Instrumentation amplifier 3

[abad]

Hi can anyone help me with the design of an instrumentation amplifier.
it is needed to amplify a signal comming from a Hall Effect sensor, I need tips on isolation and circuit design.

 

[electricuwe]

An instrumenation amplifier doesn't provide isolation. If you need isolation you have to add an isolation amplifier aor an other kind of isolation.

One basic feature of the instrumenation amplifier is to provide extremely high input impedance, symmetrical on both inputs. To my mind a Hall Effect Sensor is a device having quite a low output impedance. So an instrumenation amplifier is not necessary. A operational amplifier connected as differential amplifier should be sufficient.

For further information on analog circuits look at the application notes from Analog Devices (

http://www.analog.com)

 

[jbartos]

Suggestion: Visit

http://www.fwbell.com/hallcatalog.pdf

http://www.ti.com/sc/docs/products/msp/control/hmc/thmc45.pdf

etc. for more info

 

[xray]

abad;
I just was reading in my recent issues of EDN magazine a design similiar to this. It utilized two opamps configured from what I can recall as non-inverting configurations with the sensor placed between the non-inverting terminals. The magazine is online and the article can be seen here:

http://www.e-insite.net/ednmag/index.asp?layout=article&articleId=CA200383&pubdate=03/21/2002.

Hope this is of help?

Cy Drollinger

 

[nbucska]

Digital: Use optocoupler, e.g. NEC's PS2801-1
Analog: send spec's <nbucska@pcperipherals.com>

 

[buzzp]

I am not sure why you need this signal isolated but optocouplers would work. Differential amp should be used to cancel out common mode noise as well as amplifying the signal (depending on what your doing with this signal). I am sure the web sites will provide some good info. Allegro is another big player in hall effects sensors.

 

[electricuwe]

Without a lot of additional circuits optocouplers like suggested by nbucska and buzzp will not be very helpful if want you to have an analog signal from the hall sensor. Optocouplers are only useful to isolate digital signals. If you really need isolation look for an isolation amplifier IC which contains a modulator, a coupler (maybe an optocoupler) and a demodulator.

 

[buzzp]

Electricwue: nbucksa listed digital for optos and analog he wanted more info. I assumed isolation would be for communications.

Why do you need isolation? I guess I have not heard of a hall effect application that required isolation (I have only used for current measurement).

 

[abad]

Hi
Buzzp Well I think i made myself unclear what i was trying to do.
I am using the hall effect sensor to measure current in a conductor by using a splittable ferrite core that will clamp around the wire. the sensor will be mounted in the air gap and the signal needs to be amplified.
I needed some advise on the circuit design of the amplifier circuit and what considerations should be taken into account, what about noise in the cable will the amplifier have sufficient CMRR to cancel the noise etc.

 

[buzzp]

You should not need to isolate the signal, by the way it is installed. The differential amp will take care of any common mode noise(use a real diff amp and not a standard amp configured as one as the offsets will be a pain to deal with- I did something exactly the same about 2-3 years ago but I can get you a part number if I dig some). If the sensor is mounted away from the electronics then use shielded cable and terminate one end of shield to ground. The most critical part of your design will be the mounting of the sensor in the gap. There is little room for error. The center of the cross sectional area of the split core needs to line up with the center of the hall effect sensor(size of gap is real critical as well). This is not to major of a deal if you are not going to production with this as you can calibrate out the errors. But if you are going to production with this then you want assure the same field is seen by the hall effect everytime. Now you may be able to calibrate out the little differences even for the production units this will depend on how far off the sensor and split core are. You may choose to buffer the signal or filter it depending on what your doing and how accurate you need it to be and the junk on the signal. If this is for a VFD then I would recommend filtering out the carrier frequency unless this is important(won't affect overall reading noticably). I can't recall anything else but if I think of more I will post it here.

 

[jbartos]

Suggestion: There are locations where a strong Electromagnetic Interference is present. In that case, some isolation would be useful for the Hall sensor.

 

[buzzp]

The hall effect will not be much different than a standard component as far as picking up junk because the effective hall plate is generally very small on the part itself (This is why consistent and correct placement is so critical of the sensor in the gapped toroid). However, filtering is required due to the low level signals from sources such as those that jbartos spoke about.
Do not use to long of cable if you are indeed using cable to the hall effect and toroid. It may be better to mount the electronics around the toroid on a PCB, unless the current measurement requires a large toroid (gapped). Good luck.

 

[abad]

Thanks a lot Buzzp you have been a great help, the sensor i am making is for a project and it is not too critical where the noise is concerned, all i want to do is be able to detect the current from its magnetic field and look at the corresponding voltage output on an Ammeter or scope.
and from the results be able to accurately or within reason tell how much current is flowing in the wire.
I wonder if you know what is the best Hall sensor to do this job. does it have to be a linear Hall sensor, or would any do. I ask this because i have seen many different types some are three legged devices and others are four legged ones. if you can help it would be appreciated.

regards

 

[buzzp]

I believe you want the 4 legged variety, if I remember correctly. 2 terminals for excitation current and 2 for voltage measurement. I was using the linear type over a predetermined magnetic field (based on core and current).

I have the information but it is out in my office.I am in the house right now and being lazy. Next time I am out there I will get some part numbers.

 

[cal222]

Iam using an LT1101 Instrumentation Amplifier in my circuit and have been unable to find application notes on how to work it effectively. The Linear Technology website is not helpful. I am interested in amplifing the signal from a crystal microphone. If you know of any websites or if you have experience in hooking up instrumentation amplfiers to crystal microphones I would appreciate your help.

 

[nbucska]

Cal222:

For xtal mike you don't want instr.amp -- you want a non-inverting, low noise, high gain stage. <nbucska@pcperipherals.com>

 

[cal222]

Is that type of amplifier availiable in a chip. I am attempting to make something portable and it needs to be as small a chip as possible (Not surface mount however--I don't have the resources to solder a surface mount chip). Thanks so much!