[{"data":1,"prerenderedAt":369},["ShallowReactive",2],{"blog-list":3},[4,179,285],{"id":5,"title":6,"author":7,"body":8,"category":163,"date":164,"description":165,"extension":166,"featured":167,"meta":168,"navigation":169,"path":170,"readingTime":171,"seo":172,"stem":173,"tags":174,"__hash__":178},"blog\u002Fblog\u002Fwhere-to-place-cold-storage-sensors.md","Where to place sensors in a walk-in (and where not to)","FreshControls",{"type":9,"value":10,"toc":153},"minimark",[11,15,20,23,26,43,47,50,76,80,83,103,107,110,129,133,150],[12,13,14],"p",{},"The fastest way to lose trust in a monitoring system is to mount the sensors in the wrong spot.\nA reading that swings ten degrees every time the compressor cycles, or a \"freezer\" sensor\nsitting in the warmest pocket of the room, teaches everyone to ignore the alerts. This is a\nfield guide to placement — the decisions that make a reading mean something.",[16,17,19],"h2",{"id":18},"measure-the-product-not-the-air","Measure the product, not the air",[12,21,22],{},"The number you actually care about is the temperature of what you're storing, not the air\naround it. Air temperature reacts instantly to a door opening or a defrost cycle; product\ntemperature moves slowly. If your alarms are tuned to raw air readings, you'll chase false\nalarms every time someone opens the door for thirty seconds.",[12,24,25],{},"Two ways to get closer to product temperature:",[27,28,29,37],"ul",{},[30,31,32,36],"li",{},[33,34,35],"strong",{},"Use a buffered probe."," A probe in a glycol or glycerin buffer bottle tracks how the product\nresponds, not how the air spikes. It's the single biggest upgrade you can make to a cold-room\nsensor, and it turns a jumpy reading into a stable one you can set a threshold against.",[30,38,39,42],{},[33,40,41],{},"Place at the point of risk."," Put the sensor where the product is most likely to go out of\nrange — not where it's easiest to reach. In most walk-ins that means away from the evaporator\nand away from the door.",[16,44,46],{"id":45},"where-readings-lie","Where readings lie",[12,48,49],{},"A few spots consistently produce readings that don't represent the room:",[27,51,52,58,64,70],{},[30,53,54,57],{},[33,55,56],{},"Directly under the evaporator fan."," You'll read the coldest, most turbulent air in the\nspace, punctuated by warm defrost cycles. It's the least representative spot in the room.",[30,59,60,63],{},[33,61,62],{},"Beside the door."," Every door opening dumps warm, humid air right onto the sensor. You'll\nsee spikes that the product two meters away never experiences.",[30,65,66,69],{},[33,67,68],{},"Against an exterior wall or on the floor."," Walls and floors have their own thermal\nbehavior and pool cold air. Mount at product height, on an interior surface.",[30,71,72,75],{},[33,73,74],{},"Buried behind stock."," A sensor walled in by pallets reads a dead-air pocket and can't be\nserviced. Keep it accessible and in circulating air at the product zone.",[16,77,79],{"id":78},"match-the-device-to-the-space","Match the device to the space",[12,81,82],{},"Placement and hardware choices go together:",[27,84,85,91,97],{},[30,86,87,90],{},[33,88,89],{},"Within reach of an access point,"," a Wi-Fi sensor mounts anywhere you have coverage and\njoins on its own. The catch is that freezer walls and racking eat Wi-Fi quickly — test the\nsignal at the actual mounting spot, not from the doorway.",[30,92,93,96],{},[33,94,95],{},"Deep freezers, far corners, and outbuildings"," are where long-range LoRa earns its keep.\nA gateway can extend coverage where the existing Wi-Fi network is limited. Test the actual\nmounting points with doors closed and normal stock in place; metal walls and racking affect\nboth radio systems.",[30,98,99,102],{},[33,100,101],{},"Racks with existing probes and door contacts"," are a job for a multi-input board. One board\nfolds several thermistors, a door switch, and an analog signal into the same fleet, so you're\nnot hanging a separate wireless sensor on every unit.",[16,104,106],{"id":105},"dont-forget-the-door-and-the-power","Don't forget the door and the power",[12,108,109],{},"Temperature is the headline, but two other signals catch problems before the temperature ever\nmoves:",[27,111,112,123],{},[30,113,114,117,118,122],{},[33,115,116],{},"Door contacts."," A propped or failed-to-latch door is the most common cause of a slow\nexcursion. Monitoring the door tells you ",[119,120,121],"em",{},"why"," a reading is drifting, often before it drifts.",[30,124,125,128],{},[33,126,127],{},"Power."," A monitored power signal turns \"the freezer is warming up\" into \"the freezer lost\npower at 2:14 a.m.\" — which is the difference between a mystery and a work order.",[16,130,132],{"id":131},"a-quick-placement-checklist","A quick placement checklist",[27,134,135,138,141,144,147],{},[30,136,137],{},"Is the sensor at product height, in circulating air, away from the fan and the door?",[30,139,140],{},"Are you buffering the probe so it tracks product, not air spikes?",[30,142,143],{},"Did you test connectivity at the real mounting location, inside the closed door?",[30,145,146],{},"Is the sensor accessible for a battery change without moving stock?",[30,148,149],{},"Are you also watching the door and, where it matters, the power?",[12,151,152],{},"Good placement is unglamorous, but it's what separates a system people trust from one they mute.\nGet the sensor reading the right thing, and every threshold, alarm, and audit record downstream\ngets more honest.",{"title":154,"searchDepth":155,"depth":156,"links":157},"",2,3,[158,159,160,161,162],{"id":18,"depth":155,"text":19},{"id":45,"depth":155,"text":46},{"id":78,"depth":155,"text":79},{"id":105,"depth":155,"text":106},{"id":131,"depth":155,"text":132},"Field notes","2026-07-08","A field guide to sensor placement in coolers and freezers, so your readings reflect the product instead of the nearest air vent or door.","md",false,{},true,"\u002Fblog\u002Fwhere-to-place-cold-storage-sensors","7 min read",{"title":6,"description":165},"blog\u002Fwhere-to-place-cold-storage-sensors",[175,176,177],"hardware","cold-chain","installation","Uh0icFQdX6xULRgd-L7GyOKExegQedNiHvok14vwG0c",{"id":180,"title":181,"author":7,"body":182,"category":273,"date":274,"description":275,"extension":166,"featured":167,"meta":276,"navigation":169,"path":277,"readingTime":278,"seo":279,"stem":280,"tags":281,"__hash__":284},"blog\u002Fblog\u002Fhaccp-temperature-logs-that-survive-an-audit.md","HACCP temperature logs that survive an audit",{"type":9,"value":183,"toc":267},[184,187,191,194,197,201,204,236,240,243,247,264],[12,185,186],{},"Most temperature records are written to be filed, not to be defended. They sit in a binder\nuntil an inspector asks for them or a customer rejects a load — and that is exactly the moment\na twice-a-shift clipboard tends to fall apart. This is a practical look at what a temperature\nlog needs to contain to hold up under scrutiny, and how continuous monitoring gets you there\nwithout adding to anyone's rounds.",[16,188,190],{"id":189},"a-record-is-only-as-good-as-its-gaps","A record is only as good as its gaps",[12,192,193],{},"HACCP treats temperature as a critical control point: you define a limit, you monitor it, and\nyou keep the evidence that you did. The weakness in a manual program is almost never the limit —\nit's the monitoring. A reading taken at 6 a.m. and again at 2 p.m. says nothing about the eight\nhours in between, and those are usually the hours that matter.",[12,195,196],{},"When a record has gaps, every gap is a question you can't answer. Continuous monitoring closes\nthem by design: a reading every few minutes, every day, whether or not anyone is on the floor.",[16,198,200],{"id":199},"what-a-defensible-log-actually-contains","What a defensible log actually contains",[12,202,203],{},"Whether you log by hand or continuously, a record that survives an audit or a claim tends to\nshare the same properties:",[27,205,206,212,218,224,230],{},[30,207,208,211],{},[33,209,210],{},"Unbroken time coverage."," Readings at a known interval, with no silent stretches. If a\nsensor drops offline, that gap is itself recorded and flagged, not left blank.",[30,213,214,217],{},[33,215,216],{},"Trustworthy timestamps."," Every reading is stamped in the site's local time, so the log\nlines up with shifts, deliveries, and defrost cycles without mental timezone math.",[30,219,220,223],{},[33,221,222],{},"The limit, alongside the reading."," The record shows not just the temperature but the\nthreshold it was measured against, so a reviewer can see compliance at a glance.",[30,225,226,229],{},[33,227,228],{},"Evidence of response."," When a reading breached the limit, the record shows the alarm fired,\nwho was notified, and when it was acknowledged. Catching an excursion is only half the story;\nproving you acted on it is the other half.",[30,231,232,235],{},[33,233,234],{},"Tamper resistance."," Values are captured by the sensor and stored automatically, not\ntranscribed by the person the record is meant to hold accountable.",[16,237,239],{"id":238},"turn-the-data-into-the-document","Turn the data into the document",[12,241,242],{},"Continuous data is only useful at audit time if you can hand it over in a form the reviewer\naccepts. The goal is a report, not a raw export — a compliance PDF for a date range, per site,\nin local time, showing readings against limits and the alarms that fired. Scheduled monthly for\nthe binder, or generated on demand when someone asks, it turns a question about history into a\none-click answer.",[16,244,246],{"id":245},"a-short-checklist-before-your-next-inspection","A short checklist before your next inspection",[27,248,249,252,255,258,261],{},[30,250,251],{},"Can you produce a full temperature history for any asset, for any date, with no gaps?",[30,253,254],{},"Does that history show the limit each reading was measured against?",[30,256,257],{},"If an excursion happened, can you show who was alerted and when it was acknowledged?",[30,259,260],{},"Is every timestamp in the site's local time?",[30,262,263],{},"Can you export all of it as a single tamper-evident document — one carrying a verification hash\nan inspector could check against the original — without rebuilding it in a spreadsheet?",[12,265,266],{},"If the answer to any of these is \"not without some work,\" the gap is worth closing before an\ninspector finds it for you. Continuous monitoring isn't about collecting more data — it's about\nnever having to explain a blank line in the log.",{"title":154,"searchDepth":155,"depth":156,"links":268},[269,270,271,272],{"id":189,"depth":155,"text":190},{"id":199,"depth":155,"text":200},{"id":238,"depth":155,"text":239},{"id":245,"depth":155,"text":246},"Compliance","2026-07-07","What a monitoring record needs to hold up when an inspector or a rejected-load claim asks for history — and how continuous logging gets you there.",{},"\u002Fblog\u002Fhaccp-temperature-logs-that-survive-an-audit","6 min read",{"title":181,"description":275},"blog\u002Fhaccp-temperature-logs-that-survive-an-audit",[282,283,176],"haccp","compliance","dZKm7QaSSU7d7YDPeoPBWhPh5clhwxBRpuqkoLSRD58",{"id":286,"title":287,"author":7,"body":288,"category":357,"date":358,"description":359,"extension":166,"featured":169,"meta":360,"navigation":169,"path":361,"readingTime":362,"seo":363,"stem":364,"tags":365,"__hash__":368},"blog\u002Fblog\u002Fhello-freshcontrols.md","Introducing FreshControls",{"type":9,"value":289,"toc":352},[290,293,297,300,303,307,310,336,340,343],[12,291,292],{},"FreshControls is real-time environmental monitoring for temperature-critical operations —\ncold storage, restaurants, grocery, agriculture, pharmacy, and the facilities that keep them\nrunning. This first post is a short introduction to why the product exists and what we are\ntrying to do differently.",[16,294,296],{"id":295},"the-problem-we-kept-running-into","The problem we kept running into",[12,298,299],{},"Most temperature-critical work still runs on trust and clipboards. Someone remembers to walk\nthe freezers twice a shift; someone hopes the reading between checks stayed in range. When it\ndoesn't, the cost shows up later as spoiled inventory, a failed inspection, or a claim nobody\ncan defend with a spreadsheet.",[12,301,302],{},"The tools on the market didn't close that gap. The serious monitoring systems were built for\nenterprises with integrators and budgets. The cheap sensors could show a number on a screen,\nbut they went quiet exactly when you needed them to shout.",[16,304,306],{"id":305},"what-we-set-out-to-build","What we set out to build",[12,308,309],{},"We wanted a monitoring tool that a small team could set up in an afternoon and actually keep\nrunning — one that treats the alert as the whole point, not an afterthought.",[27,311,312,318,324,330],{},[30,313,314,317],{},[33,315,316],{},"Status first."," The person on the floor should know if something is wrong in a single\nglance. Worst-status rolls up to the top; the detail is there when you need it.",[30,319,320,323],{},[33,321,322],{},"Alarms that reach a person."," Threshold and escalation alarms by email, SMS, and an automated\nvoice call, with automatic escalation to the next on-call until someone acknowledges.",[30,325,326,329],{},[33,327,328],{},"Records that survive an audit."," Every reading is timestamped in the site's local time and\nexportable as a compliance PDF with a report ID, data digest and file-verification instructions.\nThe SHA-256 checksum of the exact PDF file is available in Reports for comparison.",[30,331,332,335],{},[33,333,334],{},"Silence is a signal."," Alarms are evaluated on the platform, so a sensor that stops reporting\nraises an offline alarm instead of quietly disappearing — and monitoring data and login credentials\nare stored in your chosen region. Notification providers process delivery content.",[16,337,339],{"id":338},"where-this-blog-goes-next","Where this blog goes next",[12,341,342],{},"We'll use this space for practical guidance: how to place sensors so they read\nwhat you actually care about, how to turn continuous data into records inspectors trust, and\nwhat we're learning as we build. No hype, no invented statistics — just notes from the work.",[12,344,345,346,351],{},"If there's a monitoring problem you'd like us to write about, tell us on the\n",[347,348,350],"a",{"href":349},"\u002Fcontact","contact page",". We read every message.",{"title":154,"searchDepth":155,"depth":156,"links":353},[354,355,356],{"id":295,"depth":155,"text":296},{"id":305,"depth":155,"text":306},{"id":338,"depth":155,"text":339},"Company","2026-07-01","Why we built a monitoring platform for the operations where a few degrees decide whether product ships or spoils.",{},"\u002Fblog\u002Fhello-freshcontrols","4 min read",{"title":287,"description":359},"blog\u002Fhello-freshcontrols",[366,367],"company","monitoring","sEM0_h--AbAqwoheW5QuE75e2hceVnexSR_bMjV4MVA",1793363097752]