Although the DCS TIU and remote controller can be used to control a model locomotive of any size that has a DCS decoder and a maximum 10 amp current draw it might be considered as overkill for Ho models. The DCS Commander can be regarded as a cheaper single unit combination of TIU and handset which is perhaps better matched to Ho though it can readily be used to control the larger locomotives that MTH produces.
Like the TIU the Commander is a controller rather than a power supply and needs to be fed from something. Fortunately suitable power supplies are easy to obtain. If you happen to have a 110-120v step-down transformer (either a domestic version as can be obtained from many of the mail-order electronic goods suppliers or one of those large yellow units intended for builders powertools) then you can use the 110v transformer available direct from MTH. A fair bit of money can be saved by obtaining one of the larger multi-voltage power supplies that are sold as spares for laptop computers or, if you are prepared to wield a soldering iron, a simple domestic lead-acid battery charger. The recommended input voltage is 18 volts though 15v should be adequate, a/c or rectified dc or constant voltage dc, it doesn't matter. The Commander has a single output, rated at a maximum of 5 amps which should be adequate for three or four Ho locomotives and will just cover the gauge-1 Triplex.
The Commander has three "modes"
Pass-through simply cuts out the DCS controller, the purpose of this is to allow anyone with a DCC controller to have this connected permanently in line. In this mode the Commander has no effect and all locomotives are controlled by the DCC unit.
Conventional mode configures the Commander to act as a variable voltage (rectified dc) controller, the thumbwheel acting as the control knob with the direction button reversing polarity. In conventional mode the display indicates output voltage and current draw (please remember that your older locomotives may have a 12v rating). There is a momentum simulation function but it is not as easy to use as a dedicated throttle/brake + momentum unit.
DCS mode is naturally what the Commander is intended to be used for. When operating as a DCS controller a fixed dc voltage is fed to the rails with DCS commands being sent down one wire. If operating Protosound 2 locomotives on 2-rail it is necessary to have this signal aligned to the locomotive but the Ho Protosound 3 locomotives are able to pick up the control signal on either side. For 3 rail systems this was never a limitation but it is a nuisance for MTH's gauge 1 and finer scale O gauge locomotives.
Rather sensibly the Commander starts up in conventional mode, zero volts. MTH locomotives assume that they are operating under conventional command unless they recieve the DCS initiation command which means that if a DCS locomotive is placed on to a track section that has already initiated under DCS it will interpet the track voltage as a legacy command to get going at full power, careful how you go! As with the TIU the Commander does have a panic button (though the small Commander Remote does not).
My suggestion is that you consider the Commander if you run MTH gauge 1 or O gauge locomotives on an individual basis or just a few Ho simultaneously as it is considerably cheaper than the combined TIU with handset. If you want to run more than one large locomotive or more than a handful of Ho at a time under DCS then the TIU is capable of twice the power on each of its four outputs. If your intention is to obtain a DCS controller at the lowest possible cost then you should also consider the DCS Commander Remote, which is considerably cheaper, has fewer features but still has a 5 amp capability.
Neither the TIU/handset or Commander are perfect designs either in terms of ergonomics or total capabilities but it is fair to say that the Protosound 3 locomotive decoder is easily the match, and I would deem superior, to DCC.
At heart DCS is a form of cruise control, rather than setting a power level (as you might with a variable voltage controller) you set a speed which the locomotive should match whether up hill or down. This is very effective at slow speeds but for me it rather ruins the natural aspect of trains slowing down at slopes or high friction curves and the manual control aspect of applying greater voltage (with consequent increased output from any smoke units) for greater effort. The "cruise" aspect acts on all the functions which can have some tedious effects in any areas where the track does not give adequate control signal propagation. To give an example you might press the whistle button with the locomotive on a good section of track (which I suppose sends the start whistle command) and whistling starts but if, before you release the button (which I image sends a stop whistle command), the locomotive moves onto a worse section of track then releasing the button will have no effect. The Commander will send the stop whistling code but if the locomotive does not receive the code clearly it will continue whistling and will continue to do so until you press the whistle control whilst the engine is on a good section of track once more. If you issued a change of speed command on a dodgy track section this would also be ignored, the locomotive would continue with all the previously received commands still active. If the locomotive subsequently reaches a better piece of track you would have to issue all the commands once more to have them put into effect. I will point out that the Commander display does indicate when the locomotive is not responding, and it can be put into a mode which, with the locomotive set to a fixed speed, can indicate signal quality along the track. I think that matters would be much improved if the controller stored all the commands that were issued when the locomotive was not responding and retransmitted them as soon as contact is reestablished. If it were easier to implement then a repeated looped broadcast of each of the latest "on" "off" and "speed" commands would have much the same action. Really tight track joiners and clean track does help but strangely I have found that my oxidised brass gauge 1 track transmits a more reliable signal than my better looking nickel silver OO/Ho stuff. I also am not a fan of thumbwheels, much preferring a control lever or knob with fixed zero and maximum positions. Amongst other matters a thumbwheel really does not lend itself to precision shunting. I understand why a thumbwheel is the easy option for a handset (though other brands have provided the "fixed zero" control knob handset) but I feel it is a cop-out on a full size controller.
Criticisms aside the sound and smoke units are world class and the ability to select forward and rear couplers, whistle, horn and suchlike is vastly easier with dedicated pushbuttons compared to using function keys, numeric codes or manual pulsed commands, MTH are not far off the controller that I would like to have.
[ If you are intent on investigating a household car battery charger as possibly the cheapest method of powering a Commander (or TIU) from the mains then my suggestion is to get a basic model that can provide at least 5 amps , there is no need for auto-shut off or continuous float. If you have sufficient common sense to take appropriate care when dealing with mains voltages and you are able to put a voltmeter across the input to the rectifier (ie the transformer secondary output) you should register something around 16 volts a/c. Assuming that is providing a suitable voltage you can disconnect from the mains and then make appropriate connections for a cable to power the Commander (please remember to reassemble the casing etc.etc., oh lord help me, why is it neccessary to put all this rubbish in? Look you are considering adapting something that is intended to be connected to the mains so if you get it wrong and touch a live wire or the whole thing bursts into flames and burns your house down because you created a short circuit it is your own responsibility and not mine) ]