• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 31 subscribers
  • Views 305 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMP1826: Compatibility with parts from "others".

Keefy
Prodigy 10 points
Part Number: TMP1826
Other Parts Discussed in Thread: TMP107

A long post I'm sorry. It was originally a technical query but I was told to post it here to get answers!

We're quite excited about the new 1-wire thermometer part TMP1826. We have had a product using the Maxim DS28EA00 for about 10 years but I've always been hoping for a replacement with better performance all this time. I was hoping this would be a drop in replacement but frustratingly it only goes about 95% of the way there. I have some queries about just how compatible it is and hope you can help. The product is an enviromental monitor for sea-ice and snow which uses 240 sensors on 5m long thin flexible PCB chain which has just three lines (Vcc, GND and 1-wire). The bus master is an ibutton OEM device (https://www.ibuttonlink.com/products/linkoem) which claims to adapt to network conditions but details are vague.

A few queries...
 
Drive capability. Is the 1-wire drive similar to the DS28EA00. We have 240 devices on a 5m PCB. The input capacitive loads look similar. The devices have a Vcc supply (i.e. not powered from 1-wire line).

Is the 1-wire system identical? Will existing (Maxim style) bus master still work? The timings all seem close for both parts.

Chain mode. The DS28EA00 chain mode is brilliant and works well for us with a PIO link between adjacent sensors all the way down the chain. However, it seems the IO pins on the TMP1826 are open-drain so we need to add (240) extra pull-up resistors to the chain. Is there a way around this?

Also, the DS2800EA chain mode allows for only the currently active device to respond to a global query by the master to identify it's position and address. The TMP1826 doesn't appear to have a chain mode so for each iteration of the search for current active device  we will have to interogate every device present to ascertain which has an active input from the previously discovered device. The search operations increase from 240 to 57600 (order of N-squared). It'll take hours but only needs doing once if we can write the chain position into the EEPROM of each device. Have I got this all correct?

What happens below -40C? Is that the bottom of the range or does the output coding still go lower? We operate in conditions to -60C. If the device still works below -40C then that would be great but obviously the accuracy is not guaranteed and of course at some point it'll give up anyway as it gets colder

I'd be keen to see drift data if/when it's available (it's TBD in data sheet).

I'd also be keen to know about linearity. We perform a single point calibration at 1C in a controlled bath. While that guarantees high accuracy at the calibration point it doesn't help accuracy at the extremes if there is a gain error or non-linear response.

And lastly, why's it not pin compatible with the Maxim part!!!?

We're really keen to use this part as our customers ask for greater resolution above absolute accuracey as this helps in the modelling of 1-D heatflux in the snow or ice body the chain is embedded in. The writeable EEPROM and automatic offset compensation is also handy. 

  • +1
    TI__Mastermind 42895 points

    Hi Keefy,

    Yes, TMP1826 is designed to be compatible with other 1-wire devices in market. You should find that it has comparable drive capability and timing specifications.

    You are correct that TMP1826 does not support the DS28 chain feature. Yes, the IO pins are open drain, so they need pull-up resistors when acting as an output. 

    If you connected TMP1826's IOs similar to the DS28 chain mode, you could toggle the IOs in order to decipher the physical order of the chain. You would still need to start by using SEARCHADDR to find the 8 byte addresses (also known as the 64 bit ROM ID) of all the devices on your chain. Once you have the addresses, you could go about toggling IOs and recording which TMP1826 sees the change. This way you could know the physical order. When you know the order, you can store it as the STACKADDR or in the 2k user EEPROM (Scratch2.) 

    Table 7-1 shows that the temperature format is capable of reporting -55C and lower. Operation is guaranteed down to -55C, accuracy is guaranteed to -40C. 

    Could you tell me, in a private message if you prefer, the company that you represent? We may be able to share additional information about drift and linearity privately.

    TMP1826 is pin compatible with MAX31826.

    Finally, regarding chain mode, have you considered TMP107? This device is intended to be used in a 3-conductor 'rope.' It has indexing as part of its protocol, which is not compatible with 1-wire.

    thanks,

    ren