The Micro-Tech 9105 is a belt feeder controller, not a plain belt weigher integrator: it weighs the material on the belt, then drives the feed rate to a setpoint. The 9205 is the same controller on a premium A/D board. Thermo discontinued the 9105 in June 2023, with service continuing subject to parts availability. We keep them running, and when one is past saving, talk to us about replacement options. This page covers its alarms, calibration and why feed rate wanders. We’re an independent belt weigher service company, not a Thermo Fisher agent.
What does a Micro-Tech 9105 do that a belt weigher integrator does not?
It closes the loop. An integrator measures load and speed, multiplies them into a rate in t/h and totalises it. The 9105 compares that rate against a setpoint and moves a control output until the gap closes, so it is part of how the plant runs.
The setpoint comes from the keypad or, switched to remote, from the plant, and the controller can run in automatic or manual. The output drives the feeder through an analog output (one analog I/O board comes standard) or increase and decrease contacts. Two control modes are offered: PID, and P.E.I.C. (Periodic Error Integrating Control), a periodic mode with an adjustable delay of 1 to 100 seconds.
For a conveyor scale that only measures, see Micro-Tech 9101 and 9201 error messages and calibration.
What do the common 9105 and 9205 alarms mean, and which stop the feeder?
Any of them can stop the feeder, and that is the first thing to establish: each alarm is configurable to warn, to shut the feeder down, or to be ignored.
They fall into three groups.
Deviation alarms. The rate has sat too far from setpoint for too long. No belt scale has these, and they usually report the plant: a bridged bin, a blocked chute, a drive at its limit. Check material is there and the drive is answering, then call us if it won’t settle.
Scale alarms. Load cell, speed and zero problems. Check the junction box for moisture and the cabling first, confirm the speed sensor turns and doesn’t slip, and clean the frame before any zero. Where they’re set to shut down, these stop the feed as well as the tonnes.
Controller and comms alarms. Faults in the unit itself or its link to the plant. On a feeder taking a remote setpoint, a lost link isn’t just lost data, it’s a lost demand signal. Check the far end and the settings at both ends, then call us.
What does a Micro-Tech 9105 calibration involve?
Zero, then span, then a material check, with one difference: this box drives the feeder, so isolate the feeder drive as well as the conveyor.
Work on or near a conveyor, or inside an electrical enclosure, needs isolation under your site’s procedures and trained, authorised people. If that’s not you, stop and call us on 1300 101 666.
Zero. The belt runs empty at full speed while the unit averages the empty belt load over whole belt revolutions, not a fixed time. On a feeder the plant must hand it over at full speed with nothing on it, which is why feeder zeroes get skipped. Clean the frame and check for carryback and a binding tie rod first: a zero on a dirty belt teaches the scale that dirt is normal (zero drift).
Span. A known load, one of three ways: an electronic reference that unbalances the load cell bridge with a precision resistor, test weights on the weighbridge, or a calibrated chain along the belt. The electronic route proves the cells and the constants, not the mechanics (which method suits which scale).
Material check. Enough material run without interruption to mean something, weighed independently on a scale good to about a tenth of a percent, after zero and span.
The numbers. At commissioning, the first R-Cal should come in under about 0.75%. On routine checks the manual allows 0.5% on a 9105 and 0.25% on a 9205; we work to 0.25% on both. Between our quarterly visits, site can run a zero about every two weeks and after big weather changes, and check the span, but record the span result rather than accepting it (zero, span and repeatability explained).
Why does feed rate wander or refuse to hold setpoint?
Because either the measurement is wrong or the loop cannot do what is asked, and the two need separating before anyone touches a tuning constant. Put the controller in manual, hold a fixed output and watch. Steady in manual but wandering in automatic means the loop; wandering in manual means the weighing.
Hunting usually means corrections too large or too frequent for a feeder with material in transit, where the P.E.I.C. mode and its adjustable delay are worth a look. An output pinned at its limit is a feeder out of range, not a tuning problem. On our visits we set the weighing side right and hand loop tuning back to the site's control people.
The measurement causes are the ordinary ones, and they read as control faults because the loop chases a bad number: build up, carryback, a dragging idler, moisture in a junction box, a slipping speed pickup (common belt weigher faults, erratic totals).
What speed sensor and load cell points matter on a 9105?
Two numbers catch people out: the minimum pulse rate, and the load cell cable length.
Speed. Both speed inputs are optically isolated and powered from the unit’s own supply. Aim for well over the manufacturer’s minimum of 1,000 pulses per minute at normal belt speed; we recommend about 2,000. Feeders are often slow, and a slow belt with a low pulse per revolution sensor falls under that easily, giving jumpy control and odd calibration behaviour (speed sensors guide).
Load cells. Up to six 350 ohm cells in parallel, with sensitivity selectable. Without sense leads the manufacturer allows about 61 metres of cable; with six-wire sense cable, about 914 metres. If the load cells aren’t wired into a junction box close by, run sense cable. The sense jumpers inside must match the field wiring, or the scale reads off by a temperature dependent amount that looks like drift (load cells guide).
The manuals are written in imperial units, while the display can be set to metric. We set them up in metric, and one dimension left in the other unit family is the classic quiet commissioning error (more on that trap and the belt scale codes).
How does a 9105 talk to the plant, and what happens if the link drops?
Through more routes than a belt scale needs, because it receives as well as reports. Analog outputs carry rate, load or speed, and one is usually the control output to the drive; one analog I/O board (two in, two out) comes standard. Serial runs Modbus RTU, Allen-Bradley DF1 or Siemens 3964R; Ethernet carrying Modbus TCP and EtherNet/IP is built in, and Profibus DP is an option.
The part people miss is that the setpoint may arrive over one of those links. Losing comms on a belt weigher costs you data; on a feeder it costs you the demand signal (more on plant comms).
Need a manual for your equipment?
Ask for it by exact model: the 9105 and 9205 have their own manuals, separate from the 9101 and 9201, which are written for a box that measures without controlling anything.
Talk to us about manuals for your equipment. Send the make, model, serial number and software version (a photo of the label is fine) and we’ll help you find the right document.
We don’t publish access codes for any integrator. If you’re locked out, see locked out of your integrator?
When should you call us rather than keep resetting it?
Call when the alarm comes straight back: that is the controller saying the condition never went away. Call sooner on a feeder than on a belt weigher, because a wrong number here does not just misreport the tonnes, it feeds the plant at the wrong rate. So a deviation alarm with the drive responding, a load cell fault, a zero that will not settle, or anything needing the enclosure opened (booking a belt weigher service).
Frequently asked questions
Is a Micro-Tech 9105 the same as a 9101?
No. Both weigh, but the 9101 only measures and reports, while the 9105 also drives the feed rate to a setpoint, with control alarms and a control output a 9101 has no need of.
Can you calibrate a weighfeeder without stopping production?
Not properly. The zero needs the belt running empty at full speed and the span needs the frame clear, so the feeder has to be handed over. We plan around a shutdown or a changeover where we can.
Why does my feeder hold setpoint at high rate but not low rate?
Usually the span is right for full rate but not across the lower part of the range, or the load signal at low belt loading is too small to control on cleanly. See why a belt weigher is wrong at low flow.
What happens to the feeder if the 9105 loses its remote setpoint?
That depends on how your unit was configured, which is the point. The alarm can warn, shut the feeder down, or be ignored. Check how yours is set before it matters.