Services
What it costs, and what you get.
Six services, each with a fixed scope and a fixed price. What is written here is what you would actually pay: the scope is agreed before the work starts, and nothing is billed by the hour.
| Service | Price | Duration | Buy it when |
|---|---|---|---|
| Data feasibility assessment | €2,400 | 5 working days | You do not yet know what your data can support. |
| Test-campaign design | €4,500–7,000 | 1–2 weeks | You have cells and cycler time, but no test matrix yet. |
| Battery passport SOH readiness | €12,000 | 3–4 weeks | Your passport SOH figure has to hold up by February 2027. |
| Silicon–graphite diagnostic | from €25,000 | After feasibility | You need the physics behind the drift, and a spec you can implement. |
| Training day | €5,500 / day | By arrangement | Your team needs the methods, not only the answers. |
| Advisory day | €1,600 / day | By arrangement | You want a design review or a second opinion, with no analysis. |
Most clients start with the feasibility assessment. It is deliberately small enough to sign off without procurement, and it means neither of us is guessing at scope.
Data feasibility assessment
€2,400Report within 5 working days
You send a representative sample of your cycling or field data. I tell you what can be extracted from it, at what uncertainty, and what is missing.
Deliverable: a written assessment covering data quality, achievable uncertainty on OCV reconstruction and SOH, identified gaps in the test matrix, and a scoped proposal for further work. If the data cannot support the analysis, the report says so plainly and sets out what would have to be measured to change that — which usually points to a test-campaign design rather than to a larger analysis.
If you commission further work within six months, the €2,400 comes off its price. So if we go ahead, the assessment has cost you nothing.
Test-campaign design
€4,500–7,0001–2 weeks
Buy this before the cycler starts, not after. A campaign that was never designed to support OCV reconstruction usually cannot be rescued afterwards at any price, and a mis-specified matrix wastes months of cell time.
Deliverable: a full test matrix — checkup intervals, rates, temperatures, depth-of-discharge windows, rest durations and sampling requirements — and, above all, the design of the checkup itself, which is where nearly all the diagnostic value sits: the quasi-OCV measurement and the rate it is run at, any dynamic segment, and pulse sequences along charge, discharge or both, chosen against what you actually need to estimate. Every choice comes with its reasoning, the degradation modes the design can and cannot separate, and the uncertainty you should expect on the resulting SOH and OCV estimates. Written so your test engineers can run it without further input.
Price depends on the number of cells and conditions, and on whether the campaign has to serve a passport filing as well as internal development.
Battery passport SOH readiness assessment
€12,0003–4 weeks
An independent re-estimation of state of health from your data, compared against what your battery management system reports at the same points, with the divergence quantified across cell life.
Deliverable: a report covering which Annex VII parameters apply to your battery type and which Annex XIII passport fields they feed; a data availability audit against each; the measured divergence between reported and independently estimated SOH; degradation-mode attribution for that divergence; a defensible accuracy envelope with the conditions under which it breaks; and a ranked remediation roadmap with effort estimates.
The relevant deadline is 18 February 2027 for electric-vehicle and industrial batteries above 2 kWh placed on the EU market. The Article 14 obligations on BMS data have applied since August 2024.
This is a technical assessment, not a conformity assessment. I am not a notified body. Legal determination of conformity rests with you and your notified body.
Silicon–graphite diagnostic engagement
from €25,000Scoped after the feasibility assessment
Full open-circuit voltage reconstruction and hysteresis characterization for your cell, degradation-mode decomposition across the aging set — separating loss of lithium inventory, loss of active material and resistance growth — and a specification for an on-board estimator that holds its accuracy as the cell ages.
Deliverable: parameterized OCV and hysteresis model for your cell, a degradation-mode breakdown over the aging campaign, quantified SOC error under your duty cycles, and an implementable algorithm specification with the compute and memory footprint stated.
Price depends on cell count, chemistry variants and the size of the test matrix. Quoted as a fixed sum after the feasibility assessment, not as an open-ended hourly engagement.
Training day
€5,500 / dayOne or two days, remote or on site
An in-house course for a cell-testing, BMS or validation team: the multi-species multi-reaction model, differential voltage and incremental capacity analysis, degradation-mode separation, and open-circuit voltage reconstruction on aged cells — worked through on real data rather than presented as slides.
Up to twelve participants. The two-day version adds a hands-on session on your own dataset. On-site delivery is quoted with travel; remote is the price above.
Advisory day
€1,600 / dayBy arrangement, remote — half day €950
Design review of an estimator architecture, a second opinion on someone else's analysis, or a working session with your team. No data analysis, no written deliverable beyond notes.
What I need from you
Published here so you can check the fit before writing. If your dataset falls short on several of these, the feasibility assessment will say so quickly and cheaply rather than after a large engagement has started.
Minimum useful dataset
- Time series of current, voltage and temperature, ideally at 1 Hz or faster for characterization tests.
- Cell format and nominal capacity, cell chemistry of each electrode if known.
- Periodic capacity checks distributed across the aging set, not only at end of test.
Helpful but not essential
- Periodic, low-rate quasi-OCV charge and discharge cycles.
- Half-cell or three-electrode data, if any exists.
- Aging at more than one temperature, C-rate or SOC window.
- Field data from deployed units alongside the laboratory set.
Formats read directly
Arbin, Maccor, Basytec, BioLogic, Neware and Digatron exports, plus CSV and ASAM MDF (.mdf and .mf4). Other formats are usually a day of work, quoted separately.
How engagements run
- If you need an NDA, send yours or ask for mine and it will be signed before any data moves.
- You own the deliverable and any analysis specific to your cells. I retain my pre-existing methods, tooling and published work.
- Nothing from your data appears in any publication without your written agreement.
- Fixed price, invoiced 50% at start and 50% on delivery. Liability is capped at the fees paid for that engagement.
- Prices exclude VAT. Invoiced from Armenia; EU business clients account for VAT under the reverse charge, so nothing is added to the invoice.
Start with a conversation, not an invoice
Thirty minutes, free, no obligation. Bring a description of your cells and your data, and you will leave the call knowing whether any of this is worth buying — including if the answer is no.