How to find/make an AC to DC converter for variable AC voltage

Thread Starter

capivara

Joined Nov 8, 2018
10

Please guys I need help. I am sorry if this is obvious, I am a beginner, and I don't have anyone I can ask around.

I'm trying to connect an LVDT/displacement transducer (Omega LD 320) to a controller (Omega CNiS16D). The LVDT has AC voltage output and the controller only has DC inputs. What do I need to use to convert the variable AC voltage to DC voltage that can be read by the controller? Please be specific. The LVDT's excitation voltage is 3 V rms which I will provide I am thinking by using an isolation transformer (I found one that might work, the Hammond 166 series with 115VAC primary and 2.5VAC secondary) to step down the building electricity.
 

AlbertHall

Joined Jun 4, 2014
12,344
It means that you can't use a mains transformer as the source of the excitation voltage - you will need to buy/make a 5kHz sinewave oscillator to drive it.
 

Reloadron

Joined Jan 15, 2015
7,501
Your process controller a DPiS16 Strain Gauge Process Monitor is designed for a DC input of voltage or current. Your displacement sensor LD320 Series is designed for AC excitation using 5 KHz at 2 to 4 volts RMS and puts out an AC signal proportional to travel. I don't know how important accuracy is to you but considering what you have you may want to consider a signal conditioner which also provides your correct excitation and conditions the LVDT AC output signal to a DC which is linear to your transducer travel. DIN Rail Mount Signal Conditioners for AC LVDT Transducers would be a consideration. You can also consider a DC output LVDT. The linked signal conditioner is merely an example as there are plenty to choose from.

Ron
 

MisterBill2

Joined Jan 23, 2018
18,167
You als need to look up a lot of application information about LVDT devices, because the Omega controller is not able to work directly with an LVDT. In addition, the phase of the signal is also part of what needs to be converted into the DC position signal. There are also LVDT assemblies that include all of the electronics to provide a nice linear DC output from a DC supply input. I have used such devices and they work quite well unless you get them way too hot. Plus you save about 150 design engineering hours by getting the whole package at once.So if you are only making one system follow this advice. But if you are going to build ten thousand ot more, then keep on experimenting.
 

Thread Starter

capivara

Joined Nov 8, 2018
10
You als need to look up a lot of application information about LVDT devices, because the Omega controller is not able to work directly with an LVDT. In addition, the phase of the signal is also part of what needs to be converted into the DC position signal. There are also LVDT assemblies that include all of the electronics to provide a nice linear DC output from a DC supply input. I have used such devices and they work quite well unless you get them way too hot. Plus you save about 150 design engineering hours by getting the whole package at once.So if you are only making one system follow this advice. But if you are going to build ten thousand ot more, then keep on experimenting.
Hey thanks for your reply. I'm making just one system. I have decided against the AC LVDT (LD320) because I don't have the device for excitation/conditioning. But I have a DC LVDT (LD620) that I can use. Would I be able to connect that one directly to the Omega controller?
 

Thread Starter

capivara

Joined Nov 8, 2018
10
Your process controller a DPiS16 Strain Gauge Process Monitor is designed for a DC input of voltage or current. Your displacement sensor LD320 Series is designed for AC excitation using 5 KHz at 2 to 4 volts RMS and puts out an AC signal proportional to travel. I don't know how important accuracy is to you but considering what you have you may want to consider a signal conditioner which also provides your correct excitation and conditions the LVDT AC output signal to a DC which is linear to your transducer travel. DIN Rail Mount Signal Conditioners for AC LVDT Transducers would be a consideration. You can also consider a DC output LVDT. The linked signal conditioner is merely an example as there are plenty to choose from.

Ron
Hi Ron thanks, this is awesome. I'm making just one system. I have decided against the AC LVDT (LD320) because I don't have the device for excitation/conditioning. But I have a DC LVDT (LD620) that I can use. Would I be able to connect that one directly to the Omega controller?
 

Reloadron

Joined Jan 15, 2015
7,501
Hi Ron thanks, this is awesome. I'm making just one system. I have decided against the AC LVDT (LD320) because I don't have the device for excitation/conditioning. But I have a DC LVDT (LD620) that I can use. Would I be able to connect that one directly to the Omega controller?
Yeah, that should work fine. Your process controller:
Input Types: Analog Voltage and Current Voltage: 0 to 100 mV, 0 to 1V (+100 mV), 0 to 10Vdc
and your LVDT offers: Output options are either ±5 Vdc (LD620) or 0 to 10 Vdc (LD621).

Additionally you may be able to use the 10 VDC Excitation Out on the process controller for your LVDT but I did not read that far into it.

Ron
 

Thread Starter

capivara

Joined Nov 8, 2018
10
Yeah, that should work fine. Your process controller:
Input Types: Analog Voltage and Current Voltage: 0 to 100 mV, 0 to 1V (+100 mV), 0 to 10Vdc
and your LVDT offers: Output options are either ±5 Vdc (LD620) or 0 to 10 Vdc (LD621).

Additionally you may be able to use the 10 VDC Excitation Out on the process controller for your LVDT but I did not read that far into it.

Ron
Hi Ron, I have a follow-up question. My LVDT (L620) datasheet states that the excitation voltage should be 0 to 30 VDC at maximum 25 mA. I have it connected to a power supply that has 12 V DC and 0.85A output. It seems to be working, I am getting numbers and it is responding well when I move the LVDT up and down. What are the effects of me having a higher current than the one specified in the datasheet? Should it be working if I use it with this power supply?
 

MisterBill2

Joined Jan 23, 2018
18,167
Hi Ron, I have a follow-up question. My LVDT (L620) datasheet states that the excitation voltage should be 0 to 30 VDC at maximum 25 mA. I have it connected to a power supply that has 12 V DC and 0.85A output. It seems to be working, I am getting numbers and it is responding well when I move the LVDT up and down. What are the effects of me having a higher current than the one specified in the datasheet? Should it be working if I use it with this power supply?
The only truly negative effect of using a power supply with a higher current rating than the application requires is that you spend more money buying it. Getting a supply rated for the load current exactly assures that the supply is running at max, and that usually means a shorter time until it fails. The load will only draw the current that it needs, so that is not an issue. And it may be that next week another change will need to utilize some of that additional current capacity.
 

Reloadron

Joined Jan 15, 2015
7,501
Hi Ron, I have a follow-up question. My LVDT (L620) datasheet states that the excitation voltage should be 0 to 30 VDC at maximum 25 mA. I have it connected to a power supply that has 12 V DC and 0.85A output. It seems to be working, I am getting numbers and it is responding well when I move the LVDT up and down. What are the effects of me having a higher current than the one specified in the datasheet? Should it be working if I use it with this power supply?
OK, the LD620 Series should have a max 5 VDC output. The part number LD620 should have a suffix something like LD620-2.5 and the suffix defines the travel range, in this example the travel range is +/- 2.5mm (+/- 0.1") for a total travel of 5.0mm (0.2"). You are applying 12 VDC Excitation and the transducer is drawing 850 mA? Something does not sound right about that. Would you just be saying the power supply is capable of or rated for 850 mA?
My LVDT (L620) datasheet states that the excitation voltage should be 0 to 30 VDC at maximum 25 mA.
I believe you wanted to say 10 to 30 VDC. Even if you drop that 12 VDC to 10 VDC I don't see very much change in current. You are sure about that current? Your LD620 transducer should have 4 wires. The Red and Blue should be + and - Excitation respectively and the Yellow and Green should be + and - Output respectively. The output should change from -5.0 VDC to 5.0 VDC (a 10.0 volt change) from 0 travel to full travel of the shaft. You are sure you have the transducer connected correctly? You are also sure you have a L620-XXX transducer? What do you see on the output between 0 and full travel?

One more thing:
The transducer output is electrically isolated from the input supply. Note: to achieve calibrated accuracy the load must be 10 K Ohm between Yellow and Green. So if measuring with a DMM for example with a 10 Meg Ohm input make sure you have a 10 K resistor across the output of the transducer.



Ron
 
Last edited:

Thread Starter

capivara

Joined Nov 8, 2018
10
OK, the LD620 Series should have a max 5 VDC output. The part number LD620 should have a suffix something like LD620-2.5 and the suffix defines the travel range, in this example the travel range is +/- 2.5mm (+/- 0.1") for a total travel of 5.0mm (0.2"). You are applying 12 VDC Excitation and the transducer is drawing 850 mA? Something does not sound right about that. Would you just be saying the power supply is capable of or rated for 850 mA?


I believe you wanted to say 10 to 30 VDC. Even if you drop that 12 VDC to 10 VDC I don't see very much change in current. You are sure about that current? Your LD620 transducer should have 4 wires. The Red and Blue should be + and - Excitation respectively and the Yellow and Green should be + and - Output respectively. The output should change from -5.0 VDC to 5.0 VDC (a 10.0 volt change) from 0 travel to full travel of the shaft. You are sure you have the transducer connected correctly? You are also sure you have a L620-XXX transducer? What do you see on the output between 0 and full travel?

One more thing:
The transducer output is electrically isolated from the input supply. Note: to achieve calibrated accuracy the load must be 10 K Ohm between Yellow and Green. So if measuring with a DMM for example with a 10 Meg Ohm input make sure you have a 10 K resistor across the output of the transducer.



Ron
Sorry I meant the power supply is rated for 0.85A. I didn't realize this doesn't matter, so that's good news. And I've got a 10K resistor and everything. I think it's sorted now. The output ranges from -5 to 5V, so it all looks good. I have to figure out how to calibrate it now. Thank you all.
 

Reloadron

Joined Jan 15, 2015
7,501
Calibrating can be tricky and it really depends on how much uncertainty (error) you can afford. We had quite a few we calibrated on a regular schedule. You consider the allowable error of the LVDT and the Display System and what you can live with. We used a simple small jig and used gauge blocks by inserting them and noting the change. You want to note the uncertainty and linearity. If you have a machine shop at your disposal making a jig is easy.

Good deal on the power supply and what you are seeing.

Ron
 

Thread Starter

capivara

Joined Nov 8, 2018
10
Hey guys, I've got my DC LVDT working, and I am now purchasing gage blocks to calibrate it. Thank you so much for your help.

Another question has come up. My lab is running 12 pieces of equipment that each require an AC LVDT that measures deflection vs time. The current LVDT's we have are Schaevitz Sensors PCA-116-300 AC LVDTs with the following info:

Output at stroke ends: 360 mV/V
Input voltage: 3 V rms
Input frequency range: 50 Hz to 10kHz

Would anyone have any recommendations to the best way to connect all 12 AC LVDT's? Has anyone done something similar? Including how to provide excitation voltage, how to condition the signal, how to transform the signal to a reliable DC output, and have the system be digital early on. Is there an Omega controller that would be perfect for this? Eventually our goal is to automate our lab, and be able to send the readings to our computer network so that they can be accessed by everyone.
 

Reloadron

Joined Jan 15, 2015
7,501
This is your sensor data sheet. As to excitation voltage and signal conditioning? There are plenty of AC LVDT signal conditioners out there similar to this one. Problem is you have a dozen LVDTs and I am trying to think of an inexpensive solution. Here is another example by the same guys who make your LVDT. Just about any AC LVDT signal conditioner will provide 3 volt excitation at the frequencies you want and give you a proportional DC voltage or even current output. Personally I like the DIN Rail mounted stuff as seen in the links. I also suggest an off the shelf turn key solution rather than attempting to roll your own building circuits from scratch.

You can also find AC LVDT signal conditioners which will provide a data out rather than a voltage or current, I believe that is mentioned in my second link. This can be useful for what you mentioned:
Eventually our goal is to automate our lab, and be able to send the readings to our computer network so that they can be accessed by everyone.
When you say "eventually" and considering what this stuff cost you may want to consider eventually about now. :) When eventually comes you don't want to have to buy new stuff.

There are dozens of ways anymore to put all your data online. How would I go about it? I would take the DC outputs of my LVDT signal conditioners and scale them so your +/- 0.3 inch becomes +/- 10 VDC. Just about all of this stuff allows you to scale it. I would run my signals into a data acquisition system which allows for differential inputs. You need one channel per conditioner. Many of these things have built in web server capabilities so putting the data literally online is relatively simple these days. If you have an IT (Information Technology) group you want to talk to them. Just on a mini scale I have used an Arduino Uno (Micro Controller) with a simple cheap Ethernet Shield and run it into my wireless router so I could look at my data anywhere in the house. The stuff I used at work was a little more pricey. :)

Your main consideration will likely be budget but everything you just mentioned is very doable and there are plenty of solutions.

Ron
 

MisterBill2

Joined Jan 23, 2018
18,167
For any data acquisition system you also need to consider the secondary aspects of your selection, which include file type and format, in addition to software costs and maintenence charges. Files saved in a proprietary encrypted format are not so useful if they can't be accessed for whatever reason. That is one challenge that is usually not considered but really can be a pain point. And the yearly license charge from some places makes their cheap system very costly. These are just some cautions that may not have been so obvious initially. RR is right, but caution and close attention to details is vital.
 
Top