Pinned post
Druk
Verified account This account is verified. Learn more@Druk_RN
Pro Halo Player | Trust the Process 🇬🇾 | metafy.gg/@druk84/sessio…
821 Following 6.4K Followers 33K Posts
Latest posts 5
$eose
If they are already at 24/7 operations this early in the quarter compared to the “end” , what will revs be?
To me it seems the picture is becoming quite clear, they need to produce to get the field data to get more orders to run at higher utilizations and produce more volume for better margins, at the same time without burning too much cash. But TH/L2 is efficient and maybe higher utilizations on that alone without TC gets them to plug the hole.
Benefit of the L1 move early is having all those already trained employees ready to swap over to Line 2 and ramp shifts fast without (much) training
If they are already at 24/7 operations this early in the quarter compared to the “end” , what will revs be?
To me it seems the picture is becoming quite clear, they need to produce to get the field data to get more orders to run at higher utilizations and produce more volume for better margins, at the same time without burning too much cash. But TH/L2 is efficient and maybe higher utilizations on that alone without TC gets them to plug the hole.
Benefit of the L1 move early is having all those already trained employees ready to swap over to Line 2 and ramp shifts fast without (much) training
$EOSE
If Eos closes the year at $350M revenue, the unit economics get very interesting.
Q4 would need ~$142M, at ~670 MWh.
Revenue per kWh: $212 (ASP, the lowest recent print) + $45 (45X tax credit) + $3 (ETC, electrolyte tax credit) = $260/kWh
COGS per kWh: ~$255 (my estimate: COGS needed to hit (15%) adj. GM at that volume)
$260 vs $255 → crossing breakeven on a per-unit basis, and getting very close to contribution-margin positive.
If not Q4, then Q1. This is done !!! , done !!!.
All Q4 figures are my own estimates, not company guidance. Company guides FY'26 to the lower half of $300–350M. ASP held flat at $212; tax credit treatment assumed.View quoted post on X
$EOSE
Q1/Q3
20-100-20 SOC RTE was avg 76.9% for DawnOS only on the Q1 release.
On the Q3 release, the RTE for 20-100-20 SOC is 79% including BOTH NON-DawnOS and DawnOS on avg 3.5hr duration (higher RTE at longer duration apparently). Don't know how many cubes still are operating without DawnOS but they said retrofitting will go on
Avg before DawnOS on SOC 20-100-20 is 41.9% (lol)
Retrofitting with DawnOS circuit board per site on average should more than double (79% avg even with the avg 41.9% non dawn OS performance included, so those with DawnOS have to be higher than 79%) field data produced per cycle (discharge?).
About half of the 3.5GwH of "late-stage opportunities" is awaiting financing, which is contingent on the field data being more consistent overall. > a 2.1% increase in efficiency in 2 quarters of field data, which feeds back into creating more field data and creates a positive feedback loop which makes it more bankable and should solve the financing issue.
Don't know why Joe didn't mention this at all, very significant progress in DawnOS which is key to orders because they are about to churning through that backlog when L1 is up at TH. If orders were there maybe Cerberus lets them use the revolver for L3 so they get more of the tax credits before their phased out.
Q1/Q3
20-100-20 SOC RTE was avg 76.9% for DawnOS only on the Q1 release.
On the Q3 release, the RTE for 20-100-20 SOC is 79% including BOTH NON-DawnOS and DawnOS on avg 3.5hr duration (higher RTE at longer duration apparently). Don't know how many cubes still are operating without DawnOS but they said retrofitting will go on
Avg before DawnOS on SOC 20-100-20 is 41.9% (lol)
Retrofitting with DawnOS circuit board per site on average should more than double (79% avg even with the avg 41.9% non dawn OS performance included, so those with DawnOS have to be higher than 79%) field data produced per cycle (discharge?).
About half of the 3.5GwH of "late-stage opportunities" is awaiting financing, which is contingent on the field data being more consistent overall. > a 2.1% increase in efficiency in 2 quarters of field data, which feeds back into creating more field data and creates a positive feedback loop which makes it more bankable and should solve the financing issue.
Don't know why Joe didn't mention this at all, very significant progress in DawnOS which is key to orders because they are about to churning through that backlog when L1 is up at TH. If orders were there maybe Cerberus lets them use the revolver for L3 so they get more of the tax credits before their phased out.
