Any electrical engineers out there who can help.

[mgreefer]

Given that I have the chiller outside I wanted some kind of a monitor to be placed inside the house next to the tank which tells me when the chiller is on. I searched online and found a hobby circuit diagram which I have put together using diode/rectifiers and an LED. When the chiller kicks on the LED light up. The circuit essentially senses when there is a load on the circuit.

When designing the circuit I had to change the diode/rectifiers used to a more powerful ones to withstand the chiller load. I went from 3A rectifiers to 6A. The problem I am having is the current coming out of the rectifiers to the LED is so low and barely light up the LED. The AC voltage coming out of the rectifiers is 1.3Vac and the DC is 400Vdc.

My question is how to increase the current so the LED have the appropriate power to be visible.

The rectifiers I am using are the NTE5817, the LED is the standard which are usually 3.5v with a max of 20mA. I also have 12V LED although I am not sure about their Amp.

Anyone?

 

[Jennifer]

What are you using to trigger the chiller to turn on? (thermostat, controller, etc)

 

[Jim Tansey]

I think you may be way over engineering this, just put a 110 volt pilot lamp across the chiller feed. Radio Shack has lots to choose from.

Jim

 

[mgreefer]

Jennifer,

The chiller has it own thermostat that triggers it on. The circuit I described above works of the main 220V feed to the chiller, in other words it is in between the 220V outlet and the chiller plug. It detects if there is a load on the circuit, the load only happens when the chiller turns on.

Jim,

This is what I was looking for but I couldn’t find anything out there, do you have any links to share. Also the chiller is 220V.

Thanks.

 

[Dragon2115]

I'd need to see a schematic to fully understand what you have but you might want to consider using a current sensor transformer. The advantage is that if the sensing circuit were to fail it wouldn't affect the chiller. Here's a link to one of the sensors I'm thinking of. The other advantage is there's no high voltage in the circuit.

http://media.digikey.com/pdf/Data%20Sheets/CR%20Magnetics%20Inc%20PDFs/CR9500%20Series.pdf

I would take the output of the sensor and feed it into an op-amp and then use the op-amp to drive the LED.

 

[mgreefer]

Dragon2115, thanks, this is a link to the schematic. The 2 things I have changed is I replaced the 1N5404 with NTE5817 given that the 1N5404 is rated @3A, the chiller is 8A, I also eliminated the resistor because the current is so low that the LED does not light at all with a resistor.

You are right, I am not an electrical engineer but if you look at the schematic I think if one of diodes fail it would probably affect the chiller.

http://www.discovercircuits.com/PDF-FILES/LAMPMONITOR1.pdf

http://www.discovercircuits.com/H-Corner/lamp-mon.htm

I am open for suggestions, I like the current sensor transformer idea. I just want you to be aware that the chiller operates @220V 2 phase circuit.

Thanks.

 

[mgreefer]

On more thing I forgot to mention, this sensing circuit is connected only to one of the 2 hot 110v wires.

 

[Dragon2115]

Where are you taking your measurements? The 1.3vac sounds right. That should be the drop across the two series diodes. (It's also why the 3.5v LED won't light up.) But the 400 vdc doesn't sound right because with the top diode the circuit shouldn't be a rectifier.

 

[mgreefer]

Yes, both readings are being taken between the 2 diodes in serious (kind where the LED is connected according to the schematic without the resistor). The 400vdc fluctuates rapidly anywhere between 325 & max of 390+. The digital display ramps up and down rapidly.

When I tested this circuit using the original 1N5404 the LED did light up perfectly but I was afraid that over time the diodes would blow since they were rated for 3A.

 

[mgreefer]

One more thing the LED does light up but very dim and hard to see with the NTE5817s, but with the 1N5404s and the resistor it lights up perfect.

 

[Dragon2115]

Looking at the diode datasheets I see that the NTE diode has a lower forward voltage than the 1N5404, 0.9 volts @ 6A vs. 1.2 volts @ 3A for the 1N5404. So it looks like you're just not generating enough voltage across the LED to drive it. Try adding a third NTE5817 in series with the other two and reinstall the resistor. That should give you the voltage across the LED that you need to drive it properly.

 

[mgreefer]

Thanks, I very much appreciate your help, I’ll try that and see what happens. I have couple of other questions, is there a way to boost the 1.3V coming out to let’s say 12V so I can drive a bigger LED, I have couple that are 10mm with built in resistor that I’d rather use. If yes, how do I do that?

Second, was I right in changing the Diodes to the NTE5817 given the lower Amps on the 1N5404 or is this amps for output not input. Sorry dumb questions I know, I have no clue when it comes to this.

Thanks.

 

[Dragon2115]

Glad I could help.

No. I would not recommend trying to get 12 volts. If it were a single phase chiller you could do it but with two phases you'd be imbalancing them by around 10% which the motor might not like. To run those I would go with the current sense transformer and supporting circuitry.

I think you were correct in going with the higher current rating. The 1N5404 is a little light duty for your application. It would work for awhile because you're probably at or just slightly over the upper current limit and would probably burn out the diode unless well heatsinked. Especially if the power company reduces the line voltage down to 85-90 VAC as they like to do from time to time during the summer.

 

[mgreefer]

Thanks, how about just boosting the 1.3 to like 3.5 enough to light up the LED properly. Is it possible or is it going to mess up the motor as you mentioned. I was thinking taking the 1.3 wires and hooking it up to something to boost then to the LED.

 

[Dragon2115]

If you're going to go to all that bother then I'd suggest starting over with the current sense transformer. That would isolate the circuit from the AC mains and allow you more freedom to do whatever you want. It would probably require a seperate power supply like a small wall wart to power the board though.

 

[mgreefer]

I am sorry I didn’t know that it is going to be a hassle to try to boost it, I have no clue about this stuff. I agree my first choice would have been an external current sensor, however I didn’t know much about this stuff and this is where I ended. Thanks for the info you provided.

I’ll be heading to this place called “you do it yourself electronics” in Needham, my first choice would be to look for a current sensor transformer, if they have it I’ll definitely go that route, if not I’ll try to add another diode and see what happens.

Do you think another one in series would be enough though, I am just guessing since 2 are giving 1.3 instead of 1.9 which is aprox 31% signal lose, so doing the math maybe I need at a minimum 5. I am sorry I know I am way over my head in this I am sure it is not as simple as that.

 

[Dragon2115]

Hey, no problem you're doing fine. I know You-Blew-It er... I mean, You-Do-It-Electronics. They can be a little expensive but when you need only one or two items what can you do? Get three diodes so you have them but I think one will probably do it. We don't need to get too close to the max forward voltage across the LED because it's easier on the LED and allows for power fluctuations that will affect the voltage drop across the diodes. Twenty milliamps is the max current for the LED and it should be screaming bright at that level. However, visually it can sometimes be hard to tell the difference in brightness when you run the LED at 15 mA instead of twenty. So start by adding one more diode and see if that does it or if you need to add a second.

 

[mgreefer]

Thanks it worked, I added another diode and everything seems to be fine.

Surprisingly when I was @”You-do-it-yourself Electronics” they never heard of the current sensor transformers or any kind of sensors like that for that matter which would have been my first choice to avoid current interruption if any of the diodes died. It doesn’t appear to be a popular item I couldn’t even find much people selling it online.

 

[mgreefer]

So thanks again for your help much appreciated.