What does each exit actually realise?
Whole-unit sale is the only exit priced against what a buyer would otherwise pay for new equipment. The other three are priced against materials, against spares demand, or against nothing at all. On a heavy frame the spread between the best exit and the worst runs to two orders of magnitude, which is why the decision deserves more than a phone call.
| Exit | What it realises | What it costs to pursue | How long |
|---|---|---|---|
| Sell whole | Priced per kW against new equipment. Used heavy frames with hours on them are advertised at roughly $59 to $216 per kW; low-hour and never-installed units at $425 to $712 per kW | Diligence access, records assembly, and usually the removal window as a negotiated term | 60 to 120 days on a prepared asset |
| Part it out | Hot-section alloys carry real value: Inconel 625 around $8.50 per pound, Waspaloy around $9.00. Unused spares sold with a complete plant are valued near 35 per cent of original procurement, and materially less once the plant is broken up | Teardown labour, inventory, storage, and a long tail of small transactions | Open-ended |
| Mothball | Nothing now. It preserves the option to take a different exit later | Under 0.5 per cent of replacement value to set up a one-year programme, then ongoing monitoring, plus storage and insurance | Indefinite, if the preservation is real |
| Scrap | Ferrous at roughly $0.09 to $0.11 per pound at the yard; copper windings at $3.00 to $3.50 per pound once stripped | Dismantling and haulage, which for gas plant decommissioning averages around $15,000 per MW | Weeks |
The advertised prices are asking prices, not clearing prices. No public database records what used heavy frames actually transact at, so treat the band as the shape of the market rather than a valuation.
What is my unit worth as scrap?
Less than almost every owner guesses. A GE 7EA gas turbine weighs about 293,000 pounds and its generator about 333,000 pounds. At national yard prices for prepared steel, that 626,000 pounds is roughly $56,000 to $69,000 of ferrous, before anyone is paid to take it apart or haul it away.
Set that against the same machine sold whole. An 85 MW 7EA at the bottom of the advertised band is a seven-figure asset. The scrap number is not a floor under the negotiation; it is a rounding error against it. The only situation where scrap is the right answer is one where the machine genuinely cannot be redeployed and the site needs clearing.
Two things do carry real metal value and are worth separating rather than throwing in the skip. The hot section is nickel superalloy, not steel, and is priced per pound rather than per ton. And the generator and any step-up transformer hold copper: stripped transformer windings fetch $3.00 to $3.50 per pound against $0.08 to $0.15 for the whole unit as delivered. One surplus transformer buyer states publicly that it pays two to five times scrap value for units it can redeploy, which tells you what the recycling route leaves on the table.
How much does mothballing cost, and what does it buy?
A properly specified one-year mothball programme costs under 0.5 per cent of the unit's replacement value to set up. Reactivating a well-preserved unit costs 5 per cent or less. Restarting one that was simply switched off and abandoned costs 20 per cent or more. Mothballing is not storage. It is a maintained engineering condition.
What "properly specified" means is specific and measurable. Relative humidity held below 40 per cent, and below 30 per cent inside the generator, because corrosion products form at an exponential rate above about 60 per cent. Nitrogen or dry air blanketing on water circuits, fuel gas and feedwater piping at around 7 psig, with oxygen concentration checked quarterly and the blanket bled and refilled. Shaft rotation monthly, or at minimum quarterly on extended layup. Humidity controls read weekly.
Most idle plant does not get this. An industry survey found only 37 per cent of utilities routinely nitrogen-blanket the boiler and only 6 per cent protect the turbine. That is the gap between a unit that holds its value and one that quietly loses it.
Done properly, layup duration stops being the constraint. A Brazilian combined-cycle plant laid up for four years was recommissioned in seven weeks and subsequently ran above 98 per cent availability. Three never-installed 171 MW 7FA units built in 2002 and 2003 are still advertised today, kept under dehumidification with generator internal heaters for more than twenty years. Nobody publishes a point after which a laid-up heavy frame stops being restartable, and the evidence suggests preservation quality matters far more than elapsed time.
When does parting it out make sense?
Rarely, and almost never as a first choice. Parting out turns one transaction into dozens and destroys the thing a buyer is paying for: a machine that can be installed and run. It earns its place when the unit is already incomplete, when the rotor is at the end of its life, or when the rest of the machine has no home.
There is an asymmetry worth knowing. Unused spares sold alongside a complete plant are valued at roughly 35 per cent of their original procurement price. Sold separately after the plant is broken up, the same inventory holds much less value, and the usual recommendation is to move it to recyclers at a substantial discount. Spares are worth most when they are attached to the machine they belong to.
What does the decision really turn on?
Three things: whether the machine can be made to run, whether anyone needs that frame right now, and how long you can wait. The market answers the second question emphatically at the moment, and the first is usually a question of money rather than possibility.
New heavy-frame lead times run to six years and beyond. GE Vernova's gas equipment backlog stands at 116 GW, with customer conversations reaching into 2032. New equipment prices have risen 195 per cent since 2019 and are forecast near $600 per kW by the end of 2027. An installed unit that can be at a site in months is not competing with scrap. It is competing with a six-year wait, which is why the whole-unit exit has rarely been further ahead of the alternatives than it is now.
The third question is the one only you can answer. Selling whole takes 60 to 120 days once the asset is prepared. Scrap takes weeks. If the site has to be clear by a date, that constraint can quietly make the decision for you, and it is worth knowing that before it does rather than after.
What is the honest first step?
Establish whether the machine can run, and get the records together. Both exits that realise real money depend on evidence, and the two that do not need any are the two you should be trying to avoid. Then decide how much of the work and the risk you want to carry yourself.
Genloop is one of the counterparties on the whole-unit exit. The first step is an assessment on site by our own engineers, valuing the machine as it stands and the project it could power. From there an owner chooses between direct acquisition, where we take title and carry the risk from that day, and a recovery mandate, where we take the unit through restoration and redeployment and the owner captures the spread. For a fleet there is a divestment mandate, sequenced unit by unit; where the owner would rather keep title, the unit can be restored and put into contracted service instead of sold. Either way the asset never goes to the open market.