- Channels
- 8, 16, 24, 32, 40, 48, 56 or 64
- Accuracy
- 0.2% ±2 digits, thermocouple excluded; the model cards state 0.2% + 0.5 °C
- Display range
- −999.9 to 1999.9
- Probes
- Type K supplied; S, R, N, E, J, T, B, PT100, PT1000 and Cu50 by type
- Sampling
- Fastest refresh 100 ms; all channels recorded once per second
- Memory
- 8 GB flash, retained through a power loss
- Display
- 5 in resistive touchscreen with keys, or 7 in capacitive touchscreen
- Live-circuit rating
- 500 V DC and 380 V AC
- Supply
- AC 85 to 265 V, 50 or 60 Hz
- Alarms
- Upper-upper, upper, lower and lower-lower per channel, with the events stored
Sixty-four channels is a different class of instrument from eight. It is not the same box with more terminals: the wiring, the scan, the memory budget and the way the data is read all change. This range is built as one family across that whole span, which is unusual at this price level.
Product overview
The range runs in eight-channel steps from eight to sixty-four, in two panels. The S series has a 5-inch resistive touchscreen with physical keys beside it; the W series is a 7-inch capacitive panel with no keys at all. Both use the same inputs and the same memory, and above 32 channels the touch panel is the only option.
The front carries the screen, a USB host socket for a flash drive, the power switch and the keypad on the keyed version. The rear carries the channel terminals in blocks of eight, the mains inlet and the fuse. Grouping the terminals in eights is a small thing that matters at 64 channels: a technician wires four banks and checks each bank, instead of working through one continuous row.
Data appears three ways. The table shows every channel with its name, its value and its running maximum, minimum and average; the bar graph puts the channels side by side against a span; the curve plots them against time. Channel names are typed at the panel, so a record reads as a set of locations rather than a list of numbers.
How the 64-channel temperature recorder works
A sampling board sits between the terminals and the 32-bit processor, and the two talk over a fast link. That architecture is why the scan can be short: the processor is not doing the analogue work itself, it is collecting from a board built for it.
Each input is isolated from its neighbours, so a switching transient on one probe does not appear on the channel beside it. The reading is compensated for the terminal temperature, linearised against the probe type assigned to that channel, and stored with the channel name and a time stamp from a hardware clock.
The memory is flash, which is why a power cut does not end the record. The trade is that the retention depends entirely on how often the instrument writes: a fast scan fills 8 GB in days, a slow one takes months. The instrument does not warn about that in advance, so the interval has to be chosen with the run length in mind.
Alarms are evaluated as each sample is taken and the events are stored alongside the data, which means an alarm can be reconstructed after the fact rather than only witnessed at the time.
How to select the right recorder
Start with the panel, not the channel count, if the instrument will be used by operators rather than engineers. Keys are easier with gloves and in a dirty environment; a touch panel is easier to read across a room and has more space for the curve. Above 32 channels the question answers itself.
Then count the points, and count the terminals as well as the probes. Sixty-four channels means sixty-four leads arriving at one cabinet, and the cable route and the panel space often decide the layout before the channel count does.
Set the interval against the length of the run. A heat-up test measured in minutes wants the fast refresh; a warehouse survey measured in weeks wants minutes, and the memory will then last the whole survey. Trying to do both with one setting is the most common way to lose a record.
Finally, list the probe types with their working ranges. The instrument will display a span far wider than any single probe can survive, and a channel that cannot reach the temperature of the process is a channel that reports nothing useful. Send the point list with types and ranges, the run length, and the panel preference, and a configuration can be proposed.
Measurement limitations and selection considerations
The accuracy figure describes the instrument and excludes the probe. At 64 channels the probe cost dominates anyway, and a matched set of probes is worth more than another decimal place on the display.
The wide display range is not a measurement range. It is the span the display can show, and it is deliberately larger than any probe, so a number on screen tells nothing about whether the probe at that point can survive the process.
Scanning is sequential, so the channels are not sampled at the same instant. At high channel counts the interval between the first and last channel of a scan is at its longest, which matters when a fault moves quickly through a machine.
The 5-inch keyed panel is limited to 32 channels, and that is a hard boundary rather than a preference. A requirement that grows past 32 forces the touch panel, and with it a different mounting depth and a different way of working. Finally, a 64-channel record is a large file: plan how it leaves the instrument, because a USB drive transfer at the end of a long run takes longer than the run itself suggests.
Every value below is taken from source documentation. Where a parameter is not stated in the available material it is marked as information required rather than estimated.
