OP asked for "Power Surge protector or a proper system to protect my rigs on 240v."  That digital logger PDU does not claim any such protection.  

If it did, a post would have defined that protection and why it works - with numbers.  That recommendation was made without any reasons to believe it.  And no numbers.  A perfect example of how to suspect it is a bogus or scam recommendation.

It does not even claim to  do what the OP requested?  So why was it recommended?  A question that asks for the missing and required whats and whys.

SPDs do not prevent fires.  SPDs can create fires.  Power strip protectors are especially dangerous.  If one is found in luggage on a cruise ship, it may be confiscated.  Ships do more to protect from fire - such as confiscate those near zero joule power strip protectors.

Article says, "install them per Art. 285."  That does not make a protector effective - do appliance protection.  That only addresses a human safety issues - do human protection.

An effective protector must make a low impedance (ie less than 10 foot) connection to single point earth ground.  Citation does state wires must not have sharp bends.  But does not say why.  It is written for electricians - who typically do not know how electricity works.  Who are mostly taught what must connect to what for human safety.

Article does not say why a protector must connect to earth ground (not safety ground).  Plug-in protectors are ineffective (for so many reasons) because safety ground has numerous sharp bends - has excessive impedance.  'Whole house' protector is effective when it makes a low impedance (ie no sharp wire bends) connection to single point earth ground.  All four words have electrical significance - not defined by the article.

Article does not discuss what or why is critical.  It mentions no sharp wire bends and wires as short as possible (ie less than 10 feet to earth).   Article notes things an electrician must know.  For example, it accurately says 'Type' does not define an effective protector.  Type (1, 2, 3) addresses human safety issues.  Article does not discuss other critical considerations (ie single point earth ground).  It is written for installers - ie electricians.

A protector is only as effective as its earth ground.  Article does not mention that.  Single point earth ground is critical for effective protection. That earth ground (not a protector) should have most of your attention.  Only better electricians would know that.

If anything needs that protection, then everything needs that protection.  Just another reason why a 'whole house' solution is the only proven solution.  Electricians (and every homeowner) should know that.  Many do not.

Many electricians do not know how or why a 'whole house' protector is so effective.  Electricians are taught what must connect to what - as defined by a code that only addresses human safety.  Fundamental concepts that apply to surge protection (impedance, equipotential, single point earth ground, etc) are not taught to electricians.  Since those concepts are not relevant to protecting human life or meeting the National Electrical Code (NEC).

Concept is simple.  No protector (not even a 'whole house' protector) does protection. Protectors are only connecting devices to what does protection - just like a lightning rod.

Lightning rod connects to protection to protect a structure.  Protector connects to protection to protect appliances inside that structure.  Both must be installed to not threaten human life.  Code (and Type 1, 2, 3) defines that aspect. But code (and Type 1, 2, 3) says nothing about why 'protection' makes an effective solution.

For example, follow a bare copper wire that goes from breaker box to earth ground.  If that hardwire goes up over a foundation and down to an earthing electrode, then it satisfied code (for human protection). And may compromise surge protection.  Wire my be too long (ie more than 10 feet).  Multiple sharp bends over a foundation are code acceptable but compromise protection.  Many electricians would not know this. Training is only about codes for human safety.  Their training typically says nothing about appliance protection.

Another critical number is a protector's ampere number.  Lightning is typically 20,000 amps.  So a minimal 'whole house' protector is 50,000 amps.  Number says little about protection during each surge.  It defines protector life expectancy over many decades and many direct lightning strikes.  Only better informed electricians would know that.

An effective protector is only a connecting device to something completely different that actually absorbs hundreds of thousands of joules.  Best protection at each appliance is already inside each appliance.  Your concern is a rare transient (maybe once every seven years) that can overwhelm protection is anything (including robust protection already inside computers).

That Eaton 'whole house' protector is one well proven solution.  But again, it is only as effective as its eaarth ground.  Care is required to make sure that breaker box protector makes a low impedance (ie less than  10 foot) connection to single point earth ground.  Also all other incoming utilities (TV cable, telephone, satellite dish, invisible dog fence) must make a low impedance connection (ie hardwire ha no sharp bends) to that same ground.  Otherwise all protection is compromised.

That is your 'secondary' protection layer. Again, each protector is only as effective as its earth ground (that plug-in protectors do not have and will not discuss).  Inspect your 'primary' protection layer.  Relevant pictures (ignore text) about half way down and after the expression "more safety hazards" demonstrate what must be inspected to protect computers and everything else.

Ignore Type 1, 2, and 3 descriptions.  Those define human safety.  A type 3 protector in a type 1 location seriously increases fire risk.  Effective protectors are not defined by 'Type'.  They are defined by a connection to and quality of earth ground - for protection during each surge.  And are defined by a spec number (ie 50,000 amps) - for life expectancy of the protection.

Protection is always an answer to this question.  Where do hundreds of thousands of joules harmlessly dissipate.  A protector is only as effective as its earth ground.

A properly engineered system does numbers so that fire cannot happen.

If a power supply is oversized (ie runs at 50% capacity), then safety features to avert fire are set too high.  This (and for better life expectancy) is why a power supply in engineer designed system is so much smaller.

Connectors have current numbers.  For example, those Molex connectors (that typically connect a PSU to the motherboard) can only carry 6 amps maximum.  Nylon connectors got longer as motherboards started consuming above 200 watts.  Many more Molex pins carry same voltages so that a power supply outputting maximum current would not come anywhere near to fire temperatures.

Same with capacitors.  A failing capacitor must trigger current foldback limiting or trip a fuse long before creating a fire.  But again, when the PSU is grossly oversized (as is typical with computer assembler designed systems), then foldback current limiting cannot protect from a potential fire.

Locate a source of that fire to first identify the human mistake that made fire possible.  Then install necessary corrections (ie a smaller PSU, inline fuses, etc) to avert future fire.  Solutions always require both steps.

Even a keyboard typically gets power from a motherboard that often includes a fuse or something similar such as an automatically resetting fuse.  Because things that cause fires only exist when a human makes a mistake. Every part is designed also so that fire cannot happen.

In one case, the kid says, "Dad, I cut something.  And it make a noise!"  That cutting was the keyboard cable.  A  motherboard fuse (to avert burning) was replaced with an automatically resetting type fuse.  Most computer assemblers would not even know that fuse exists let alone know why.  Averting fire is always part of every design. Fire occurs because a human made a mistake.

Many posts ago in this thread, discussed were functions that help prevent such damage such as current foldback limiting.  But when a PSU is grossly oversized, that protection function is compromised.  The PSU will simply pump out excessive current without triggering protection.

Power supply designs are why, for example, some supplies are under greatest stress when at 50% capacity. Under least stress when at 0% or 100% capacity.  To say more requires numeric specification unique to that PSU.  Without numbers, then the 60% capacity to keep it cool is only wild speculation justified by feelings - not facts.

Same applies to fans.  Too many fans create reliability problems.  Fans must be selected based upon power consumption, acceptable hardware temperatures, and CFM.  Those numbers designed for what must be a good room temperature for any computer - ie 100 degrees F.

Since most who assemble computers do not know how electricity works, then computer assemblers are told to buy a PSU double what is required.  Then help lines are not clogged teaching how electricity works.  Only way to know a power consumption number is measuring with tools that any layman or pre teen can use.  On-line calculators are mostly useless.

Some components can maintain a flame.

For example, some resistors have ignited and flamed.  (I have made them flame like a candle.)  Those resistors are routinely used in location where no such current will create flame even with other failures.  Other resistors are flame retardant types - do not flame.  Often used both as resistors and as a safety fuse.

Most components are constructed of materials that might burn; but not flame.  But again, anyone with a 'dead body' and a propane torch can learn this by doing.

Another item with a history of creating fires are plug-in protectors.  APC recently admitted some 15 million protectors must be removed immediately due to fire. Flame is a problem when a protector is grossly undersized.

Electronic parts are designed to burn but not support flame.  Exceptions exist.  But a flame initially supported by arcing quickly extinguished once arcing did not support (maintain) that flame.

To better understand this, take a propane torch to the dead body to learn what will and will not support fire.  But I guess you also need a shovel to dig it up that dead body.

Dust is never conductive to those low DC voltages.  Dust would never create that spark.  However a stray metal fragment might.

Normal is for electronics to be heavily caked in dust without failure or arcing.

More likely something internally failed inside a semiconductor creating a physic defect.  Most all electronic failures have no indication.  But some rare ones (such as this one) do. So a short circuit caused the PSU to enter current foldback limiting - and shutdown.  Later, the same failure was now a tiny short circuit that burned open during a new power up only creating smoke.  Not flame as originally suggested.  Just arcing and smoke.  A major difference.

A properly designed PSU has foldback current limiting.  If properly sized, then a short circuit causes excessive current.  So the PSU cuts back power without flames.

However, many computer assemblers grossly oversize their PSU using a myth that more power makes a better PSU.  So that PSU turns a short circuit into flames; does not perform foldback current limiting.

Is your computer so hot as to also toast bread?  If not, then a 700 or 1000 watt PSU is excessive and unnecessary.  If a short circuit occurs, this grossly oversized PSU turns that short circuit hot as a bread toaster's toating wire.

ATX specs are quite clear.  All outputs from a PSU can be shorted together.  EVen that does not damage a PSU.  ATX specs even say how thick that shorting wire must be - so that every computer assembler will do it to verify a good PSU.

But if wires inside a computer are not big enough for an oversized PSU, then fire can result.

Molex connectors: each contact is speced to support up to 6 amps.  An assembler must add specification numbers to confirm that connector will not be overloaded.  A molex connector can be overloaded and still not cause melting.  Melting occurs when the connector is so grossly overloaded that the assembler should have know better just from basic knowledge. Never ignore numbers.

Nobody said the cable must be inside a wall or that a wall must be ripped open. Provided were some examples well defined by safety codes.  Wire must be installed to be protected - including from motion.  Cable can be routed on a wall's surface.  Hardware exists to do this - define by 100 years of science, experiment, and experience. How to install that wire is clearly defined by safety codes and standards.  Since your life is only secondary. Primary concern is for other's lives - including future guests, children, neighbors, and firemen.  That is why safety codes exist.

In a datacenter that must meet safety codes, a receptacle must be secured to the building.

You have described what is considered an extension cord.  Extension cords used in anything but temporary service are also considered a human safety risk.

Properly installed cables in a data center remain inside walls, underneath floors, or selected and secured in a manner that protects that cable from traffic or other movement.

For human safety, wires inside walls must not move.  Therefore wire must be secured by an appropriate device where it exits a wall.

They do not.  Anyone can read its spec numbers.  Those do no surge protection until voltage well exceeds 500 volts.  They do not claim to protect from surges that typically cause damage.  For example, a hundreds of joules surge is routinely converted by a computer's power supply into rock stable, low DC voltages to safely power its semiconductors. Where is a threat?

Surges that are a concern are hundreds of thousands of joules. That protection (rated at 1200 joules) means it can only absorb 400 joules and never more than 800.  Near zero joules.  Protection from surges (that can overwhelm protection inside any appliance) is by something completely different and called a surge protector.

Anyone can read spec numbers.  It claims to correct voltages that are not problematic.  Computers (like all electronics) are perfectly happy even when incandescent bulbs dim to 50% intensity.  How often is your voltage dropping that low?  Never?  If voltage is dropping that low, then a power conditioner must be installed on at risk appliances - furnace, refrigerator, dish washer, air conditioner, etc.

I never saw a specific request.  But then I also stop reading as soon as emotion is expressed.

Assuming a digital meter is available - since software that reads a meter inside a motherboard is insufficient.

Restore every connection as when the computer worked. AC power cord connected to a receptacle.  Computer not on.  Set a digital meter to 20 VDC. Attach its black probe to the chassis (bare metal; not paint).

Locate a purple wire (pin 9 ) from PSU to where it attaches to the motherboard. Use a red probe to touch that wire inside a nylon connector that attaches to motherboard. If necessary, make that connection using a needle or paper clip. It should read somewhere around 5 volts.  Record that number to three digits.

Next, do same with a green wire (pin 16 ). Then press computer's Power On button. Monitor how meter changes and what it eventually settles to. First number should be something well above 2.6. Second number should be something near to zero, Actual numbers and time to change (behavior) are relevant.

Repeat same to a gray wire (pin 8 ). Note a higher starting voltage, a lower final voltage, and its behavior. Report those three digit numbers and behavior.

Setup computer to execute as much software as possible. IOW it should be outputting sound loudly, while searching the disk, while playing complex graphics (ie a move or game), while powering a USB device, while accessing the internet, etc.  Having it access many peripherals simultaneously is important.  If it cannot power up, then monitor any one red (pin 4,21-23 ), orange (pin 1,2,12 or 13 ), and yellow (pin 10 or 11 ) wire for what each does as and after its power button is pressed.

Report all three digit numbers from those six wires. Next reply will identify or exonerate suspects.

BTW, if wires are not colored, then a PSU may not be ATX Standard. See  www.smpspowersupply.com/connectors-pinouts.html    for color and pinouts.

Snarky is the emotion that exists only in your mind.   "An answer means minutes of labor, a meter, and some requested instructions."  If you want help, obtain and request three things.  If you don't want help, then stop posting.  It could not be simpler.

Why are you being emotional to a post that is 100% technical.  Snarky is an emotion you have invented.  One sentence says exactly how to obtain assistance.  You must do three things.  "An answer means minutes of labor, a meter, and some requested instructions."  How much simpler can that be?

Record nothing.  If a defect exists, it exists constantly.  But the symptom of that defect occurs intermittently.

A meter is so simple (requires no training) that junior high science students use it.  An answer means minutes of labor, a meter, and some requested instructions.  Measuring 12 volts says little that is useful.  Instructions obtain in minutes numbers that results in an answer without any more speculation.

Symptoms no longer exist. But the defect does.  Normal is for a defective part to still boot and run a computer.

A meter is so ubiquitous as to be sold even in stores that also sell hammers.  Often for less than the cost of a good hammer.  Even available in Walmart for $15 or in Harbor Freight for $5.

The word 'surge' has many and completely different meanings.  For example, a surge on a USB port is due to insufficient current.  A surge on AC mains is a massive current spike.  A surge reported by a computer's motherboard is a low DC voltage problem.

A plug in protector is for an AC anomaly.  Your message is about a DC anomaly. To  say more about why requires number from a meter using some requested instructions.  That DC anomaly can result in intermittent computer instability.