Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[toasterking]

I recently needed wireless on/off control of a fluorescent lighting circuit running on 277V AC (a single phase of commercial 480V 3-phase service) by a PC.  After weighing all the options (which were very few), and considering that the PC already had a CM19A connected and I was already using X10 RF there, I decided on modifying a TM751.  (There are no other PLC devices running on 277V there, so PLC "rudeness" of the TM751 isn't a concern, the RR501 is too large to fit in my enclosure, and I can live with being stuck with unit code 1.)

This procedure is based on a procedure on Ido Bar-tana's web site for modifying the RR501 for 220V:
http://www.idobartana.com/hakb/modifying_remote_control_transce.htm#Modifying%20110v%20to%20220v
I had to make some minor changes to get it reliable with 277V.  I ended up with a 100Ω 7W resistor for the latching relay coil mostly as a result of trial and error since I don't have all the specs on the latching relay used in the module.

The date code on the TM751 module I modified is 9F25 (25th week of 1999 during June).

   1.   Desolder the 110V plug blades and solder insulated, stranded wires in their place.  Use an appropriate wire size for the load.
   2.   Desolder the 110V socket contacts and solder an insulated, stranded wire in place of the HOT out contact.  (Neutral output is already connected to the neutral input so you shouldn't need an extra lead there.)
   3.   Cut the plastic out of the inside wall of the area of the plastic case where the socket contacts were.  This allows you to route the new wires out through the openings previously used for the 110V socket.
   4.   Optional:  If the TM751 will be mounted in an area where the telescoping antenna would touch a metal surface or the telescoping antenna cannot be used, replace the antenna with a wire.  Follow the black wire from the case's antenna to the RF daughterboard.  Desolder the wire from the daughterboard and replace it with a 20-24AWG wire exactly 9 1/16" (1/4 wave) or 18 1/8" (1/2 wave) in length.  Note that this poses a potential shock hazard if the module will be mounted in a place where this wire could be touched with the power on!  The wire must be well-insulated because it is referenced to the mains.  Add a piece of shrink tubing to the end of the wire to prevent it from shorting to other conductors.  Slide the telescoping antenna out of its pocket in the case and remove the old wire and the metal plate or foil coupler.
   5.   Change the large blue capacitor (2E225) from 2.2µF/250V to 1µf/400V (Digi-Key part # EF4105-ND).
   6.   Change the smaller blue capacitor (2E224) located close to the output connector from 0.22µF/250V to 0.1µF/400V (Didi-Key part # EF4104-ND).
   7.   Check that the 330kΩ resistor, close to the big capacitor, is 1/2W. It should be physically bigger than all other standard resistors. If not change it to 330kΩ 1/2W.  (Did not need to change in my TM751; already 1/2W.)
   8.   Change the 220kΩ 1/2W (in some models 1/4W) close to the output connector to 470kΩ 1/2W.
   9.   Change the 130V MOV near the space that the input plug blades formerly occupied to 281V (Digi-Key part # F4010-ND).
   10.   If there is a resistor of 330kΩ 1/4W going to the output connector, change this one to 470kΩ 1/2W.  (Did not exist in my TM751.)
   11.   Add a 100Ω 7W wirewound resistor (Digi-Key part # 985-1143-1-ND fits nicely) in series with one of the wires for the coil of the cam relay. The resistor could get a little warm, so try to ensure that it does not touch any other components.

My modified TM751 has been toggling its relay 6-10 times daily for 4 weeks and no failures yet!

Edit: Correction to circuit wiring for the MOV.

[Brian H]

Relay is a 56 OHM coil.
Pulsed by a SCR that hits it with a single 120 volt DC pulse. Yes it is a brief 2 amp pulse if my math is not wrong.
The small MCR100-8 SCR has a voltage rating of 800 volts. 800 ma continuous current and 10 amp nonrepetitive single pulse. So it is probably OK in your mod.
The relay coil also has a small 140 volt MOV on it but your tests seem to show it maybe OK.

There is a schematic of the TM751's Main Board on the web site where you found the 220 modifications.

[toasterking]

Thanks, Brian, for the additional info!

Yes it is a brief 2 amp pulse if my math is not wrong. [...] 800 ma continuous current and 10 amp nonrepetitive single pulse. So it is probably OK in your mod.

I originally was using a 100Ω 2W metal oxide film resistor.  It seemed to work initially, but within about 20 activations, the resistor failed internally; not completely open but high resistance in the Megohms.  I tried 2 of them with the same result.  So far, the 7W wirewound resistor is holding up.

The relay coil also has a small 140 volt MOV on it but your tests seem to show it maybe OK.

That may be because I replaced the MOV with a 281V one. 

[Brian H]

I thought there where two MOVs.
Even for X10. No MOV across the incoming power pins is not typical.

[toasterking]

I thought there where two MOVs.
Even for X10. No MOV across the incoming power pins is not typical.

Maybe there are two MOVs in some versions of this model.  Mine only has one.  I do have "130V MOV in parallel with the input" in my procedure, but that part of the text was copied from the original procedure at http://www.idobartana.com.  I will change it to read "130V MOV near the space that the input plug blades formerly occupied".

[Brian H]

I opened a Date Code:10E19. The later hardware version. With the RF Daughter Board that looks like the one in a CM15A.
It has two MOVs in it. One near the relay coil across the SCR and the other one on the AC Input pins.

[toasterking]

I still have the innards of another TM751 with date code 02K46 on my workbench.  Whilst modifying the one with a 9F25 date code and documenting this procedure, I remember comparing the two modules and determining that the circuit layout was identical, so this info applies to the one I modified also.

I can confirm this now.  The only MOV -- the one I changed to 281V -- is across the SCR that pulses the relay coil.  There is no MOV across the AC input pins.  So if anyone attempts this procedure on a TM751 with 2 MOVs, it would be prudent to change them both to 281V or higher (although I'm not sure a higher one would fit; the 281V is already pushing it on space).

[toasterking]

Relay is a 56 OHM coil.
Pulsed by a SCR that hits it with a single 120 volt DC pulse. Yes it is a brief 2 amp pulse if my math is not wrong.

Slightly off-topic now, but Brian, I wonder if you could enlighten me as to how one would do the math there.  I have a new project where I'm toying with the idea of replacing the latching relay in the UM506 with a dual-coil latching reed relay.

The pulse is a 120V DC pulse just as in the transceiver.  The reed relay I am looking at has a resistance of 1750Ω in each coil and intended for 12V DC.  There is no current rating on the coil from the relay's datasheet, but I know that it is intended for 12V.  12V into 1750Ω would be 7mA.  Is it safe to assume, then, that the coil's current rating is 7mA?  Using Ohm's law, I find that 120V (the pulse from the UM506) into 1750Ω would be 69mA or 8.2W across the coil, so I would need to drop the current by about 62mA to make it safe for the relay coil.  120V ÷ 62mA = 1935Ω and 120V x 62mA = 7.44W, so maybe two 2kΩ 8W resistors would do it (one for each coil in the relay).  But will the voltage 10x the coil's rating make a difference?

I may have just arrived at the answer while asking the question, so if anyone would like to check my work and confirm and/or correct me, please do! 

[dave w]

I can confirm this now.  The only MOV is across the SCR that pulses the relay coil.  There is no MOV across the AC input pins. 

At one time there was a MOV across the input, but now there isn't so there is no surge/spike protection. This continuous "cost reduction" practice is one of the reasons the old X10 WTI company is no more.

[Brian H]

I just thought of a possible issue.
If you remove the AC Ratchet relay and the On Off sensing circuit does not see the AC output going on and off. When you send it an X10 On it will pulse a few time trying to turn it On. If you send an X10 Off. It will think it is Off and not pulse at all.

The Home Automation Knowledge Base from idobartana.com had a modification for a momentary pulsing of a 12 volt relay. In a X10 Appliance Module. Used a dropping resistor, 12 volt zener diode and a filter cap. Basic pulsing circuit should be the same as the TM751.
I did one and it works just fine.
Last time I looked the site was down but I will look and see if I can find the circuit.
I will look at your circuit and get back.

[toasterking]

The Home Automation Knowledge Base from idobartana.com had a modification for a momentary pulsing of a 12 volt relay. In a X10 Appliance Module. Used a dropping resistor, 12 volt zener diode and a filter cap. Basic pulsing circuit should be the same as the TM751.
I did one and it works just fine.
Last time I looked the site was down but I will look and see if I can find the circuit.

Brian, is this the circuit?
http://web.archive.org/web/20140531004739/http://www.idobartana.com/hakb/appliance_momn.htm
I was going to try that one at one point, but I could not find a source for a reed relay with a 4650Ω coil.

I just thought of a possible issue.
If you remove the AC Ratchet relay and the On Off sensing circuit does not see the AC output going on and off. When you send it an X10 On it will pulse a few time trying to turn it On. If you send an X10 Off. It will think it is Off and not pulse at all.

This should not be an issue in the UM506 because it does not have an on/off sensing circuit.  It has two coils in the latching relay:  one coil to close the contacts and a second coil to open them.  I was wanting to replace it with a latching reed relay with two coils because it will need to close and open frequently and quietly and was hoping I could do this by just adding resistors of the right value.

[toasterking]

At one time there was a MOV across the input, but now there isn't so there is no surge/spike protection. This continuous "cost reduction" practice is one of the reasons the old X10 WTI company is no more.

I agree with the latter statement in general and typically the chronology is representative of that. However, in this case, Brian's module has an MOV across the input and mine, which was made almost 11 years earlier, does not!

[toasterking]

Last time I looked the site was down but I will look and see if I can find the circuit.

I did not realize prior to your post that this site has disappeared. When I started this topic about 5 weeks ago and linked to it, it was still working!

[Brian H]

It may have moved to a different server.
Now I get a warning of an Untrusted Connection. Do you want to take a chance or get out of here now.