MTH Z1000 Brick outputs 20V AC

No lionel only allows 19 volts on there engines.

What does your meter indicate for the 14V AC Accessory Output on the brick?

The Accessory terminals read 14.0v.   

I run 17 or 18 vac. I think I would use the controller, seems like the MTH transformers do put out a few volts more than Lionel.

Thanks for all the input, I think I will just stick with the control unit.   

If you want to lose the controller, find yourself a 10A bridge rectifier and use it to create a double diode pair voltage drop from the 20V output.  Here's one from a US seller for $6.50 shipped, eBay: 181265965528

Connect the + & - terminals together, connect the two AC terminals in series with one transformer lead.  This will drop the voltage about 1.4 volts and bring your 20 VAC down to safe levels for TMCC.

I'd bolt this piece to a chunk of aluminum for a heatsink, the 10A is with a proper heatsink.

Thanks @gunrunnerjohn I am going to stick with the controller!   

If the objective is to reduce the voltage applied to Lionel electronics, be advised that controllers like the MTH Z series do NOT reduce the peak or maximum voltage on the track.  That is, they "chop" off selected portions of the brick's sine wave so that a meter will indeed read a lower voltage.  Even at full Clockwise throttle on a Z controller (the ones I've observed on an oscilloscope), the sine wave is chopped a bit so the meter-measured AC voltage is indeed lower...but the peak voltage (top of the sinewave) is still there. 

Due to how engine electronics  converts AC to DC (using rectifiers and capacitors), it's the peak track voltage (vs. meter-measured AC voltage) that is the concern.  This has been discussed before but I don't remember the "final word" with respect to the Lionel limitation.  Anyone?

GRJ's idea of a bridge-rectifier drops both the peak and the meter-measured AC voltage.

An excellent point Stan, I didn't even think of that, but it's certainly true.

If it's considered rude to over-voltage Lionel engines it's just my opinion (I don't have Lionel engines) but I would spend a couple bucks and reduce the peak/max voltage using a bridge rectifier as GRJ suggests.  I poked around and see you can even get a 10 Amp bridge for $1 shipped from Asia or under $3 shipped from the US.  And with low-cost adapter jacks you don't have to splice/cut into the cable from the transformer brick.  If there is no other application for the Z controller, I'm pretty sure you can sell just the controller on eBay or the OGR buy/sell forum and more than make up the difference.

Again, this is based on the opinion that, it's not the meter-measured AC voltage, it's the peak AC voltage:

A couple things to add:

First is that while you are correct, Stan, about the controller still delivering the full peak to peak voltage even when it is turned down, I believe the controller still acts just like the method GRJ suggested where it drops the RMS of peak to peak to about 18.6 or so through it's electronics.  This is defiantly true for other controllers that use triacs, but with MTH's use of MOSFETs in the controller the drop may be a little lower than expected.  Still If I recall correctly the Z-controller will output about 18.6 with a 20v input.  (after a little research, the MOSFETs used in the Z-controller have 1.3 volt drop across them:  (FZ48NS)

Next, I'll trust the wisdom of folks that have much more experience with TMCC/Legacy electronics to know what the limitations of those boards might be.  For LionChief, however in the boards I have examined, I've been unable to find anything that would cause a failure with 24.7 VAC or less (35V peak to peak).  The caps in the power supply are rates for 35VDC and the regulator for 40v if I recall.  On LionChief Plus engines the voltage input for conventional control is limited with a zener diode to prevent excessive voltages from damaging the micro-controller.  The only possible point of failure I can think of below 24v RMS is perhaps the Gate to Source voltage on the motor mosfets (Vgs), with the 20v Ggs of the Fet in use, there must be some protection in use for this, however, or it would be damaged by AC input over 14 volts.  Next time I have an LC engine apart I'll have to investigate this further, but it's the only thing I can think of, as every other part is after the power supply, which is good up to 24.7VAC.  

As to the original question, I like the simple and sure-fire protection offered by the method GRJ suggested above.  

JGL

I don't know about 20 volts, but I saw an early Legacy locomotive cook on 22 volts, the victim was the DCDS board.  He was using one of the earlier Z-1000 bricks direct, some of them put out as much as 24 volts.

I was thinking that this morning, that the power dissipation inside products could be the limiting issue. But then I thought, how does that change anything? Best to use an 18 vrms brick, or the equivalent in all respects.

Add another pair of diodes if you're still above the limits.

Last transformer I know of in that range is the post War ZW at 21 volts max. Glad i run at half throttle most of the time, have yet to swap out the 12 volt bulbs on many things.