A Current Probe’s Essential Sensor Became Its Purchasing Problem

Little Bee is an open hardware probe that measures magnetic fields and, with an attachment, current in a wire. Its maker, Weston Braun, designed around a particular magnetic sensor, then found that obtaining that sensor was harder than obtaining much of the rest of the product.

His published production notes expose a distinction worth making before a hardware launch: a supplier's willingness to discuss an order is different from an order the supplier can accept and deliver. A founder needs to test the purchasing process, including account approval, quantities and ordering channels, before putting its lead time into a customer promise.

The risk was named before it became a delay

The Little Bee campaign identified the Honeywell HMC1041Z sensor as a supply risk. Braun wrote that distributors were out of stock and the normal alternative was a large reel with a long lead time. He had spoken to Honeywell about supplying a smaller quantity.

That arrangement mattered because the sensor was central to the design. An anisotropic magnetoresistive sensor, abbreviated AMR, converts a magnetic field into an electrical response. Little Bee uses that response to produce a signal an oscilloscope can display. An oscilloscope shows how an electrical signal changes over time.

The campaign also distinguishes its two measurement modes. Looking at a circuit's magnetic field can help identify behavior without inserting a measuring element into the circuit. The wire attachment provides a different, more controlled current-measurement arrangement. The campaign does not present free-position field sensing as an exact measurement of current amplitude.

The maker's public design repository identifies the B1 as an AMR-based current and magnetic-field probe. Open files make the design inspectable. They do not make its essential component easy to buy.

A purchasing conversation had several more steps inside it

In his March 19, 2021 production update, Braun said the sensor had no equivalent and described difficulty turning the manufacturer's assurance into a purchase. Account setup and an ordering-channel mismatch consumed time. He then planned to use non-authorized distributors despite price and authenticity concerns.

This is Braun's account of a historical procurement problem. It is not an independent investigation of Honeywell's systems, a current stock notice or a recommendation to buy components from an unverified seller.

The account is useful because it separates physical production from procurement readiness. Some accessories were already obtained or validated while the main board waited on the sensor. A product can look increasingly complete without moving closer to the one missing item that controls its shipping date.

A supplier representative may be helpful and sincere while another department controls approval or order processing. Treat each handoff as a dependency. Name the team, required document, expected response and escalation path. The part's technical lead time starts to matter only after the commercial process can actually release the order.

Use a ladder of purchasing evidence

For each hard-to-replace component, record the strongest evidence you possess. Do not collapse these levels into the word available.

1. A datasheet proves a specified component exists.
2. A distributor listing suggests a possible sales route.
3. A written quotation defines a price, quantity and validity period.
4. An approved account and accepted purchase order establish a real transaction.
5. A confirmed allocation and delivery schedule support planning.
6. Received, accepted parts support production.

Each step answers a different question. A quote can expire. An order acknowledgement may contain qualifications. Received parts may fail incoming checks. Keep the exact document and its date beside the planning assumption.

Ask the assembler what it needs, too. A quantity sufficient for finished units may omit setup losses, rejects, retained samples or repair stock. These quantities should come from the actual process and contract, not a generic percentage copied from another project.

Before launching, try the proposed route on a small order when commercially feasible. Completing account setup and purchasing a representative quantity can reveal administrative friction that another engineering prototype will never uncover. It cannot guarantee that later supply will remain available.

Price the fallback without pretending it is equivalent

Here is illustrative US-dollar arithmetic for a fictional 200-unit instrument batch. Assume an essential component costs US$6 through the planned supplier and US$15 through a qualified fallback. Both prices, the qualification and the quantity are hypothetical.

The planned component bill is US$1,200. The fallback bill is US$3,000, adding US$1,800. Suppose verifying and qualifying the alternate purchasing route requires another US$800. Its extra cash requirement reaches US$2,600 before any additional rejects or handling.

If postponing production creates US$650 of continuing weekly payments, four weeks costs US$2,600. That is the simple cash comparison point under these assumptions. It does not mean the fallback should be chosen automatically. Paying more for unusable or inauthentic parts can make the situation worse.

Separate three possibilities in the sheet: the same genuine component through a different seller, a genuinely substitutable component, and a redesigned product. They have different verification, engineering and test requirements. Calling all three a second source hides the work.

Also separate cash from profit. Continuing payments may include costs the business would have paid during production anyway, while a delayed shipment can postpone collections. Build a dated forecast alongside the incremental-cost comparison. SkeinMinder's batch economics explain why a cheaper parts bill can still leave the maker without enough cash or paid time.

Redesign is an option with a calendar

For a sole-source design, ask what would have to change if the component disappeared. The answer may include the circuit, enclosure, firmware, calibration, documentation and acceptance tests. A pin-compatible package alone does not establish functional equivalence.

Put the redesign tasks in dependency order. An alternate sensor might require new calibration equipment before production can be accepted. A firmware change may need verification against the new hardware. Price a usable replacement design rather than the first alternate part someone finds in a catalog.

These are planning questions for competent technical staff. This article does not recommend a replacement for the HMC1041Z or certify the Little Bee design. The value is learning what a replacement would entail while there is still time to make a deliberate decision.

MNT Reform's manufacturing map offers another view of dependencies distributed across suppliers. Little Bee adds a purchasing question inside that map: has the critical supplier relationship progressed far enough to support the promised date?

In the next hour, choose the three least replaceable components in your bill of materials. For each, attach the latest purchasing evidence, identify the next unfinished step and write the fallback's qualification work. Leave uncertain delivery dates visibly uncertain.

Braun's notes do not prove a universal stocking rule, and one surviving crowdfunded project cannot represent all small hardware launches. They show a narrower mistake that a founder can prevent: treating a helpful conversation as if the essential part were already on its way.