# Looking for Industrial Grade 24-bit Wireless Digital-to-Analog converters with analog outputs for a machine condition monitoring project

#### dashanan99

Joined Mar 8, 2021
4
Hi,

I am looking for Industrial Grade Wireless Digital-to-Analog Converters (DACs).

These DACs should be capable of receiving and accepting wireless digital transmission from industrial grade wireless vibration sensors, convert it into its corresponding analog/voltage signal, and send this signal out through analog/voltage output channels. The analog/voltage signal should be in 0 to ±10V (i.e. both AC and DC voltage).

I am primarily looking for 3 basic configurations:

1. 4-Channel - the DAC should be able to receive data simultaneously from 4 wireless vibration sensors with 1-axis measurement (or 1 tri-axial wireless accelerometer and 1 uniaxial accelerometers) and should be able to convert them into 4 corresponding analog/voltage outputs.

2. 8-Channel - the DAC should be able to receive data simultaneously from 2 wireless vibration sensors with 1-axis measurement (or 2 tri-axial wireless accelerometers and 2 uniaxial accelerometers) and should be able to convert them into 8 corresponding analog/voltage outputs.

3. 16-Channel - the DAC should be able to receive data simultaneously from 16 wireless vibration sensors with 1-axis measurement (or 5 tri-axial wireless accelerometers and 1 uniaxial accelerometer) and should be able to convert them into 16 corresponding analog/voltage outputs.

I am also open to purchase a combination of sensor + DAC both in case if sensor independent DAC is not available.

Any help or pointer in right direction will be greatly appreciated.

Many Thanks,

#### sagor

Joined Mar 10, 2019
439
24 bit? I must ask, what are you doing with the 0 to +/- 10V signal afterwards? if feeding some meter, you will never see 24 bit resolution, not even 16 bit. If feeding some other electronic device, it will likely be digital, or converted to digital, anyway, so why have the precise 0 to +/- 10V? Does the receiving instrument have 24 bit A/D?
24 bit resolution is almost 0.06 ppm (60 ppb). Such instrumentation is extremely hard to find, and would cost extreme amounts of money.
"Wireless" is irrelevant. It is just a medium to connect the source of a digital signal to a destination. You could use any wireless link device that can pass data packets, just as a wired interface would do. Even a wireless serial port at each end will do the data transfer. Since it is not clear what the wireless "data" looks like, or how you plan on sending it, hard to come up with suggestions for that.

#### Papabravo

Joined Feb 24, 2006
16,111
What's with the "industrial grade" crap. What does that even mean?
Over the industrial temperature range your bottom 8-bits will be meaningless hash.

#### dashanan99

Joined Mar 8, 2021
4
24 bit? I must ask, what are you doing with the 0 to +/- 10V signal afterwards? if feeding some meter, you will never see 24 bit resolution, not even 16 bit. If feeding some other electronic device, it will likely be digital, or converted to digital, anyway, so why have the precise 0 to +/- 10V? Does the receiving instrument have 24 bit A/D?
24 bit resolution is almost 0.06 ppm (60 ppb). Such instrumentation is extremely hard to find, and would cost extreme amounts of money.
"Wireless" is irrelevant. It is just a medium to connect the source of a digital signal to a destination. You could use any wireless link device that can pass data packets, just as a wired interface would do. Even a wireless serial port at each end will do the data transfer. Since it is not clear what the wireless "data" looks like, or how you plan on sending it, hard to come up with suggestions for that.
Hello Sagor,

I hope you are doing fine.

Thanks a lot for your response. I really appreciate it.

Well, I am not a very technical person, but let me try to explain what we are trying to achieve. Here is a brief summary of the project we are working on:

I need help in completing the conversion design of system where we will be converting an existing fully wired 24*7 continuous Vibration Measurement & Monitoring system, into a completely Wireless 24*7 Continuous Monitoring System as a part of one of our projects by a very large & prestigious customer who is insisting on keeping most of his existing and decade old fully wired 24*7 continuous Vibration monitoring system. This existing decade old Vibration Monitoring System (which is a proprietary system developed exclusively for this client) has a Vibration Monitoring and Analysis Software installed on a Server in control room. This software is connected to Data Acquisition Systems (DAQs) via TCP/IP network. These DAQs come in 4-Channels, 8-Channels and 16-Channels configuration. There are hundreds of DAQs installed in entire plant. Each DAQ connects to sensors via BNC cables and connectors for receiving and collecting / receiving vibration data. These DAQs can connect and communicate only with those sensors which provide Voltage/analog output via BNC cables and connectors, e.g. Piezoelectric Vibration Sensors, or ICP Accelerometers (both 1-Axis and 3-Axis). The distance between sensor locations and DAQs varies between 500 meters to 800 meters. The DAQs cannot accept input in any other form - they have been designed to accept data only through BNC cables and connectors from Piezoelectric / ICP sensors only (both 1-axis and 3-axis).

Now, the client wants us to convert this existing system into an Wireless Vibration Monitoring System without removing this proprietary software and DAQs. He basically wants us to convert the wired connection between DAQs and Vibration Sensors into a wireless one by removing these wires. The only change he has allowed us is to remove existing sensors by wireless MEMS sensors.

I researched a lot, and came to a conclusion that there is only one solution which looks practical enough to implement in this environment, and which can be described in the following steps:

1. We need Industrial grade Wireless Wireless Accelerometers/Vibration Sensors. These sensors will collect vibration data from motors, pumps, turbines etc. and will pass on this data to Wireless receivers / DACs (Digital to Analog Converters) through wireless transmission (on 2.4 GHz frequency).

• wirelessly receive vibration data from Wireless Vibration sensors,
• will convert this data from digital to analog format,
• and then pass on this Analog data output to the legacy DAQs.

3. These legacy DAQs will receive this Analog data as input, process it, and pass on the processed output to software installed in Server room for analysis, through ethernet based TCP/IP network.

Please find attached the technical specifications of 16-Channels Continuous Monitoring System (CMS), and conceptual diagram of proposed system for your kind reference.

I was searching for an Industrial grade Wireless DAC / Gateway with voltage (0 - 10v)/Analog outputs when I came across this forum and I posted my question here.

I am also open to purchase a combination of Industrial Grade Wireless Vibration Sensors / accelerometers & Wireless DAC/Gateways.

I hope this description provides enough information to make any suggestion.

Any pointers or help in resolving this issue will be greatly appreciated.

#### Attachments

• 48.9 KB Views: 15
• 249.2 KB Views: 17

#### bogosort

Joined Sep 24, 2011
674
I am looking for Industrial Grade Wireless Digital-to-Analog Converters (DACs).
You don't specify the minimum sample rate required, which is a crucial parameter. If your application can tolerate a relatively slow sample rate (<= 192 kHz), there are many 4-, 8-, and 16-channel 24-bit DACs that target the audio interface market, but can of course be used in other applications (seismology, etc.). For example, this 16-channel DAC is graded for automotive use (better than industrial grade) and costs $13 US per chip: https://www.digikey.com/en/products/detail/analog-devices-inc/ADAU1966AWBSTZ/4915075 You'll have a difficult (impossible) time finding a 24-bit DAC with a 20 V peak-to-peak output, but that's what amplifiers are for. Likewise, wireless capability is up to you to design it into the system. It should go without saying that to utilize anywhere near the 24 bits of precision, you'll need to use ultra-low noise parts with ultra-low temperature coefficients, and excellent PCB design. Even with all of that, you won't get better than 20 ENOB. Thread Starter #### dashanan99 Joined Mar 8, 2021 4 What's with the "industrial grade" crap. What does that even mean? Over the industrial temperature range your bottom 8-bits will be meaningless hash. That's a term suggested by my engineer. Thread Starter #### dashanan99 Joined Mar 8, 2021 4 You don't specify the minimum sample rate required, which is a crucial parameter. If your application can tolerate a relatively slow sample rate (<= 192 kHz), there are many 4-, 8-, and 16-channel 24-bit DACs that target the audio interface market, but can of course be used in other applications (seismology, etc.). For example, this 16-channel DAC is graded for automotive use (better than industrial grade) and costs$13 US per chip:

You'll have a difficult (impossible) time finding a 24-bit DAC with a 20 V peak-to-peak output, but that's what amplifiers are for. Likewise, wireless capability is up to you to design it into the system.

It should go without saying that to utilize anywhere near the 24 bits of precision, you'll need to use ultra-low noise parts with ultra-low temperature coefficients, and excellent PCB design. Even with all of that, you won't get better than 20 ENOB.
Hi Sir,

I may be able convince my customer as well as engineers to make some adjustments in terms of peak-to-peak output.
Rest of the details have been posted in my answer to response given by Sagor. Kindly go through that. That should explain everything. The attachment has all technical specifications for your reference.

#### bogosort

Joined Sep 24, 2011
674
1. We need Industrial grade Wireless Wireless Accelerometers/Vibration Sensors. These sensors will collect vibration data from motors, pumps, turbines etc. and will pass on this data to Wireless receivers / DACs (Digital to Analog Converters) through wireless transmission (on 2.4 GHz frequency).

• wirelessly receive vibration data from Wireless Vibration sensors,
• will convert this data from digital to analog format,
• and then pass on this Analog data output to the legacy DAQs.

3. These legacy DAQs will receive this Analog data as input, process it, and pass on the processed output to software installed in Server room for analysis, through ethernet based TCP/IP network.
FYI, the precision of your DACs need only be slightly better than the worst precision in the system. For example, if your wireless vibration sensors have only 10 bits of resolution, then there's no point in using anything above 12-bit DACs. I see in the PDF that the DAQ's analog inputs use 24-bit ADCs, which I assume is why you're considering using 24-bit DACs, but that would almost certainly be unnecessary and wasteful. First pick your sensors, then design the rest of the system specs around that.

#### bogosort

Joined Sep 24, 2011
674
I may be able convince my customer as well as engineers to make some adjustments in terms of peak-to-peak output.
You can continue to use 20 Vpp output, you'll just need an amplifier stage after the DAC, which is standard practice. For example, a given DAC may produce an output in the range -2.5 V to +2.5 V. The customer's DAQ is designed to see voltages in the range 0 to +20 V. So, following the DAC with an amplifier with gain of 4 and a 10 V dc offset will convert this 5 V peak-to-peak output to a 20 Vpp output in the 0 V to +20 V range.

#### sagor

Joined Mar 10, 2019
439
I notice the datasheet mentions "RS485 communications interfaces to server". Would it suffice to use a wireless RS485 to wireless 485 adapter? All that does is replace the wiring with a wireless serial connection.

#### Papabravo

Joined Feb 24, 2006
16,111
That's a term suggested by my engineer.
Did you ask him what it means?

#### anniel747

Joined Oct 18, 2020
1,043
Well, I am not a very technical person, but let me try to explain what we are trying to achieve.
Why are you in charge of finding a solution?

#### jeffl_2

Joined Sep 17, 2013
44
In your search you might try using the term "seismic". Texas Instruments makes seismic A/D chips with 31 bits of resolution as the ADS1228x series, and D/A chips with 24 bits of resolution as the DAC128x series. Can't help you on number of channels though because the A/D only has two and the D/A only one, you could potentially extend the D/A with high-performance sample-and-holds but you'll need to drop the output rate correspondingly, and YOU'LL be responsible for seeing that the system performance isn't adversely impacted by any design issues. These parts are essentially "state of the art" so don't expect to find much better than these for the sample rates that they support, whether or not those rates meet your needs of course at this point is unknowable.

#### Papabravo

Joined Feb 24, 2006
16,111
In your search you might try using the term "seismic". Texas Instruments makes seismic A/D chips with 31 bits of resolution as the ADS1228x series, and D/A chips with 24 bits of resolution as the DAC128x series. Can't help you on number of channels though because the A/D only has two and the D/A only one, you could potentially extend the D/A with high-performance sample-and-holds but you'll need to drop the output rate correspondingly, and YOU'LL be responsible for seeing that the system performance isn't adversely impacted by any design issues. These parts are essentially "state of the art" so don't expect to find much better than these for the sample rates that they support, whether or not those rates meet your needs of course at this point is unknowable.
I think we drove him off. I doubt he'll be back. I think @anniel747 had it right when he asked the guy why he was responsible for finding a solution. Especially after referring to "my engineer". I didn't realize people could still be possessed. There are many possible answers to the question, and each of them is more frightening than the previous one. If he had managed to tough it out he might have had a chance to learn something and might have been able to turn a 'setup' for failure into a win. I don't think we'll ever know.