VFD and Titanium shaft in sea-water 2

[Skogsgurra]

Shaft corrodes in environment where it survived before use of VFD.

Does anyone have deep knowledge in this field? Suspect stray currents from frequency inverter (same thing as you have in ball bearings) that somehow affect the titanium.

Common wisdom says that AC should not cause electrolytic damage (no net charge transportation), but observations indicate that corrosion is a fact.

Theory: Streaming water spools ions away so they don't find their "way home" again when the current changes direction. Anyone done any observations?

 

[Skogsgurra]

Anyone? Please!

 

[Metalguy]

Dunno what VFD means (variable fan drive?), but half the time with AC is in the wrong direction corr.-wise, and you don't regain anything the other half.

 

[Skogsgurra]

Thanks Metalguy,

VFD = Variable Frequency Drive aka Frequency Inverter.

The frequencies involved in the corrosion are in the several hundred kilohertz region (HF part of the switching frequency) and my thinking was (is?) that the ions do not move much in a couple of microseconds and hence "fall back" when polarity reverses.

So I did not expect to lose tens of mm due to that kind of corrosion. Do you know about any publications that describe this?

A razor's edge is quite another story. Any progress?

 

[JimMetalsCeramics]

Its sounds lke you are enhancing the galvanic coupling between the titanium and the surrounding environment. You may also be ionizing species in the water.

 

[hacksaw]

you may have a grounding problem. VFD(variable freq drive) in and of itself does not cause corrosion.

what is you fluid, is the pump direct drive, have you grounded the motor, where is the corrosion occuring, and what sort of mechanical coupling are you using?

 

[Skogsgurra]

Sorry hacksaw, I am awfully late in answering your question:

The liquid is North Atlantic seawater. Current seems to flowing through shaft (steel coupling, not insulating) into pump and from internal shaft and impeller to pump housing and from there to ground. The affected parts are internal shaft and impeller. We obviously have to insulate the coupling. Or find some other solution.

Have you experienced similar problems?

Thanks for your interest!

 

[stevenal]

I've heard of the swept away ion hypothesis before. Also heard that some metallic joints will tend to rectify the induced current. I cannot confirm either one.

A few decades back I was involved with the salt water system of an aquarium. After changing from rubber coupled pumps to close coupled pumps, the corrosion of the cast iron impellers increased dramatically. I could measure both AC and DC potentials on the fan end of ther motor shaft to ground while it was running. 60 Hz, no VFD involved. Pump manufacturer tried various coatings with no luck. Finally replaced impellers with SS. I had thought of rigging a brush to ground the shaft, but never got the chance to try it.

 

[moltenmetal]

I presume that the speed after the installation of the VFD was LOWER rather than higher, but that may not be true. Why was the VFD installed? If the new operating speed is higher than before, have you considered that you may be experiencing velocity-induced damage to the shaft and impeller, due to removal of the passive layer on the titanium- I have heard of erosion-corrosion phenomena with titanium before. Just a thought...

 

[Skogsgurra]

Thanks stevenal and molten,

I should have mentioned that there are several pumps involved. The older ones are DOL and the newer ones are VFD. The old ones have normal lifetime but those with VFD have only months before they need repair/changing.

The speed of the new ones is lower than that of the DOL pumps (flow regulated by adjusting valves).

I had another case today. A small motor (4 kW with VFD) and I was finally able to measure about 20 mA AC in shaft between motor and pump. Sounds not so serious, but I guess that it will have some effect over time. N.B. that this is not the pumps in the original posting, but much smaller ones.

Have also started to think about shaft grounding with a carbon brush.

Have started a test with Titanium bars in salt water and feeding them with AC to find out what happens. Keeps the water circulating around one of the bars and still around the other. Central common bar is the counter-electrode. Any comments to that kind of setup?

 

[hamma1]

I have encountered something similar to this on boiler feedwater pump with variable speed gear drive - stray currents in shaft jumped across gaps of gear teeth and pitted teeth making excruciating noise. Solution was a simple insulated coupling

 

[Skogsgurra]

Thanks hamma1,

Yes. An isolated coupling would do the trick. But not possible in this case (motor shaft carries impeller and seals). No intermediate bearings and no room for it.

We have had several cases this year where PWM inverters have caused damage to seals, tubing, bearings and gears. We try to use inverters with sine-wave output whenever possible. They solve all these problems.

 

[EdStainless]

Since these are fairly small units I wonder how good the output filtering on the VFD is. There should be no stray harmonics.
I have had VFD/pump problems in the past. They turned out to not be galvanic, but corrosion fatigue. The VFDs were causing a small amount of high frequency pulsing that was wearing the heck out of couplings. Pitting, fretting, fractures, the whole works. It was solved with better output filtering.
I have trouble believing that this is galvanic. Almost everything else in you system is less noble than Ti. If you have some stray current or induced voltage issues I would expect the non-Ti parts to be going away very fast.


= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[YBG58]

Hamma1

Request to give more inf on insulated coupling ! esp. gear type coupling & other jaw type.It will help us esp in problems like above.

YBG 58
ybgandhi@gnvfc.com

 

[Skogsgurra]

Ed,

"I have trouble believing that this is galvanic. Almost everything else in you system is less noble than Ti. If you have some stray current or induced voltage issues I would expect the non-Ti parts to be going away very fast."

All components that are in touch with the liquid are Ti and diamond coated silicon carbide. And it is only where current can pass from component surface to liquid that corrosion has occured. Structural components like motor frame, exterior of pump etc are not affected. And we do not expect them to be affected either.

 

[btrueblood]

YBG:

Try:

http://www.tbwoods.com

and look for: Sure-flex and Dura-flex couplings. The HD coupling can provide isolation as well.

http://www.lovejoy-inc.com

has a coupler that can provide isolation as well.