Symptom
The driver statistics show a retry rate above a few percent for one site or for all sites; polling cycles take longer than designed; sites flicker between good and bad quality; the retry count rises at particular times of day or has grown over months.
The short answer
High Retry Count
A retry means a poll went unanswered and was sent again, and the count says how often. One site retrying while the rest are fine has a weak path, an antenna problem, a controller slow to answer, or a configuration mismatch; every site retrying has a problem at the master: its antenna system, its radio, a timeout set too short, a data rate too high, or collisions on the channel from a second master or a device transmitting unsolicited; retries at certain hours are interference; retries that grow over months are coax taking water. The radio and the SCADA driver both keep statistics per site, and the trend of retries and signal against time and against the site list is the diagnosis. Fixes follow: margin for a weak path, a corrected timeout or data rate, removal of the collision, a filter or a channel change for interference, and an antenna system repair.
Key points
- One site retrying is that site; all sites retrying is the master, the timeout, the data rate, or a collision.
- A few percent of retries is normal on radio; ten percent is a problem; fifty percent is a site about to drop.
- Timeouts shorter than the radio turnaround create retries out of nothing.
- Collisions come from a second master, a stuck transmitter, or unsolicited messages on a polled channel.
- Trend retries with signal strength: retries with strong signal are interference or timing; with weak signal, the path.
Possible causes and what to check
| Possible cause | What to check |
|---|---|
| Weak or fading path at one site | That site retries and its received signal is low or wandering. Path margin, antenna, coax, as for an intermittent path. |
| Antenna system problem at the master | Every site retries and every site shows lower signal than the record. Master coax, connectors, arrestor, antenna. |
| Timeout shorter than the round trip | Retries on every site with strong signal; the timeout in the driver against the radio turnaround plus the controller response time. Lengthen the timeout. |
| Data rate too high for the path | Retries on the longer paths with adequate signal; the radio data rate against what the path margin supports. Lower the over-the-air rate. |
| Collisions from a second master or a device transmitting unsolicited | Retries across all sites in bursts; a second poller on the channel, a site configured for unsolicited reports, or a radio with a stuck transmitter. Listen on the channel; find the extra transmitter. |
| Interference at certain hours | Retries rise at the same times daily with signal steady; the noise floor at those times. Find the source; filter or change channel. |
| Controller slow or overloaded | One site retries with strong signal and the radio hears the poll; the controller answers late or not at all. The controller scan and its port settings. |
| Serial settings or address mismatch after a change | One site retries constantly after maintenance; the port settings and the address on both ends. |
| Coax taking water at one end | Retries growing over months at one site; return loss test. Replace the coax. |
| Poll cycle too long for the site count | Retries appear as sites time out because the cycle exceeds the timeout; the arithmetic of sites, bytes, and rate. Fewer sites per channel, exceptions, or a faster channel. |
Read the statistics
| Retries | Signal | Scope | Likely cause |
|---|---|---|---|
| High | Low | One site | Weak path or antenna system at that site |
| High | Low | All sites | Master antenna system or radio |
| High | Good | All sites | Timeout, data rate, or collisions |
| High | Good | One site | Controller slow, port settings, or address |
| High at certain hours | Good | Any | Interference |
| Growing over months | Falling slowly | One site | Coax taking water |
| Bursts | Good | All sites | Collisions: a second transmitter on the channel |
Procedure
- 1
Collect
The retry rate, the error rate, and the received signal per site from the driver and the radio diagnostics, over at least a day; the poll cycle time; the timeout and retry settings.
- 2
Scope
One site, several on one repeater, or all. That decides where to look.
- 3
Timing
Retries by hour; bursts or steady; growth over weeks.
- 4
Check the settings
Timeout against the measured round trip for the slowest site; data rate against the path margins; addresses and port settings after any change.
- 5
Listen
With the master silent, the channel activity and the noise floor: another transmitter, interference, a stuck radio.
- 6
Fix
The path, the antenna, the setting, the extra transmitter, or the interference, and watch the statistics for a day after.
Timeouts and turnaround
A radio poll takes time the driver does not see: the master radio keys up, sends, and unkeys; the site radio receives and passes the message to the controller; the controller answers after its scan; the site radio keys up and sends back. On a licensed narrowband radio the turnaround alone can be a hundred milliseconds or more at each end, and a driver timeout of a second with a slow controller creates retries on the healthiest path. The timeout is set longer than the measured round trip of the slowest site with margin, and the retry count is kept small so that a dead site does not stall the cycle.
The arithmetic
A polled channel serves its sites in sequence, and the cycle time is the sum of the polls including their retries. Retries lengthen the cycle, a longer cycle means later polls, and on a driver with a fixed poll interval the later polls collide with the next cycle. A channel that was full at commissioning with no retries becomes unstable with a few percent of retries. The cure is the same as for a slow cycle: fewer sites per channel, report by exception, or a faster channel, and the retry statistics are the early warning.
Frequently asked questions
- What retry rate is normal?
- A percent or two on a well-designed radio system, from ordinary fading; five percent is worth looking at; ten percent is a problem that will get worse; above that the site is about to go offline. Cellular paths run higher and are handled with longer timeouts.
- The retries started when we added three sites to the channel.
- The cycle got longer and the timeouts and the cycle now overlap, or the new sites collide with the old on a channel that is now too busy. Recompute the cycle time; move sites to another channel, or shorten the polls with report by exception.
- Retries are high on one site with a strong signal. What is that?
- The site radio hears the master fine; the problem is behind it or in its transmitter. The controller may be answering late, the serial settings may mismatch, or the site transmitter may be weak while its receiver is fine. Check the master received signal from that site, then the controller.
- Should I increase the retry count so the sites stay online?
- More retries hide the problem and lengthen the cycle for every site; two or three is the usual maximum. Fix the cause of the retries, and let the communication alarm tell you when a site is genuinely lost.
Related topics
- Intermittent Radio PathA site that drops and returns by the hour, season, or weather: thin fade margin, foliage, a new obstruction, a loosened antenna, coax taking water, a corroding connector, interference, or a marginal supply. How to trend signal against time and weather.
- Interference on ChannelData lost on a strong-signal path: another system on the frequency, a nearby transmitter overloading the receiver, intermodulation on a shared tower, drive noise, or a stuck transmitter. How to measure the noise floor, find the source, and cure it.
- Device Times OutA polled device that answers late or not at all: how to read the driver counters, tell a dead device from a slow one, and find the timeout setting, the bus loading, the radio latency, the duplicate address, or the device that is simply too busy to answer.
- SCADA CommunicationsHow SCADA servers talk to controllers and the field: polling and report by exception, the protocols and what each is for, media from plant Ethernet to radio and cellular, poll rate arithmetic, timeouts and quality, time sync, store and forward, and security.
- Poor Signal MarginA path with too little signal above receiver threshold to survive rain, leaves, and inversions: the link budget from transmitter to receiver, the Fresnel zone, and why the margin is smaller than the study said. How to compute, measure, and add to it.
- DNP3The protocol built for utility telemetry: master and outstation, static and event data classes, unsolicited reporting, time-stamped events, and why it suits slow links that Modbus does not.
Direct contact
Have a controls question?
Reach Eric Sullivan directly about anything on this site, a controls or automation topic, or one of his personal projects.