Every remote visual inspection starts from the same commercial question: is it cheaper to look inside, or to open it up? Opening a vessel means isolation, permits, scaffold, a confined space entry, and a shutdown that costs more than the inspection ever will. Looking inside means getting an optical instrument through an access port the size of a fist.
Here is what RVI can actually resolve, where it stops, and how to specify it so the result stands up.
The equipment, and why it matters which
"Borescope" gets used for the whole family, but the differences decide what you can reach.
Rigid borescopes are a straight optical tube. Best image quality of the lot, because the light path is glass rather than fibre or a sensor at the end of a cable. Useless the moment the path bends. Ideal for cylinder bores, straight tube runs and castings with a clear line of sight.
Flexible fibrescopes carry the image along a coherent fibre bundle. They go round corners. The trade is resolution — you are seeing the image sampled by however many fibres are in the bundle, and broken fibres appear as permanent black dots.
Videoscopes put a CMOS sensor at the tip and send an electrical signal back. Better resolution than fibre, articulating tips that can be steered to look sideways and backwards, and the image is digital from the start — so it can be measured, recorded and shared. This is what most industrial RVI now means.
Pan-tilt-zoom cameras and crawlers for larger volumes: tanks, vessels, ducts and sewers, where the instrument is placed inside rather than threaded in.
Drones for structures where the problem is height rather than access — flare tips, chimney interiors, tank roofs, jacket members above the splash zone.
The selection is driven by the access geometry, not by preference. A 6 mm port and a 90° bend eliminates most of that list before anyone discusses image quality.
What you can actually see
RVI is a surface method. It sees what is on the inside surface, and nothing beneath it. Within that, it is very good at:
- Corrosion — general wastage, pitting, and the distinctive under-deposit attack that only shows once you can see the surface
- Erosion — particularly at bends, tees and downstream of orifices where flow does the damage
- Weld condition from the inside: root profile, excess penetration, burn-through, misalignment at the root
- Cracking, where it is open enough to see and the surface is clean enough to show it
- Mechanical damage — impact marks, gouges, distortion, missing or displaced internals
- Blockage and fouling — scale, debris, product build-up, collapsed internals
Measurement, and the honest limits of it
This is where RVI is most often over-sold.
Modern videoscopes offer 3D phase measurement or stereo measurement, projecting a pattern onto the surface or using a split optic to build depth from parallax. Used properly, they measure pit depth, remaining wall at a gouge, and the length and width of an indication to a genuinely useful accuracy.
Used carelessly, they produce a confident number that is wrong. Accuracy depends on the tip being close enough, the angle to the surface being reasonable, the surface being matte rather than reflective, and the operator picking the right cursor points. A measurement taken at a shallow angle down a long dark bore is an estimate with a decimal place on it.
Two practical rules. First, treat RVI measurements as screening: they tell you what to worry about and roughly how much, and anything close to a rejection limit gets confirmed by ultrasonic thickness measurement from the outside. Second, insist the report states the measurement mode and the standoff distance, because without those the number is not reproducible.
Where RVI is not the answer
Being straight about this is what makes the method worth specifying.
It cannot see into the wall. Sub-surface defects, lack of fusion, inclusions and internal cracking are invisible. If you need volumetric information, you need ultrasonics or radiography.
It cannot see through deposit. A vessel with 3 mm of scale on the wall gives you an excellent inspection of the scale. Cleaning is part of the scope, or the scope is not achievable — and this is the single most common cause of an RVI job producing nothing useful.
It is only as good as the lighting. Large volumes swallow light. A tip light adequate for a 50 mm bore reveals very little of a 3 m tank, and the resulting footage looks like a dark tunnel with a bright circle in it.
Tight cracks may be invisible even on a clean surface. Where surface-breaking cracking is the concern and the surface can be reached, magnetic particle or penetrant inspection is far more sensitive.
Specifying a job that produces something usable
The difference between an RVI report you can act on and an hour of unusable video is decided before anyone mobilises.
Say what you are looking for. "Inspect the vessel" produces a tour. "Confirm the condition of the shell-to-nozzle welds and quantify pitting in the bottom third" produces a result. The technique, the equipment and the time all follow from the question.
Confirm the access. Port diameter, distance to the area of interest, and every bend in between. This determines the instrument, and finding out on site that the only port is 6 mm wastes a mobilisation.
Agree the acceptance criteria first. RVI generates a lot of indications. Without a threshold agreed in advance, every one becomes a discussion, usually with the crew standing by.
Deal with cleanliness. Decide who is cleaning, to what standard, and whether it happens before the crew arrives.
Specify the record. Still images at defined locations, continuous video, or both — and where it will be stored. Which brings us to the part that outlives the inspection.
What it costs against opening up
The comparison people rarely do properly.
Opening a vessel is not one cost. It is isolation and de-isolation, permits, blinding, cleaning and gas-freeing, scaffold in and out, a confined space entry with attendants and rescue provision, the inspection itself, then reinstatement and re-test. On anything of size, the inspection is the smallest line on that list.
RVI removes almost all of it. What it adds is the access requirement — a port, a nozzle, a manway that can be cracked rather than fully opened — and the cleaning needed for the instrument to see anything.
Where the numbers change is when RVI finds something. If the result is "significant wall loss at the bottom third", you may end up opening the vessel anyway, and having paid for both. That is not an argument against RVI; it is an argument for being clear about what decision the inspection is feeding. If the realistic outcomes are "fine, carry on" or "open it", then RVI is buying you the chance of avoiding the opening entirely, and that is usually worth having.
If the decision is already made and the opening is happening regardless, RVI is a survey that tells you what to prepare for — different value, still real, but worth being honest that it is not saving the shutdown.
The recording is the deliverable
A visual inspection carried out by someone standing next to the component produces an opinion. A remote visual inspection produces a file.
That difference is worth more than it first appears. The footage can be reviewed by a specialist who was never on site. It can be shown to a client, a regulator or an insurer. It can be compared directly against the same location at the next shutdown, which turns a snapshot into a trend — and a pit that has grown 0.4 mm in two years is a completely different conversation from a pit of unknown history.
For that to work the record has to be locatable. Footage with no positional reference is a video of some metal. Every image needs to say which component, which weld, which orientation, at what distance along the run — recorded at the time, not reconstructed afterwards from memory.
Our RVI results are recorded into the same reporting system as every other method, so images are bound to the specific examination, with the technician, the equipment, its calibration and the acceptance criteria attached. When someone asks in four years what the inside of that vessel looked like, the answer is a lookup.
The short version
RVI answers "what does it look like in there" without the cost of opening it. It is excellent on corrosion, erosion, internal weld condition and mechanical damage; blind to anything beneath the surface; and entirely dependent on access and cleanliness. Measurements are screening-grade unless taken carefully and confirmed. And the recording, properly indexed, is usually worth more over the asset's life than the inspection that produced it.
If you have a vessel or a system you would rather not open, tell us the access and what you need to know, and we will tell you honestly whether RVI can answer it — including when it cannot.
TECHNICAL REVIEW — DELETE FROM THE LINE ABOVE, DOWN
Scheduled for 2026-08-25. Not to go live until a Level 3 or the RPA has read it.
Check: 1,905 impressions and 1 click across the cluster. THERE IS NO RVI SERVICE PAGE — this article is a stopgap; the real fix is a /remote-visual-inspection/ service page. Confirm which equipment we actually hold before the equipment section implies capability we do not have.
Image to shoot: Borescope in an access port with the live feed visible alongside, both in one frame.
Internal links already in the text: /non-destructive-testing-ndt/, /digital-tools-inspection-technology/, /ultrasonic-testing/ — confirm they read naturally, do not add more.
