Cavitation -- formation of an air or vapor pocket (or bubble) due to lowering of pressure in a liquid, often as a result of a solid body, such as a propeller or piston, moving through the liquid; also, the pitting or wearing away of a solid surface as a result of the collapse of a vapor bubble. Cavitation can occur in a hydraulic system as a result of low fluid levels that draw air into the system, producing tiny bubbles that expand explosively at the pump outlet, causing metal erosion and eventual pump destruction.

Cavitation erosion -- a material-damaging process which occurs as a result of vaporous cavitation. "Cavitation" refers to the occurrence or formation of gas- or vapor- filled pockets in flowing liquids due to the hydrodynamic generation of low pressure (below atmospheric pressure). This damage results from the hammering action when cavitation bubbles implode in the flow stream. Ultra-high pressures caused by the collapse of the vapor bubbles produce deformation, material failure and, finally, erosion of the surfaces.


Cavitation Damage

Initial Bubble


Initiation of Bubble Collapse



Forming of Liquid Jet


Impact and Metal Extrusion

Cavitation occurs when the pressure acting on a fluid is below the saturation pressure of the dissolved gas in the fluid. At this point, if the bubbles which are flowing with the fluid as it passes through the system encounter a region of higher pressure, they will collapse as illustrated in Figure 1. The process may be violent, depending upon the load pressure on the hydraulic pump. This can cause broad, high-frequency vibrations, noise, material damage and thermal degradation of the oil.

This process requires only nanoseconds and the localized temperature may be 2012ºF (1100ºC) or higher.
source


An asymmetrically collapsing bubble next to a wall. Due to the bubble's asymmetry, jets can develop that are directed toward the wall. A bubble's collapse can be extremely violent, as revealed in light emission, called sonoluminescence.

Excessive amounts of water in the fuel will affect engine performance, dilute oil, and can lead
to the formation of acids and varnish in the engine, as well as reduce the life of diesel fuel
pumps and injectors. An authority on the rebuilding of diesel fuel pumps stated that since fuel
supplies have become more critical, service life of pumps in many fleets has declined as much
as 50%.
Source

Water damage is a leading cause
of premature failure of fuel injection systems. Diesel fuel containing excessive water that
enters the injection system can cause irreversible damage in a very short time. Many
diesel engines are equipped with water separators that cause small water droplets to
coalesce until they are large enough to drop out of the fuel flow where they can be
removed. There are some reports that these water separators are not effective when used
with biodiesel.

Source


source Bosch Service Telegram “Damage to Distributor Injection Pumps

The pics of water damage show a perfect example of cavitation damage. Cavitation erosion results from rapid fluctuations in oil film pressure. When the pressure in an isolated area of the oil film drops below the bulk vapor pressure of the oil, a small, vapor-filled cavity is formed. This cavity then travels to an area of higher pressure where it collapses with the surrounding oil impinging on the adjacent bearing material. This action can eventually erode the bearing surface, as shown.

Cavitation is damage to metal and plastic components of liquid handling systems caused by the collapse of gas bubbles (steam, vapor, air, or other gases) on a metal surface. The repeated collapse of a large number of bubbles damages the protective metal oxide and removes the metal by the mechanical action of the minute shock waves generated by the collapse. The result is thinning of the attacked components which, in the case of piping, can lead to leaks and even a catastrophic failure.

Cavitation is often misdiagnosed as erosion-corrosion (flow-accelerated corrosion). It is most likely to occur when the liquid is near its saturation point, when there are dissolved gases in the liquid, or when the local flow velocity and turbulence are high. The flow velocity reduces the static pressure, bringing the liquid closer to saturation where gases come out of the solution and vapor bubbles form. Pumps, valves, orifices, and piping components which produce high local flow velocities are susceptible.

source

There are two types of cavitation..gas cavitation and steam cavitation.
since the probability of gas cavitation is extremely low due to the small amount of soluble gasses normally resent in VO we mainly need to be concerned only with the second...steam cavitation.

Steam cavitation results when water (micro) droplets turn into steam due to heat and/or pressure reduction inside our Injection Pumps. These bubbles of steam can then collaps violently..or implode...causeing damage to watever surface they are adjacent to when they do. This damage is microscopic but is cumulative in nature. Any IP subjected to cavition damage on a constant basis WILL experience a dramtically shortend useful life. Since IP replcement can cost in excess of $3000 it is prudent to limit cavitiosn as much as possible for economic reasons.

With VO conversions we have three strikes aginst us as far as the likelyhood of steam cavition damage in our IPs.

The first is that cavition is made more likely the higher the viscosity of of the pumped fluid is. VO is significantly more viscous than diesel fuel.

The second is that heat can also contribute to the likelyhood that "steam" cavitions will take place" if water is present in any quantity. Typically we tend to thin VO by heating it prior to the IP.

And the third strike is that WVO nearly always contains SOME suspended microdroplets of suspended water unless it has been treated to remove it. Sometimes the droplets are SO numerous theyare actually visible as a haze in the wvo even at temperature warm enough to completely liquify all of the oils and fats it is composed of.

We cannot do much about viscosity with out raising the teperature of our VO fuel and trading one one contributing factor for the other. So the only real option we have to lower the likelyhood of the formation of bubbles of steam in our VO fuel as it passes through our IPs is to make certain that as little water is present in our as is possible.

Damnnnn... thats a powerfull little jet that bubble has. Hmmmmm... nowwww can that power be harnessed? If so we may have something better than WVO for fuel? We can have Cavitation powered Jet cars

You should see some of the research papers on "supercavitation" and cavitational nano pumps. Far out stuff. And yes the energy of steam cavitation bubbles is being researched as a way to initiate low energy input hydrogen generation using catalyst surfaces...and theres even some stuff on using cavitaion to weld/cut nano parts for nano machines.

The stuff I had to wade though to find what we are interested in with VO conversions was an enormous pile. And I knew what I was looking for since I have been researching how water microdroplets can damage IPs for a LONG time.
I pity anyone who might try that search from scratch.

Dana: do you have any references specifically siting cavitation damage to IPs? The Bosch report doesn't mention cavitation, and only shows conventional damage, nothing specifically due to cavitation. (UPDATE: See later posting where a report was actually found, from Bosch)
The steam engineering cavitation report was good, but also wasn't specifically IP related.
I haven't heard from the Bosch folks yet, but will report when I do, and will continue looking. I'm anxious to either support or disprove the IP cavitation theory. I'd be happy to help your search in any other ways, and am open to suggestions.

Johno,

Actually what is shown in the Bosch report IS cavitaion damage. It shows the "bearing" surface of the part of the IP which actuates the high pressure pumping of fuel in a rotary (Bosch)pump. The areas which show pitting are the spots which undergo them most extreme pressure variation during operation..AND use the fuel as the lubricating fluid.

AS the "wheels" roll over the wavy "bearing surface" they create a temendous amount of pressure between themselves and the wavy
track they roll on. This is due to the "backpressure" transmitted to them via the fuel being compressed to injector actuating pressures. If you go to the report and copy the image you can blow it up a bit more and this will be a bit more obvious.

Let’s imagine the process in slow motion.
As the roller bearings proceed from one of the “peaks” of the “wavy” track or “bearing surface” the oil between them is at its lowest pressure. This creates the optimum conditions for steam bubbles of water to form from microscopic droplets of water if they are present. But as the rollers begin to proceed “up” the next peak the pressure in the oil is dramatically increased causing these micro bubbles of steam to collapse.

You will notice if you look closely at the original images that the “bearing wheels” have an even “pitting” on the surfaces which roll over/contact the wavy “bearing surface”. In contrast the “bearing surface” itself only shows similar “corrosion” on the points where the “lubricating layer” of fuel undergoes rapid pressurization.

This is classic “steam cavitation” ,
Which causes “material breakout” to occur. when water contaminated fuel is passed through an IP.

Are you saying that simply because the word “cavitation” is not used to directly describe this damage it is NOT the cause?

I don't know what you mean by "conventional damage". Cavitation is the most common damage to IPS. What do YOU think caused the IP "water contamination" damage shown in that report?

I worked for Cummins Engine Company many many years ago. The cylinder liners on the in-line six (the NT engines found in most Semi-trucks) are a wet liner. That is they are exposed directly to the coolant. There are rubber O rings around the bottom of the cylinder and a brass compression ring at the top. If the trucker was not diligent enough in his upkeep of the cooling system, the liners would pit from cavitation and soon would leak coolant into the crankcase. DCA was the chemical additive that was put into the coolant to stop this. It stood for dry chemical additive, don't know why, it was a purplish liquid. It was also alkaline IIRC. Electric discharge machining (EDM) also works on the principle of cavitation. Ever hear of Magna-Porting? A gentleman here in Michigan developed this to reduce muzzle jump in high powered hand guns. EDM was used to machine small ports in the grooves of the barrel at the end near the sights. Just before the bullet left the barrel, gas would shoot out these ports and reduce muzzle rise. These two forms of cavitation do not happen under extreme pressure. Just FYI. Sorry to ramble off-topic.

quote:

I don't know what you mean by "conventional damage". Cavitation is the most common damage to IPS. What do YOU think caused the IP "water contamination" damage shown in that report?

By "Conventional damage" I meant fretting, abrasion, galling and corrosion, etc, when talking about damage that happens to lubricated parts. The use of the term cavitation damage in that context is new to me, and is not mentioned in my technical books, most of which are several years old now. Cavitation seems to be a more recent concern. The pictures of damage are similar to the pictures in my books, but my books don't attribute the damage to cavitation, but rather to fretting, galling, etc. Perhaps my books are out of date, but I've been unable to find a report that specifically identifies cavitation as the cause of IP damage, or any kind of rolling element damage. I am finding plenty of reports that attribute the damage to the "conventional" reasons I mentioned, but none of them mention cavitation as a cause of IP damage, including the Bosch report. (UPDATE: I heard back from Bosch, confirming Cavitation damage. See my later posting, below)
Since I'm driving with both BD and SVO, and I'm already convinced that water in either fuel is very, very bad for the my unobtainable IP, I'm willing to go to great pains to dewater them thoroughly, regardless of the detailed cause of damage. Water DOES damage IPs - I don't mean to question that, only how it happens.
I'm familiar with the Cummins cylinder liner cavitation problem - somewhere I've got a copy of the report, which included good micrographs of the damage. The coolant additive fixed the problem. The cavitation problem inside the IP is different.
We do a lot of EDM machining, mostly of very weird materials. Cavitation is part of the phenomenon that removes the material - it's a combined effect with the spark that electrically heats the material in a microscopic point, then blasts it away.

The search continues...

"Fretting" is litterally "eating away" of matieral ..an accurrate description of BOTH cavitation and oxidation corrosion.

"Galling" refers to "wearing away by friction" which would best fit the damage shown in the later images of that report. As would "Abrasion".

Neither abrasion nor galling would seem to be included in the report section which refers to the images showing water damage. And I see no way oxydation damage would possibly account for the localized pitting in the images of damage due to water contamination.

If you care to include the corrosion types covered under "etc" I will be happy to addres these as well.

I appreciate that you are not questionsing that water damages IPs. How do you theorize this damage occurrs if not for cavitation?

Joe,

I do not know how the muzzle porting relates to cavitaion.

But cylinder liner cavitation is most definitely due to the passing of a transient high/low pressure wave through the liner and into the coolant from the "explosions" taking place inside the cylinder.

The first I ever heard of cavitation was in relation to propellor damage. As water with entrained air passed from the suction side to the pressure side of a propellor the bubbles would collapse with violent intensity and drill into the metal of the prop.

I took me a minute to get my head around this in relation to diesel engine cylinder walls. The IH (Ford) 7.3 litre engines are infamous for this. Being a bored out 6.9 the comparisons are easily at hand, more 7.3's eventually give up due to cavitation than do 6.9's. The thinner cylinder wall is more able to bulge and spring back as the pressure inside crests and subsides. The coolant in the relatively low pressure environment is acted on in a microscopic fashion in close proximity to the cylinder wall, due to the very high speed with which the cylinder wall bulges and retracts the the pressure in this region drops very dramaticly allowing water/coolant to boil and as the event is so brief the bubbles collapse violently and drill into the (unprotected) cylinder wall.

My conclusion : Any environment which combines boilable fluids, unprotected metals, and violent pressure changes is a prime candidate for cavitation damage.

Glenn

CONFIRMATION SUCCESS! I just received an email from the Robert Bosch Training Center, confirming that Cavitation is a cause of Injector Pump damage, and that they teach it in their curriculum. I'm forwarding the details to Dana for posting, etc.
The specific cause of the cavitation (water in fuel, supply side filter restriction, failure to floss regularily, etc) wasn't cited, but I'll try to clarify that for my own satisfaction.

In the case of prop cavitation. In modern high speed props the decrease in pressure can be sufficent to create bubbles of O2 litterally pulled from soution. This is referred to as "gas cavitation" in the scientific literature. It occurs where no entrained air is possible.

But bubbles of entrained air WIll cause cavitation as well. This is often the case in high pressure hydraulic oil pumps. But "steam" cavitation is also common. This is not surprising considering that IPs are essentially small positive displacment high pressure hydraulic pumps.

In fact..any micro bubble formed and then caused to violently and asymmetricly collapse aganist a surface can cause cavitation "pitting", fretting, or corrosion of that surface.

In the case of water contaminated fuel causing cavitation damage in IPs the phenomenon involved "steam" cavitation. This particular form involves the formation of micro bubbles of steam when microdroplets of entrained water are "flashed" into steam due to extreme pressure drop..temperature rise..or both...and then subsequently collapse against a surface where high pressure exists.

Considering the similarity of the damage caused by severe oxydation and cavitation..to the naked eye...it is little wonder that cavitaion damage is often mistaken for oxydation damage. Under the microscope however the pits etched into a surface by severe oxidation is easily differentiated by the raised edges of the "pits" created by cavitation. Pitting caused by the mechanism of oxidation have no such raised circumferential edges.

In the case of cylinder liners the bubbles can be "catalysed" by the existance of an existing pit or other roughness and so cavitaion pits can tend to litterally drill a hole through a very thick surface since they will tend to fome in exactly place each time and so create pits inside of pits. Even though the individual pits may only be a nanometer in depth...they add up over time.

Since cavitation bubbles come from the lowest vapor pressure components in a liquid, should we also consider other parts of our fuel beyond water? Dino-diesel has higher vapor pressure fractions than our biodiesel, making it more prone to cavitate, in theory. Methanol could, too. So would Gasoline (RUG). These will be good topics to discuss with the Bosch guy, if I get to talk with him.

This is a very interesting and complex subject. As the percentage of a dissolved low boiling component declines the temperature required to make it boil increases. Does this mean that the tendency to cavitate declines as the concentration is reduced?

very curious about the difference between cavitational liquids in *suspension* (i.e. water microdroplets in VO) and cavitational liquids in *solution* (i.e. ethanol in solution with VO and biodiesel)

instictively, it seems to me an ethanol/VO/BD blend should be perfectly safe to run through an IP
am I mistaken?
Dana? Neutral?

is there a world of difference between solutions and suspensions as far as cavitation propensity goes?
I am assuming there is - I'm assuming a solution, or blend, will have properties, including vapor pressure/boiling point, markedly different from the individual components - but am confused by the above posts and would very much like to know if that is not the case

(talk slow for the dummies in the room)(me, et al)

thanks

"Does this mean that the tendency to cavitate declines as the concentration is reduced?"
Simple answer: yes.
Complex answer: it depends on the substances thermodynamic and physical properties. The abrupt change that causes bubbles to pop into existance, then implode, may generate the necessary pressure change, and the necessary temperature change, but might not have enough energy to power the substance's phase change. The energy needed to change water from liquid to gas is greater than for IsoPropyl Alcohol, for instance. That would imply that IPA would more easily cavitate, but the phase change energy requirement may also be pressure dependant, making prediction less intuitive. I don't know what IPA's phase change energy requirements are at different pressures (especially very high pressures), but water's requirements don't change much up to 50,000psi, only the temperature requirement. I can probably dig up some specific numbers if people want, but I'm afraid they can't be directly tied back to a cavitation propensity ranking due to further complexity. It's just complex enough that I would distrust intuitive ranking, such as: 1)diesel, 2)water, 3)biodiesel. It might be some other order, and the order might even change depending on operating pressure.
Older diesel's operate at lower pressures and velocities than newer ones, making them less prone to cavitation, in general. This will obviously be a topic of increasing interest as new engines gain numbers. My old Rover diesel operates at something like 6000psi, yet modern common rail TDi's are operating at 20,000psi. That's a huge change.

Johno

How would you intuitively rank 3% methanol in biodiesel, 20% RUG in veg oil, 30% kero in veg oil, 40% diesel in veg oil, for tendency to cavitate?

It sounds like a trick question, but I'll give my intuitive answer, with the understanding that I'm probably wrong:
worst sensitivity cavitation: 20% RUG in SVO
then 40% diesel in SVO,
then 3% methanol in BD,
Most resistant to cavitation: 30% kerosene in SVO

My rational is that gasoline (RUG) will have the highest vapor pressure components, and kerosene the lowest, with #2 diesel in between. Methanol probably has similar properties to the lighter RUG fractions, but I needs confirmation.

The other confounding worry is what the lubricating properties of these mixtures are, since we're really concerned with damage to our expensive moving parts (IPs & injectors). RUG has no lubricating properties to speak of. Methanol is only slightly "better". Kerosene has no lubricating requirements, unlike diesel fuel, so they're blended differently, but kerosene is certainly "better" than RUG as a lubricant. Is veg oil a better lubricant than biodiesel? I really don't know, and suspect that it may depend on the source oil - palm, soy, canola, and degree of hydrogenation and fats contamination. Too many variables that I don't have a "feel" for, but also not related to cavitationability.
I'd rather have 3% methanol than water in my fuel, but not necessarily because of cavitation effects. Small amounts of water, even dissolved water, have devastating reductions in lubricating properties. One potential problem having methanol present is the increased ability to dissolve water into the fuel/methanol mixture.
Um, did I get carried away? Sorry.
back to work...

Thanks Johno, that was more than I hoped for. I don't think lubricity will be a problem with any of these fuels. VO must have a lubricity of the same order as biodiesel and there is plenty of VO in all the mixes. Perhaps one day we will know whether or not there is an upper limit to the pressure which is safe against cavitation but it will probably turn out to be more complex than that due to differing mechanical designs of IPs.

can anyone clear up the suspension v. solution question?
are the components of a solution really still acting as pure substances even though they're in solution?

here's what I mean:
my understanding of a solution is that molcules are uniformly mixed
i.e. if you have a 50/50 solution of ethanol/BD, those molecules are completely dispersed among eachother
if you could "scoop" up a "handful" of molecules, you'd find them alternating, 50/50, BD, EtOH, BD, EtOH, etc
evenly and equally distributed
no discrete pockets of pure ethanol, for example, which might more easily boil and then condense
doesn't having BD interspersed w/ the ethanol make it much less likely to turn into vapor and then collapse?
aren't microdroplets our real concern here?

or are solutions more complicated/less homogeneous than that? are there pockets of pure EtOH in a nice dispersed solution w/ BD?
so even if you have two entirely 'miscible' substances, the pockets of pure components are large enough to be a cavitation danger?

someone pls clear this up, my brain is going to overheat