Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Electrical Noise Generation

Status
Not open for further replies.

Noggin

Member
For my senior design project, we are making a device to go into a vehicle. Our instructor is pushing us to test this device against SAE standards. One of tests I'm working on is electrical noise. I need to generate 0.15, 0.5, 1, and 4v signals peak to peak and place them on all input pins of the device.

The SAE standards also call for a capacitor of a minimum 100 uF to be placed at the output leads.

The schematic below is what I came up with in PSPICE to use as a circuit, I purchased a 120v@760mA to 25v@2A transformer from Radio Shack to use as the transformer. I'm hoping it will be beefy enough to handle the current.

Everything looked great until I added in the DC resistances of the inductors.... the output at 1 kHz went from a 4v swing centered at 12v, to just under a 1v swing centered on 9.7v. The 9.7v I can live with, but is anyone familiar enough with SAE standards to shed some light on this? Is it ok for the Pk - Pk voltage to drop as frequency increases? With a 10kHz input freq, the swing drops to .1v.

By the way, the V1 sin generator will have an offset voltage to keep it from dipping below 0v. And I can turn up the amplitude, but only so much. And I don't know what the actual coupling is, so I just set it to 1, but I realize that its not.
 

Attachments

  • amp.gif
    amp.gif
    5.1 KB · Views: 437
I'm not familiar with the SAE specs, but you might want to consider using your transformer to get your sine-wave(step down) and use an opamp to switch in variable gains (to get all your peak to peak values and this also takes care of DC offsets) and then take the output of that stage into a X1 power output stage if your device under test needs to draw high current from your noise generator. With this scheme in mind, you will be able to maintain accurate peak to peak voltages out to the MHz range. It's not alot of parts either and it is reasonably cheap. Are you interested?

Maybe there is something else about your design requirements that forces you to stay with the topology that you have shown in your schematic.
If it is unclear whether or not the SAE specs allow for that much attenuation(which is alot IMHO), then I would assume that you must maintain the pk-pk out that far. 1KHz is not at all fast, but like I said, I have never seen the SAE spec.
 
Thanks for the reply, I checked over the specs again and I also thought about it while I was in class. The specs call for an optional capacitor, I thought it was mandatory. It doesn't really say exactly how to implement anything for this test. All it shows is what I have attatched below. Today, I saw where it said that the capacitor is optional ONLY on sensors, and is not to be used when testing power supplies.

This is what kind of confuses more than anything though. DUT on that would seem like it would be my project, but the load box is where it says to hook everything up....

If you can point me towards a schematic of what you're talking about, if you think its better, then please do so. I'm still in school, and I am NOT good with analog stuff like this. I'm heading to a meeting with my advisor right now, I'll see what he has to say about this.
 
forgot the attatchment
 

Attachments

  • drawing.gif
    drawing.gif
    4.4 KB · Views: 439
Oh Holy Jesus that was NOT a good idea. I smelled something burning and went to hold my hand over the BJT and transformer, accidentally touched the BJT and almost passed out from the pain. The circuit was running off of a 6v 1.5 amp limited DC supply. I may go to the hospital if the pain doesn't stop by morning.
 
Noggin,

I'm not clued in to what it is you are building/doing. But I have one suggestion to help you avoid heating up components in the future. When I was in school nobody every told me about using one of the current limited power supplies. You attach your circuit to it, turn the voltage and current all the way to 0 and slowly bring your circuit up. Since I learned that technique (which now seems rather obvious), I have never burned myself or touched anything too hot. Hope your finger is better soon.
 
Thanks, the pain is stopped but I've got second degree burns, but SAE says to use a 50 frickin watt supply into .5 ohms.... that was a 9 watt supply, or about that. I really expected that the BJT wouldn't get that hot but instead the transformer would be the main power consumer. Then again, the transformer is much larger, so it wouldn't get as hot as the BJT.

I was playing with the amplitude on the signal generator though, and found that when the signal amplitued was from 0 to x, I got a really distorted signal from the output of the transformer. From x to y I got a clean signal, from y to z I got a somewhat distorted signal, and from z to max I got a clean signal again.

I neglected to turn it back down to the x to y range and instead left it at the z to max range. Don't know if that is why it overheated or if it would have overheated if I had left it turned down.

No matter what though, I was stupid for touching it, even if it was an accident.
 
Noggin said:
Thanks, the pain is stopped but I've got second degree burns, but SAE says to use a 50 frickin watt supply into .5 ohms.... that was a 9 watt supply, or about that. I really expected that the BJT wouldn't get that hot but instead the transformer would be the main power consumer. Then again, the transformer is much larger, so it wouldn't get as hot as the BJT.

I was playing with the amplitude on the signal generator though, and found that when the signal amplitued was from 0 to x, I got a really distorted signal from the output of the transformer. From x to y I got a clean signal, from y to z I got a somewhat distorted signal, and from z to max I got a clean signal again.

I neglected to turn it back down to the x to y range and instead left it at the z to max range. Don't know if that is why it overheated or if it would have overheated if I had left it turned down.

No matter what though, I was stupid for touching it, even if it was an accident.

Can you fill in some measurement numbers for your operating conditions in your schematic? There could be all kinds of trouble with your transformer & BJT.. It may be as simple as too much power dissipation in the devices. BUT, something more devastating comes to mind. One, depending on your circuit values (write down supply volts, currents and part numbers) you might be running the BJT into second breakdown(might! I dont know cuz we need the numbers)

But more seriously, the transformer you bought likely is designed for 60Hz operation. Unless you know otherwise from specs, and it is, you will run the core into saturation at higher frequencies and that will draw LOTS of current through your BJT which in turn, will overheat via second breakdown.

This is just a guess at this point. We need to examine the specs of all the parts you are using and tell us what the intended operating conditions are with volts and currents labelled.

A quick test for some supporting evidence would be to run the thing at just a few kHz and it will still get hot. If your BJT is in second breakdown it will likely be very hot also and if it is not destroyed already it will be soon. I would replace it right away with a new one (will save you troubleshooting headaches in the future)
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top