I want to run my water heater on solar power (Newbie)

Everyone is right and wrong. Go on youtube and pretty much every one heating water with PV is a moron.  I heat water with PV and it is quite effective.  The old saying is that once PV got below $1.50 a watt that PV was better than tubes. Techluck is basically a sound idea but a poor execution, the unit heats up because it is IGBT based and the capacitors will likely go bad in a few years. So here is what I do. There are two tanks a 10 and a 20 gallon in series. It helps that the tanks were given to me.  Each tank has a 120V 2,000W element.  These aren't a bad match for a 36V, 48V or a pair of grid tie panels.  A controller made up of a capacitor bank, a couple FET and an Arduino.  That monitors the panel voltage and sets the PWM of each heater to maintain power point.  The capacitor bank stores the PV solar energy in the milliseconds that the FET is off.  Example: 50% duty cycle with an array that  can supply 5A. 5A goes into the capacitors while heating is off.  When it turns on, 10A goes into the heating element, 5A from the panel and 5A from the capacitors. Voltage isn't dropped, current is increased for a short period of time.  Basic power supply stuff.

The controller knows the temperature of each tank and the primary is heated first. Any extra goes into the second make up water tank. At a certain temp tank #1 shuts off and all power is directed to #2.  Any little extra power goes to tank #1 again till a very upper temp limit is reached. The whole system costs about $30 to make.  Here is the real kicker, All my water needs are supplied from potential PV power that would be just wasted.  I have 1,100W in a 36V string powering my entire camp, refrigeration, lights and even a dishwasher. Since I only have a car battery, I am heavy on panels as everything operates in real time.  Typically I can divert 2.5KWH a day into water heating.  The system has paid for itself just from the elimination of monthly service charges.

At home I have just installed two grid tie panels, $340 for 560W. Panels are cheap now. Compare that to the cost of buying a HPWH that may only last a dozen years. Electric savings are about the same.  For someone who doesn't want to go grid tie this is the most effective use of solar panels out there.  100% of the panels energy goes into heating water.  Nothing is wasted unlike the systems that almost all of you have.  The real trick is to keep it supplemental.  Depending on solar for 100% of your hot water needs is far too costly. Just think if about every household had just two solar panels. It is a simple cheap system to implement.  Evacuated tubes are very hardware intensive and installation can be a nightmare.  A wire is easy to run and doesn't have heat loss.

It is my only source of hot water in the summer and it is quite reliable. Highly cost effective for a "hobby project" compared to powering a home office or garden shed. Can provide backup power in an emergency and keep your boat battery charged. In conclusion it is a great idea, the hardware is not available because the solar world is draconian, and it can't be implemented by muggles.
  

Just as a guess 1ms might be enough, but in doing so you lose 15% of your power.  That can be significant since a lot of heating elements are higher resistance than you would like.  It also takes less time if the polarity reverses.    This is the trick the capacitor and resistor use in a NC/NO switch.  The capacitor charges up to the DC voltage through a high ohm resistor and the switch reverses the polarity extinguishing the arc.  There are designs out there that use a H bridge just like a MSW inverter.  Any tungsten contact is good for "one time" overheating events.  The advantage of going less than 100% is a three wire no brainer hook up. On youtube someone opening up a box gets 17K views. Something with technical content might get 100 views over a couple years.  I watched a guy plug in three solar panels and then asked you to come back for the next episode.

If anyone wants to buy themselves a Christmas gift consider buying a USB scope.  The Hantek 6022BE is only $60 and attaches to the laptop.  It gives you time, voltage and frequency.  It also allows screen save so you can share with others.  It is great at my camp and if your computer is all plastic it is floating.  Not  the greatest scope in the world but more than sufficient. It comes with two 10X probes but I suggest you get a 100X.  The bad thing about it is that it is easy to overload.  If you get a perfectly straight line with no noise it is overloaded.  Wish they would turn the trace red when it is overloaded.

Temp sensor is the way to go as long as the code recognizes a bad sensor.  Maybe I should get POWERJACK interested in a heater version.  It is still a very small market.  Techluck hasn't sold that many. Modifications are OK for me but just as much work as building something right.  The beauty is a micro can add other systems to the mix.  At my camp, I don't run the panels in real MPPT.  I use MPPC. Power point voltage only changes with temperature and within a season the slight loss of not tracking isn't that great compared to the other things it allows. There are a few chips that do this and call themselves MPPT, they have a set voltage and can vary that with a temp sensor.

I use this set power point to set priorities of loads.  Just setting a devices power point 1/4 volt lower gives that device a priority.  This is how I share power with charging and water heating.  It works slick. It can then dump even just a few watts into the heater. This method also allows you to attach a heater control to an existing charge controller.

That speed control is a little pricey and it still need a capacitor bank to work.  Why don't you wait and get a beta version if the board.