) accepts a 0-10VDC signal for controlling the fan speed. 

As a starting point I think I can use the 120V wires from the motor control board (EPL755) that controls 4 levels of fan speed: when wire 1 has 120V, the fan is on the lowest speed, wires 2 and 3 medium speed, and wire 4 receiving 120V drives the motor at the highest speed.  Then I would need to convert these signals to 0-10VDC for the EC fan speed control on the new Fantech motor.

In other words, a circuit for conversion of:

120V on wire 1: 2VDC on 0-10V signal wire and 120V power to motor
120V on wire 2: 4VDC on 0-10V signal wire and 120V power to motor
120V on wire 3: 7VDC on 0-10V signal wire and 120V power to motor
120V on wire 4: 10VDC on 0-10V signal wire and 120V power to motor
0V on all 4 wires: no power to the motor

Is there a circuit product that already exists for programming this kind of conversion?  Or would I need to make a custom circuit?  It would be nice if the four 0-10VDC voltages would be adjustable later (via programming or easy to swap in electronic parts) in case I decide to change the corresponding fan speeds in the future.

[bdunham7]

I'm not aware of any ready-made solution to your issue, but there may be one.  The manufacturer or seller of the new motor might know.

The basic idea would be to use a voltage divider with a voltage follower as a buffer, using a DC source, perhaps 24V,  that either exists now or that you add.  The lower resistor of the voltage divider would be across the voltage follower input and would represent your target voltage, the upper resistor would be different for each of the 4 inputs.  You'd then have to figure out how to translate those 4 120V inputs into an on-off control of the 24VDC (or whatever) that is applied to each resistor. 

If you are unwilling to modify the current board, then I'd just use 4 optocouplers to use the 120VAC to control the lower voltage.  If you don't mind modifying the board, there appear to be 4 relays for the speed control.  Just remove the 120V feed to the those relays, supply them with 24VDC (or whatever you've chosen) and then put the appropriate voltage divider resistor on the outputs.

[Terry Bites]

Use ac detectors such as a HCPL3700, to monitor the lines. Use the logic output transistor to create your DC levels.

[ledtester]

With the HCPL-3700 or any optocoupler you'll need some input protection.

This application note discusses designing detection circuits with the HCLP-3700:

Threshold Sensing for Industrial Control Systems with the HCPL-3700 Interface Optocoupler
Application Note 1004 - Avago Technologies

https://docs.broadcom.com/doc/AV02-3699EN

Specifically, look for "Example 2. AC Operation" which seems to be exactly your use case.

Another option is to simply use some relays with 120V coils.

[ddbear]

Great suggestions so far. 

Another one - my brother, founder/CEO of a diesel engine performance company who designs engine controls all the time, stopped by and offered another idea.  He said it would be simplest and most modular if I use four separate AC to DC converters (with adjustable dials to set the 0-10V), plus four separate relays to activate the 120V power to the motor if any one of the four output terminals get the 120V (having four separate relays provides isolation from each other).  He thought of this idea real quick during a brief visit so I'm wondering if this would work and if there is an aspect we didn't think about deeply enough while considering this?  Also, suggestions for good brands of AC to DC converters with adjustable DC output?

[ledtester]

How about this idea...

I have a feeling the control voltage doesn't need to be low impedance.

Get four DPDT relays with 120VAC coils. On each relay configure one pole to send power to the motor. Thus the motor will have power if any of the four relays are triggered. The other poles will activate the four contacts in this low voltage circuit:



The resistor values are chosen to output 10, 7, 4, and 2 volts when the corresponding switch is closed (from left to right). With no contacts closed the output is 0V.

Of course, you can get by with DPST relays.

[Ian.M]

The fan in question has a 10V output with a factory fitted potentiometer providing the 0-10V speed control voltage.  Use it to determine the control voltages required, then calculate a resistor ladder to output those voltages using fixed resistors  that totals to the track resistance.   Then its just a matter of removing the factory pot and using the relay contacts to connect each divider tap in turn to the control voltage input.

*IF* the existing pot is 10K (or it works with a 10K pot), and the O.P.'s voltages are tested with the fan that would be a 2K bottom resistor between 0V and the 2V tap, a 2K resistor between the 2V and 4V taps, a 3K resistor between the 4V and 7V taps and a 3K top resistor between the 7V tap and 10V, totaling 10K.

Due to the length of the cable run it will probably need a capacitor between the control voltage and 0V at the fan to filter RFI so it doesn't misbehave.  A 10uF 16V electrolytic should do the job.

[themadhippy]

looking at the picture of the control board you might be able to modify  it by cutting a track or 2  to the relays and adding a few bodge wires from the low voltage side

[ddbear]

How about this idea...

I have a feeling the control voltage doesn't need to be low impedance.

Get four DPDT relays with 120VAC coils. On each relay configure one pole to send power to the motor. Thus the motor will have power if any of the four relays are triggered. The other poles will activate the four contacts in this low voltage circuit:

(Attachment Link)

The resistor values are chosen to output 10, 7, 4, and 2 volts when the corresponding switch is closed (from left to right). With no contacts closed the output is 0V.

Of course, you can get by with DPST relays.

So far I like this idea best for its simplicity.  Except I am thinking of using potentiometers instead of fixed in place resistors, so I can always fine tune the voltage settings. 

The Fantech terminal has a 10VDC reference source which I can use (instead of having to use a 12VDC or 24VDC power supply).

I measured the factory pot, it reads 0.2 Ω to 1.645 kΩ.  Given this info what should the specs of the pot be (as I don't think I'll find one with exactly this range)?  I am not familiar with pot selection at all.  Since pots are relatively cheap I figure I will buy the very best.  Any recommendations for the very best pot on the marketplace under $30/each? 

I ordered some Schneider DPST relays (725BXXBM4L-120A) because they have a nice LED indicator and are DIN mount so I can use an enclosure with a rail.   You mentioned DPDT as your first choice, is there a reason that DPDT would offer some benefit over DPST for this application?

(Before I read this post I had ordered four of the DROK DC 5.5-30V to 0.5-30V 5V 12V 24V Adjustable Power Supply Regulator Modules along with four AC to DC 24V DIN rail power supplies, per my brother's idea that I described above, but I may change the design to use pots and the DP relay because that means it's simpler with less parts that can fail eventually.  What I like about the DROK adustable DC converter is that I can see a digital readout of the voltage so I know the values it was previously set to, when playing around with the settings to fine tune the fan speeds.)

[ledtester]

So far I like this idea best for its simplicity.  Except I am thinking of using potentiometers instead of fixed in place resistors, so I can always fine tune the voltage settings. 

I measured the factory pot, it reads 0.2 Ω to 1.645 Ω.  Given this info what should the specs of the pot be (as I don't think I'll find one with exactly this range)?  I am not familiar with pot selection at all.

Are you sure about those readings? Could they instead be 200 ohms and 1.645 K ohms or 1645 ohms?

You mentioned DPDT as your first choice, is there a reason that DPDT would offer some benefit over DPST for this application?

No reason other than DPDT relays seem to be more common. I don't think you'll be boxing yourself into a corner by getting the DPST relays but having that extra throw might come in handy in case your requirements change.

[ddbear]

I fixed the typo, it is 0.2 to 1.645 K ohms.

I am also curious if this is something I should account for:  "Due to the length of the cable run it will probably need a capacitor between the control voltage and 0V at the fan to filter RFI so it doesn't misbehave.  A 10uF 16V electrolytic should do the job." ?

[ddbear]

The nonstandard potentiometer readings are puzzling so I checked with a 2nd multimeter and the readings are identical.  However, I discovered that the max value of exactly 3 K ohms is reached when the dial is at the label 6, and then drops back down to 1.645 K ohms at level 10.  Is this a mislabeling of the PCB that the pot is mounted on?  How can it be that the resistance value peaks at setting 6 and then drops at higher fan speeds?  Impossible physics or mislabeling, or because I can't test it properly without a higher load current?  I took a picture of the pot, and with a close up view was able to ID that it's a Bourns.  Maybe I should buy four of the exact same.

[Ian.M]

Did you disconnect the factory pot before you measured it?  If not, do so as (judging by your readings) the fan is probably shunting it when not powered. Without the total track resistance, measured between the pins that were connected to 10V and 0V, we cannot determine the specs of the pots you need to order. Alternatively it may have a code for its value printed on one of its sides.  Show us that and we can work it out.

The trouble with multiple pots for the different speeds is: to avoid overloading the fan's 10V output, you'd either have to use higher resistance pots so the total parallel resistance of their tracks isn't much lower than the factory one, or make one contact set of each relay switch the 10V power to one pot track, and use its other contact set to connect the control input to the wiper of the same pot.   Personally, *IF* the fan control input behaves normally when using a higher resistance pot, I'd go for paralleling the tracks.

You don't need particularly expensive pots - just about any fully enclosed single turn trim-pots will do if their linear law (the most common) and with a total track resistance that's suitable.  They shouldn't cost you over half a buck each, maybe a buck if you order a premium brand. 

[ddbear]

Aha, it must be the shunting, as I left it connected. 

I was able to get a picture of the microscopic ID numbers on the side of the pot.

(Edit) I didn't think about the issue with powering multiple pots, good catch, sounds like the last piece of the design puzzle to solve.

Maybe this is part of the Bourns 91 long life series?  That's as far as I can decipher the model numbers.  It would be nice to find the version of this which I can mount to a DIN rail adapter rather than a PCB solder type.
https://www.bourns.com/docs/Product-Datasheets/91-95LL.pdf

[ledtester]

The "103" suggests it's a 10K pot.

[Ian.M]

'103' is the code for a 10K pot/preset.   Its actually Bourns 3386P-1-103LF,  which is under two bucks at Digikey, Mouser or a number of other suppliers.  You can also get the same series preset cheaper in a vertical mounting (side adjust) format: 3386B-1-103LF, currently $0.33 + tax at Arrow.

Each 10K preset will draw 1mA from the fan's 10V output.  Its highly probable that the output can supply at least 10mA, and extremely unlikely that it would be overloaded by the 4mA drawn by 4x 10K presets with their tracks in parallel.

Caution - there may not be adequate isolation between poles to safely run mains and low voltage through the same relay.  Check the relay's datasheets *CAREFULLY*.     

[ddbear]

Wow great info to find the specs, thanks Ian!  It sounds like I can power all 4 without being concerned about overloading the 10V reference, since after you found the specs you discovered they draw so little power, if I understand correctly.

The following came to mind when I was originally thinking about my brother's idea with 4 separate power supplies with adjustable DC outputs, rather than DP relays:

Caution - there may not be adequate isolation between poles to safely run mains and low voltage through the same relay.  Check the relay's datasheets *CAREFULLY*.   

I never liked mixing DC in close proximity to AC (when I do home electrical wiring, this is a big no-no).  But for consumer electronics this is done all the time.  Just to be safe I am putting all of this into a contained enclosure.  The relays are Schneider Electric 725BXXBM4L-120A.  I haven't found anything in their datasheets yet to say whether or not it's ok to have AC on one pole and DC on another pole.

But still playing around with the idea to just go with power supplies with adjustable output rather than utilizing the 2nd poles of the relays, just to eliminate any "grey areas" in the isolation.

[ledtester]

Re: isolation -- another option is to double the number relays -- i.e. two SPST relays instead of a DPST.

But still playing around with the idea to just go with power supplies with adjustable output rather than utilizing the 2nd poles of the relays, just to eliminate any "grey areas" in the isolation.

Actually, I had in mind an external 12V power supply to power the voltage dividers.

[ddbear]

How do I identity this PCB to wire connector that is on the Miele fan control board?  It would be nice to find another one instead of having to cut the existing one off the motor.  Here is a close up picture.  It is about 2cm wide. 

[ddbear]

Since having this new interface opens up the possibilities for automation, I thinking I could also solve a problem with people forgetting to turn on the vent fan during cooking.  I could use a heat sensor to also trigger the relay in addition to the buttons.  The heat sensor would be located just inside the hood opening behind the grease filter.

I haven't thought this through, so I wonder if this would be sensible from a user interface perspective.  This means sometimes the fan would turn on by itself without the buttons lighting up on the Miele vent hood.  The Miele buttons would still work normally, just wouldn't be able to override the heat sensor to turn it off (maybe a heat sensor would make it stay on too long?)

I wonder if there are good heat sensors for this type of application?  What is the general principle of how I would connect a heat sensor to the control interface I'm putting together?

Alternatively, I could connect a current sensor inside the junction box in the attic that goes to the 240V induction cooktop.  I wonder which option (heat sensor vs. current sensor) would be better.

[ledtester]

How do I identity this PCB to wire connector that is on the Miele fan control board?  It would be nice to find another one instead of having to cut the existing one off the motor.  Here is a close up picture.  It is about 2cm wide.

Does that connector slide on and off the edge of the PCB? Also, it looks like the edge connector has holes in them. If that's the case, one option is to remove the connector and solder new wires through those holes.

A photo of the PCB with the connector removed would help determine if this is feasible / desirable.

[ddbear]

Does that connector slide on and off the edge of the PCB? Also, it looks like the edge connector has holes in them. If that's the case, one option is to remove the connector and solder new wires through those holes.

Yes, it slides on and off the edge of the PCB.  Soldering would work so I might just do that.  I wasn't sure if this was a very generic connector type.