A few buddies and I have decided to start working on a startup together, and we've reached the point where we are ready to start investing in some prototyping equipment. The relevant device is an EEG, but we've hit a bit of a snag because we're a tad overwhelmed by the options available for electrical equipment testing.

We are trying to order an Oscilloscope for testing out the components of our prototype, and there is quite a range available. We're looking for something that can measure accurately 0-50Hz Signals, but at 0-200MicroVolts.
From what I've gathered, the 0-50Hz thing isn't a problem, but the scale of microvolts certainly could be...
What oscilloscopes could do this reliably, and still provide a price that won't make our college loans hurt exponentially?

We were also looking into potential circuit simulators, Multisim being a strong candidate of choice. Though we don't know whether getting a full circuit simulator (and the millions of different editions they offer at millions of different prices) will be more cost/time efficient rather than ordering the components themselves and testing them by hand (since EEG components are pretty cheap for the most part, vs the simulator being at LEAST a few hundred dollars).

Thanks for the help!

 

Houston, we have a problem.

The lowest you will get from an oscilloscope is about 1mV. Most scopes go down to 5-20mV.

You need to amplify biomedical signals with an instrumentation amplifier that has high CMRR (common-mode rejection ratio).

web page:
http://medicaldesign.com/blog/using-oscilloscopes-microvolt-biomedical-measurements

or pdf:
Making MicroVolt Biomedical Measurements - Tektronix pdf

 

Ok cool, so instead of placing it after the instrumentation amplifier stage (pre amp), I'll just use another amp afterwards to get the signal to a proper level for the O-Scope. Thanks for the heads up regarding the general nominal level for the o-scope, that's very good to know because some of us are searching around and getting puzzling results.

That being said, I'm aware of the CMRR requirements, and the theory for all stages of the prototype design. The trouble is, none of us have ever needed to specifically order components from suppliers before, hence why we are looking at simulation software to help us get a good data base of reliable components. Though at the same time, this might be overkill because the software is very expensive, especially for the suites that have huge part databases (atleast that's what they advertise).

Up to this point, we've never needed to 'buy' o-scopes, software, and components, though we've had a fair amount of experience with them. So bridging that gap into the purchasing problems/real world spec issues has been a major hurdle.

 

Ok cool, so instead of placing it after the instrumentation amplifier stage (pre amp), I'll just use another amp afterwards to get the signal to a proper level for the O-Scope. Thanks for the heads up regarding the general nominal level for the o-scope, that's very good to know because some of us are searching around and getting puzzling results.

That being said, I'm aware of the CMRR requirements, and the theory for all stages of the prototype design. The trouble is, none of us have ever needed to specifically order components from suppliers before, hence why we are looking at simulation software to help us get a good data base of reliable components. Though at the same time, this might be overkill because the software is very expensive, especially for the suites that have huge part databases (atleast that's what they advertise).

Up to this point, we've never needed to 'buy' o-scopes, software, and components, though we've had a fair amount of experience with them. So bridging that gap into the purchasing problems/real world spec issues has been a major hurdle.

Simulation software will get you only so far. This is a real world scenario. You are not going to know what you are up against until you build some real hardware.

(PS. My MSc thesis was EEG analysis.)

 

Right, I've read multiple design articles so far where they were able to produce results using simulation software but failed completely when trying to reproduce the results using the physical components (which was frustrating because I paid for these articles).

In any case though, we will need an oscilloscope either way. After looking, there are several digital models within a respectable price range. All of them seem plenty capable of a 50Hz frequency, the voltage we will amplify before leading into the scope, but is there anything else I should watch out for in particular with using oscilloscopes for this specific application when purchasing? Is there a particular model that you liked using in your applications for a specific reason?

I apologize for the additional questions, but hearing what you specialized in, I would certainly love to know!

*EDIT* - I previously purchased a USB oscilloscope and was sorely disappointed with it's functionality due to driver update issues. So I'm hoping to use a stand alone version.

 

Are there any recommendations for scopes within the $400-1k range that could have low enough noise for this application?

I’m finding a lot online, where the specs ‘seem’ identical for desktop digital scopes but cost significantly varies.

A particular brand perhaps? I’m at the point where I’m just considering ordering a brand I used from my previous lab, but aside from matching similar aesthetics and having the bare minimum specs I’m not sure what other details I should be very cautious of.

 

Are there any recommendations for scopes within the $400-1k range that could have low enough noise for this application?

I’m finding a lot online, where the specs ‘seem’ identical for desktop digital scopes but cost significantly varies.

A particular brand perhaps? I’m at the point where I’m just considering ordering a brand I used from my previous lab, but aside from matching similar aesthetics and having the bare minimum specs I’m not sure what other details I should be very cautious of.

I don't know if this carries over into bio-medical work, but when I was scope shopping, I was advised to get 4 channels instead of 1 or 2. Seemed like overkill at first, and I've only used 3 or 4 a few times, but it was a real lifesaver the few times I needed it.

I only have limited experience with one analog and one digital scope, but for what it's worth, I love my digital scope. It's a Rigol DS1054Z, and it cost ~$400.

 

I wish I had a 4 channel scope! So go with ebeonwulf17's suggestion I think.
An it may be worth looking at this...
http://www.instructables.com/id/How-to-Hack-Upgrade-a-Rigol-DS1054Z-Digital-Oscill/
I've not had any personal experience with them but thy sound pretty good.

 

Several of us on these sites use the free LTspice simulator.
It's probably among the best of the free simulators and has a reasonable selection of models, especially now that Linear Technology has been acquired by Analog Devices.
Simulation can give you a basic idea of the circuit operation, but there are many real world problems that can only be found by actually building the circuit, especially when working with such small, noisy signals as with an EEG.

 

Great to hear! I’ll look into that scope and compare specs to some of the others I have noticed. Thank you

Interesting article on upgrading as well, if I decide to buy this model I will certainly look into this! I can imagine 4 channels can sometimes be very convenient for complex projects where isolating the problem can be a challenge

I’ve heard of LTspice before, hearing that it is free is quite nice. I will check it out this afternoon

Thanks all for the suggestions it is greatly appreciated!!

 

Start up? What is that?
Why do you want to simulate then build an EEG?
There are many circuits on the web and some of them work properly, find one and copy it or re-invent the wheel.

 

Start up? What is that?
Why do you want to simulate then build an EEG?
There are many circuits on the web and some of them work properly, find one and copy it or re-invent the wheel.

We are going to be working with a large group of projects all related to EEG data analysis and robotics (using micro controllers). We had quite a big debate about this among ourselves and others, and we came to the conclusion that we will be using it so frequently, and in so many different designs/applications, that we may as well understand the ins and outs of the device as well as skip out on the risk of paying royalties for using someone else's known product. The best way to do this of course would be to make one ourselves, and since this would be used in consumer products and not clinical studies, we can probably make it quite a bit cheaper than clinical quality EEG's...I hope.

 

Just be sure to design so to severely limit the current applied to your subjects in a fault condition.