What Is a HART Polling Address?
The polling address is the small number you set in a HART transmitter that decides two surprisingly different things at once: how a host finds it, and whether it drives a live analog current or parks at a fixed value. Getting this wrong is behind a lot of confusion where a transmitter mysteriously sits at 4 mA. This page explains what the polling address does, why address 0 is special, and how nonzero addresses relate to multidrop. It is for technicians configuring HART devices and wondering why the address setting matters so much.
HART Polling Address in one line: A HART polling address is the short-frame address that identifies a device on a loop and, on classic HART, also controls its analog behavior. A device set to polling address 0 drives a normal live 4-20 mA analog output. A device set to a nonzero polling address parks its current at a fixed low value and communicates only digitally, which is what makes multidrop possible on a shared pair.
Address 0 Versus a Nonzero Address
On traditional HART, the polling address does double duty. It identifies the device for short-frame communication, but it also determines whether the device produces a live analog signal. A device at polling address 0 is in normal point-to-point mode: it drives its 4-20 mA current to represent the primary variable exactly as a plain analog transmitter would, while also answering HART digitally. This is the default and the mode almost every single-transmitter loop uses.
Set the polling address to any nonzero value and the classic HART device changes character. It stops driving a live analog current and instead parks that current at a fixed low value, typically 4 mA, freeing the loop to carry only digital HART traffic. The device now communicates purely as digital HART, identified by its nonzero polling address. This is precisely the mechanism behind the classic surprise where a transmitter that was reading fine suddenly sits pinned at 4 mA - someone changed its polling address away from 0.
This coupling of address and analog behavior is why the polling address is not just an identifier but a mode selector on classic HART. Choosing a nonzero address is choosing to give up the analog value, and choosing address 0 is choosing to keep it. The multidrop page covers the shared-wire mode this enables; the point here is that the address setting alone flips the device between live analog and digital-only operation.
Polling Addresses and Multidrop
Nonzero polling addresses are what make multidrop work. To put several transmitters on one twisted pair, each must have a distinct polling address so a host can tell them apart, and each parks its current so the shared wire is not fighting over the analog signal. The host then addresses each device by its polling address in turn, reading them one at a time as digital HART. The polling address is therefore both the name a device answers to on the shared line and the trigger that made it give up its analog output.
Because the polling address is a short-frame identifier, it is only unique within one loop - two devices on different loops can share the same polling address without conflict, but two devices on the same multidrop line must not. This is different from the globally unique long-frame address, and it is why discovery on an unknown loop involves probing polling addresses to find who is present. The long-frame address page explains that contrast; the polling address is the loop-local name, not the global one.
For anyone configuring or troubleshooting HART, the practical rule is to treat the polling address as a deliberate choice with a consequence, not a cosmetic label. Leave it at 0 when you want a normal analog loop; set it nonzero only when you intend multidrop and have accepted the loss of the analog value. When a control system or scada tag reads the analog current, the device feeding it must be at address 0, and a device that has been moved to a nonzero address for multidrop will read a flat parked value there instead of its measurement.
Frequently Asked Questions
Why is my HART transmitter stuck at 4 mA?
The most likely cause is that its polling address was changed away from 0. On classic HART, a nonzero polling address puts the device into digital-only operation and parks its current at a fixed low value, typically 4 mA, so it no longer drives a live analog reading. Setting the polling address back to 0 restores the normal 4-20 mA analog output.
What does HART polling address 0 mean?
Polling address 0 is normal point-to-point mode. A device at address 0 drives a live 4-20 mA analog current representing its primary variable while also answering HART digitally. It is the default for a single-transmitter loop. Only when the polling address is set nonzero does a classic HART device stop driving analog and communicate digitally only.
Do multidrop devices need different polling addresses?
Yes. Every transmitter on a multidrop pair must have a distinct nonzero polling address so the host can address each one separately, and each parks its current so the shared wire carries only digital HART. The polling address is unique only within a loop, unlike the globally unique long-frame address, so two devices on the same multidrop line must never share one.
Automation services
Need help turning this into a working system?
Merobix integrates SCADA, programs Allen-Bradley and Siemens PLCs, and designs and fabricates industrial control panels.
Meeting requests are reviewed before confirmation.