Wall-Wart to LPM?

[Wolfman29]

Hey guys. Question here....

So I have nearly finished my DIY <$50 LPM. It's using that one circuit on that one thread (DIY LPM for Under $50) and I have gotten a digital panel voltmeter.

I have everything wired correctly, and I am using a 5.3V 320mA wall-wart. Now, the digital panel voltmeter requires <100mA, but the TEC module requires, AT MOST, 6A and 3.8V.

Everything works fine in the circuit (tested it separately), but when I hook up my wall-wart to the panel *and* the leads to my TEC, the panel shuts off. Without the leads to the TEC connected (it's actually a LM1085 that connects to the leads of the wall wart, which then connects to the TEC, I needed the voltage to be 3.8V, not 5.3V), the panel runs fine. But once I touch the POSITIVE wire of the LM1085 driver to the power supply output, the panel shuts off and my wall-wart stops outputting ANYTHING.

Does anyone have any idea why this would be happening? Any suggestions? I want to get this thing up and running at some point, preferably today, but my wall-wart just doesn't seem to want to power it.

 

[random person]

Hey guys. Question here....

So I have nearly finished my DIY <$50 LPM. It's using that one circuit on that one thread (DIY LPM for Under $50) and I have gotten a digital panel voltmeter.

I have everything wired correctly, and I am using a 5.3V 320mA wall-wart. Now, the digital panel voltmeter requires <100mA, but the TEC module requires, AT MOST, 6A and 3.8V.

Everything works fine in the circuit (tested it separately), but when I hook up my wall-wart to the panel *and* the leads to my TEC, the panel shuts off. Without the leads to the TEC connected (it's actually a LM1085 that connects to the leads of the wall wart, which then connects to the TEC, I needed the voltage to be 3.8V, not 5.3V), the panel runs fine. But once I touch the POSITIVE wire of the LM1085 driver to the TEC, the panel shuts off and my wall-wart stops outputting ANYTHING.

Does anyone have any idea why this would be happening? Any suggestions? I want to get this thing up and running at some point, preferably today, but my wall-wart just doesn't seem to want to power it.

Have a scope? Might want to see whats happening to the voltage line when you do that. First see what type of voltage spike occurs, then scope across a small resistor and find the current response. That will give you a start.

as for the tec "But once I touch the POSITIVE wire of the LM1085 driver to the TEC"

Wouldn't that be a very bad idea? I'm probably mistaken but isn't the tec a very low impedance sensor that generates its own voltage based upon the heat transfer to its main plates? Consequently if you hook this up to a power rail you will burn out the TEC instantly and probably your supply (unless it has short protection in which case it will just shut off).

 

[Wolfman29]

Crap, that was a typo. Meant to say, once I touch the positive wire of the LM1085 to the output + of the wall wart.

I got this working now (just using two different power supplies, one is two AA batteries to power the TEC, the other is the wall-wart to power the voltmeter), but I would still like to get this all running off of a single power supply, preferably not batteries.

EDIT: And unfortunately, I don't have a scope :\ I wish I did, boy do I wish I did.

 

[JLSE]

I got this working now (just using two different power supplies, one is two AA batteries to power the TEC,

Wait, what?? You're powering the TEC?? Dont think there is supposed to be power
going to the TEC. Unless im missing something :thinking:

 

[random person]

Crap, that was a typo. Meant to say, once I touch the positive wire of the LM1085 to the output + of the wall wart.

I got this working now (just using two different power supplies, one is two AA batteries to power the TEC, the other is the wall-wart to power the voltmeter), but I would still like to get this all running off of a single power supply, preferably not batteries.

EDIT: And unfortunately, I don't have a scope :\ I wish I did, boy do I wish I did.

DS1052E::RIGOL : Best 400 usd. I ever spent.

As for what your saying.. i'm still abit confused and curious(cause I want to make one myself soon). Your trying to hook up the LM1085 output to the + terminal of your AC-DC power supply? Wouldn't you want to hook up the LM1085 input to your wall supply and then the output of the supply to whatever your trying to power? Also, what do you mean by powering your TEC? Isn't it a sensor and self powered? I only see 2 leads going from it. 03111-9J30-20CA

as for the isolation issue your running into... i've had the same happen with LCD displays... Sometimes theres some switches or mods that you can made to them to fix this issue. Problem is they don't come with a datasheet generally so you have to experiment.

 

[Wolfman29]

Okay, let me clarify some more.

I was connecting the input + of the LM1085 driver to the output of the adapter. Then, I was giving the power to the LPM op-amp board. I guess I may have been wrong, there isn't any power going to the TEC, but there is power going to the board, which is manipulated by the TEC.

Essentially, I had it like this:

AC Adapter --> LM1085 input --> LPM Op-amp board.

Because I don't really understand how the LM358 works exactly or how the board itself works (this is a single case of me building something without attempting to understand it), I was just plugging and playing (the board *does* work though with an input of 3-3.8V, which is what the LM1085 should have been giving).

EDIT:
@Random person: I don't think there is an isolating issue. I think it's just that the LM1085 + op-amp board is shorting something, drawing too much current, or something else and it's causing it to shut off.

 

[random person]

Okay, let me clarify some more.

I was connecting the input + of the LM1085 driver to the output of the adapter. Then, I was giving the power to the LPM op-amp board. I guess I may have been wrong, there isn't any power going to the TEC, but there is power going to the board, which is manipulated by the TEC.

Essentially, I had it like this:

AC Adapter --> LM1085 input --> LPM Op-amp board.

Because I don't really understand how the LM358 works exactly or how the board itself works (this is a single case of me building something without attempting to understand it), I was just plugging and playing (the board *does* work though with an input of 3-3.8V, which is what the LM1085 should have been giving).

EDIT:
@Random person: I don't think there is an isolating issue. I think it's just that the LM1085 + op-amp board is shorting something, drawing too much current, or something else and it's causing it to shut off.

Well... when it comes to analog sensors theres only 1 reason you use an op amp and that is to get signal gain. You wouldn't use it as comparator or schmitt trigger. The TEC gives out a very small voltage signal and you use the op amp to boost the gain of that signal to something read-ible by the meter. This is all high impedance(resistance) type stuff and thus your current draw should be super tiny. Thus it shouldn't be that the board is drawing to much power.

However, when I tried to create a gain circuit and feed a analog sensor signal into a LCD meter, I would always cause some high current back feed isolation issue that could only be fixed by using an external supply and or adjusting my LCD screen PCB settings such that it could handle a non isolated input. Not all screens are designed to and it messes with the logic when you use a input that is not isolated. Anyhow, i'll get back to you on that. The PCB board shouldn't draw more then 10mA max but i'd have to see the schematic first.

 

[Wolfman29]

Hmm. So you are saying that it could be a problem that the ground of the panel and the ground of the board are the same ground?

 

[Jufran88]

Whoa there, why again are you using an LM1085??

If I'm reading the OP correctly,You don't need to power your TEC, the TEC is the sensor which creates a current that the amplifier board will generate to your volt meter panel.

Laser ---> TEC sensor --> Amplifier board --> Batteries

Amplifier board--> voltmeter --> batteries

 

[random person]

"Whoa there, why again are you using an LM1085??

If I'm reading the OP correctly,You don't need to power your TEC, the TEC is the sensor which creates a current that the amplifier board will generate to your volt meter panel.

Laser ---> TEC sensor --> Amplifier board --> Batteries

Amplifier board--> voltmeter --> batteries "

Hes using it to power his op amp. However, i'm assuming he doesn't directly hook it up to his op amp since his op amp can't handle the voltage of his AC-DC convertor. Thus he is using the power regulator to step down the voltage to a voltage his op amp likes?

Hmm. So you are saying that it could be a problem that the ground of the panel and the ground of the board are the same ground?

Its possible... the circuitry in that LCD screen is far far far far more complex than anything that you would ever see on this forum. I remember reading about isolation issues with these meters before. Mind posting specifically which screen you bought?

 

[jib77]

The panel meter can not handle being on the same supply as the LPM Amp board, its not designed to just like random person said. They must be on separate supplies. There are more expensive panel meters that can.

 

[Jufran88]

LM358 has a range input voltage from 3V to 32V. I don't think you need an LM1085.

The problem could be that your voltmeter needs an isolated power supply. There are some volt panel meters that behave that way. Did you check the specs of you panel meter?

EDIT: Argh, Jib beat me to it!

 

[lasersbee]

The panel meter can not handle being on the same supply as the LPM Amp board, its not designed to just like random person said. They must be on separate supplies. There are more expensive panel meters that can.

That depend on the type/model of Digital Panel Meter he bought...
Some can have Common ground between SupplY and Signal....
and some can not... They both exists...

And I cant see why wollman needs 6A and 3.8V for the TEC....:thinking:

but the TEC module requires, AT MOST, 6A and 3.8V.

Jerry

 

[random person]

LM358 has a range input voltage from 3V to 32V. I don't think you need an LM1085.

The problem could be that your voltmeter needs an isolated power supply. There are some volt panel meters that behave that way. Did you check the specs of you panel meter?

EDIT: Argh, Jib beat me to it!

woops, my bad, you are correct on that one. Was to lazy to read the datasheet and assumed it was a 0-5 standard Cmos op amp.

Trouble with these volt/current meters is they normally have very little data specs regarding their operation. Anyone know of a good LCD display that is forsure not isolation dependent?

 

[Jufran88]

Hmm, if you want to go the cheaper route some eBay sellers offer volt panel meters that states if they need an isolated battery supply or not. Otherwise you'd probably have better luck through Mouser, DigiKey, or any other electronics distributors.

Another possibility which includes some programming is to create some code to output your readings on an arduino board similar to jibs. Or just use your multimeter :beer:

 

[Wolfman29]

Well, yes, I am using the 1085 to drop the voltage for my op-amp.

Regarding the panel... I am just confused because it's not the GND that wants to be isolated - it's the positive! The positive just seems to want to kill the wall-wart.... even though I may have figured out a solution. Gonna try a couple of things and report back later with my results.

EDIT:

Alright. So good news. I hooked up a three-pronged cable to the two prongs of my AC adapter input. Then, I used the third prong, which is GND, to connect the ground of my op-amp circuit. Further, I stuck two diodes in series with the output of the AC Adapter, and it works just fine and dandy!

Although it's weird because whenever I move the op-amp board, or move ANY wires, my LPM jumps around a lot. It may be because of shorts or something... not sure.