{"product_id":"antminer-z15k-hashboard-repair-service","title":"Antminer Z15K Hashboard Repair Service","description":"\u003cp\u003eAntminer Z15K hashboard repair covers the board that stopped hashing and the board still hashing short. The miner lists each chain with its chip count: any count short of what a healthy board reports is a fault, a chain gone from the list is the same fault further along. We repair at component level in the USA, 4-5 business days.\u003c\/p\u003e\n\n\u003cp\u003eAntminer Z15K hashboard repair is new work on new hardware. The machine began shipping at the end of August 2026, and the first boards are only now reaching benches anywhere. We have been repairing this Equihash line in our USA-based lab since 2025, and we take Z15K work from the first machine in: board-level diagnostics, component-level repair, cleaning and thermal work, then validation under load. Working out which of the boards inside has failed is not your job. You send the miner, we find it.\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eWhat we know about this model, and what we do not.\u003c\/strong\u003e The Z15K is weeks old. Bitmain has published no service documentation for it, no board reference and no chip data, and no independent lab has published a teardown. Everything in the table below that we cannot yet verify is marked as not published rather than filled in from a machine that merely looks similar. We update this page from the first Z15K on our bench, and what we add will be what we measured.\u003c\/p\u003e\n\n\u003ch2\u003eTechnical Diagnostics and Common Issues with Antminer Z15K\u003c\/h2\u003e\n\n\u003ch3\u003eQuick Symptom Checklist\u003c\/h3\u003e\n\n\u003cp\u003eThese are the readings that bring a board of this generation to us. Counts and rates differ from machine to machine, so match the shape of what you see, not the digits.\u003c\/p\u003e\n\n\u003cul\u003e\n\u003cli\u003eOne chain is gone from the chain table altogether - not zero, absent - and the miner settles near two thirds of its rated 525 KSol\/s\u003c\/li\u003e\n\u003cli\u003eOne chain reports fewer chips than the others on the same machine\u003c\/li\u003e\n\u003cli\u003eThe rate falls away as the machine warms up and comes back after it has been off for an hour\u003c\/li\u003e\n\u003cli\u003eThe miner restarts on its own within minutes of reaching full rate\u003c\/li\u003e\n\u003cli\u003eAccepted work drifts down over weeks while nothing in the interface turns red\u003c\/li\u003e\n\u003cli\u003eThe machine stops on temperature, or reports a fan it cannot find\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003cp\u003e\u003cstrong\u003eOn log lines, plainly.\u003c\/strong\u003e The exact strings a Z15K writes in its kernel log are not published anywhere yet, and we will not print lines from an older machine here and let you assume they are this one's. If your Z15K has written something you do not recognise, send the log with the miner. We publish what we have verified, and this section grows as the first machines come through.\u003c\/p\u003e\n\n\u003ch3\u003eModel-Specific Patterns We See on Antminer Z15K\u003c\/h3\u003e\n\n\u003cp\u003e\u003cstrong\u003eHeat that shows in the numbers before it shows in the interface.\u003c\/strong\u003e A Z15K puts roughly 830 W through each of its boards inside an air-cooled two-fan chassis. What the owner sees is a rate that sags as the machine warms and recovers overnight, or a chain that only drops out once the machine is hot. It is not the pool and it is not the network, both of which fail the same way cold. It is on the board, and which part of the board it is on is bench work.\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eA chain that is short, and a chain that is gone.\u003c\/strong\u003e These look like different problems to an owner and they are two stages of the same one. A dead ASIC that still passes the chain along leaves the board in the list with a lower count, hashing on what is left. A dead ASIC that stops passing the chain along takes the whole chain out of the list. The first is often mistaken for a pool problem, the second for a dead board. Neither is.\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eThe machine that restarts under load.\u003c\/strong\u003e On a machine drawing this much through three boards, a restart minutes after full rate can come from the board or from the supply feeding it, and from the dashboard the two are not distinguishable. This is the single strongest reason we ask for the complete miner on this model rather than a bare board: the machine tells us which of the two it is, and a board on its own cannot.\u003c\/p\u003e\n\n\u003ch3\u003eHardware Notes\u003c\/h3\u003e\n\n\u003ctable style=\"width:100%;border-collapse:collapse;table-layout:fixed;font-family:sans-serif;font-size:14px;margin:16px 0 24px;\"\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width:32%\"\u003e\n\u003ccol style=\"width:68%\"\u003e\n\u003c\/colgroup\u003e\n\u003cthead\u003e\n\u003ctr style=\"background-color:rgb(248,248,248);border-bottom:2px solid rgb(238,238,238);\"\u003e\n\u003cth style=\"padding:12px;text-align:left;vertical-align:top;\"\u003eSpecification\u003c\/th\u003e\n\u003cth style=\"padding:12px;text-align:left;vertical-align:top;\"\u003eDetails\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eAlgorithm and cooling\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eEquihash (Zcash), air-cooled, two fans\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eRated hashrate\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003e525 KSol\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003ePower draw\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003e2483 W at the wall, 200-240 V single phase\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eHashboards per miner\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eThree, as the machine is sold. We confirm the count on the first machine we open and update this row from what we count, not from a listing\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eWhat one board is worth\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eAbout a third of the machine, roughly 175 KSol\/s. Whether the miner keeps hashing with one board out is decided by the firmware Bitmain ships for this model\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003ePower per board\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eRoughly 830 W, which is why thermal faults are the first thing we look for on this model\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eChips per board\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eNot published by the vendor. We will state the number once we have counted it on a board ourselves\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eChip grouping\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eNot published by the vendor. The miner names its own grouping in the log, and that is the name this table will use\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eChip family\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eNot published by the vendor, and not assumed here\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eHashboard part number\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eNot published by the vendor. We read it off the board at intake and record it with your repair\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\n\u003ch3\u003eWhat the Log Is Telling You\u003c\/h3\u003e\n\n\u003cp\u003eYou do not need to act on any of this. It is here so that you know what state the board is in before you decide to ship it.\u003c\/p\u003e\n\n\u003ctable style=\"width:100%;border-collapse:collapse;table-layout:fixed;font-family:sans-serif;font-size:14px;margin:16px 0 24px;\"\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width:30%\"\u003e\n\u003ccol style=\"width:36%\"\u003e\n\u003ccol style=\"width:34%\"\u003e\n\u003c\/colgroup\u003e\n\u003cthead\u003e\n\u003ctr style=\"background-color:rgb(248,248,248);border-bottom:2px solid rgb(238,238,238);\"\u003e\n\u003cth style=\"padding:12px;text-align:left;vertical-align:top;\"\u003eWhat you see\u003c\/th\u003e\n\u003cth style=\"padding:12px;text-align:left;vertical-align:top;\"\u003eWhat it usually means\u003c\/th\u003e\n\u003cth style=\"padding:12px;text-align:left;vertical-align:top;\"\u003eWhat we do with it\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eA chain absent from the list\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eThe chain has stopped answering rather than slowed down. Several different faults produce it\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eWe bring the chain up on the bench and follow it until we find where it stops\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eA chain short of chips\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eAt least one ASIC on that board is no longer answering\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eWe isolate the dead ASIC and establish what took it out before anything is replaced\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eRate sags when hot, returns when cold\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eHeat is staying in the silicon instead of leaving it\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eWe clean the board, renew the thermal interface and then run it hot on purpose to see whether it holds\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eRestarts minutes after full rate\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eEither the board or what feeds it cannot hold the load. From the dashboard these look identical\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eWe load the board in the machine it came out of, which is why we ask for the complete miner\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"border-bottom:1px solid rgb(238,238,238);\"\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eAccepted work drifting down for weeks\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eAn ASIC degrading rather than failing outright\u003c\/td\u003e\n\u003ctd style=\"padding:12px;vertical-align:top;word-break:break-word;\"\u003eWe run the board long enough for the counters to move, because a single pass will not show it\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\n\u003ch3\u003eDiagnostics Focus\u003c\/h3\u003e\n\n\u003cp\u003eTwo things set the order of work on this model. The board is inspected before anything is switched on, because at roughly 830 W per board a fault that would be cosmetic on a lighter machine does not survive being powered twice. Then the chain is walked end to end, because the difference between one dead ASIC and a chain that has stopped answering decides whether the board needs chip work or the fault is somewhere along the chain, and nothing on the dashboard separates the two.\u003c\/p\u003e\n\n\u003ch2\u003eOur Professional Repair Process\u003c\/h2\u003e\n\n\u003ch3\u003eGotchas\u003c\/h3\u003e\n\n\u003cp\u003eWe do not test a Z15K board on a supply sized for the older machines in this family. A board starved of power behaves exactly like a board with a fault, and a false diagnosis on a machine this new is one that would follow the model around for months.\u003c\/p\u003e\n\n\u003cp\u003eWe photograph and record each Z15K board as it comes out, before any work starts. For this model there is no published reference to check a board against later, so we build that reference ourselves as machines arrive. It also means that if your board has been opened by someone else before us, we can usually tell.\u003c\/p\u003e\n\n\u003ch3\u003eTypical Service Scenario\u003c\/h3\u003e\n\n\u003cp\u003eMost boards that reach us on a new model come in for one of three reasons. The machine ran in dusty or humid air and the thermal path gave up first. Somebody chased the last few percent with an aggressive tune on hardware that already runs hot from the factory. Or a first repair was attempted elsewhere, and the board arrives with the original fault plus whatever was added to it.\u003c\/p\u003e\n\n\u003ch3\u003eWhat Happens After Intake\u003c\/h3\u003e\n\n\u003cp\u003eIntake, incoming inspection, diagnostics, component-level repair, then cleaning and thermal interface replacement. Every board is tested on our STASIC and ASIC REPAIR equipment where the board profile allows, and every repair is finished with a minimum of one hour under load in a live miner. Extended burn-in over several days is available as a separate add-on.\u003c\/p\u003e\n\n\u003ch3\u003eWhat We Need You to Send\u003c\/h3\u003e\n\n\u003cp\u003e\u003cstrong\u003eSend the complete miner, even when only one board has failed.\u003c\/strong\u003e We have no Z15K of our own to test a bare board in, and no bench fixture exists for this model, so a board sent on its own cannot be proved to work after it is repaired. Your machine is the only place a Z15K board can be validated, which is why it comes with the board. This is the one model-specific thing we ask of you, and it is why a repaired board leaves here having run in the miner it will go back into.\u003c\/p\u003e\n\n\u003ch3\u003eDiagnostics and Validation Equipment\u003c\/h3\u003e\n\n\u003cp\u003eBoards are diagnosed on STASIC and ASIC REPAIR testers, which give us the chip-level picture before any rework begins. Final validation is always under load in a running machine rather than on a tester alone, because a board can pass a bench profile and still fold when it is hot and working.\u003c\/p\u003e\n\n\u003cp\u003eAll repairs are performed personally by Alex, who has been servicing ASIC miners since 2019. \u003ca href=\"https:\/\/antminer-repair.com\/pages\/aleksandr-muranov\"\u003eSee credentials →\u003c\/a\u003e\u003c\/p\u003e\n\n\u003cp\u003eContact our repair team today and get your miner back to full power.\u003cbr\u003e\u003ca href=\"https:\/\/antminer-repair.com\/pages\/contact\"\u003ehttps:\/\/antminer-repair.com\/pages\/contact\u003c\/a\u003e\u003c\/p\u003e","brand":"Antminer Repair","offers":[{"title":"Default Title","offer_id":54683411677457,"sku":"HB-Z15K","price":350.0,"currency_code":"USD","in_stock":true}],"url":"https:\/\/antminer-repair.com\/products\/antminer-z15k-hashboard-repair-service","provider":"Antminer Repair","version":"1.0","type":"link"}