can a resistor solve this problem? 3

[mick43]

I want to operate a 24 volt device from a 48 volt power supply.

Manual says device (a carbon dioxide sensor) will operate at 18-30 vdc (or20-28 Vrms AC) and power consumption is listed at "less than 2 watts @ 24 VAC."

Most of the power consumption seems to be from a "heater" that pulses on and off (about two times per second) to keep an exact temperature in the sensor module.

When running with a 24 vdc power supply, my digital meter (set on 10 amp) flashes between .04-.05 of current.

However, at a preset level of carbon dioxide, a control relay closes. When the relay is closed, the reading flashes between .06-.07.

Is it possible to drop the voltage into the 18-30 volt range with a resistor when the load varies like this?

Thanks for any help with this question.

Mick43

 

[BrianR]

No, it is not practical to use a resistor when the load varies so much. The best solution is to use a switching regulator designed for 48V i/p. This voltage is a bit high for linear regulators and the regulator would dissipate the same power as the load and get quite hot.

 

[mick43]

Thanks Brian for your quick response. Can you point me toward as website to find a "switching regulator designed for 48V i/p'?

 

[dhwilliams]

If you fit a 440 ohm resistor in series with your sensor (2 x 220 ohm 2W in series) the sensor voltage would be about 20V with the relay on and 28V with the relay off. These both fall within the 18-30V range so, as long as the sensor can tolerate the change in voltage when the relay operates, it should work. Try it!

If you don't like that method (or it doesn't work), use a 1N5359B 24V Zener Diode in series with the sensor. That will "lose" the 24V you don't need. Connect the cathode to the positive rail of the 48V power supply, the anode to the sensor positive terminal and the sensor negative terminal to the 48V power supply negative rail.

Dave

 

[BrianR]

If your power supply is DC then a DC-DC converter is what you need. This is a switching regulator too. There are a lot of options depending on case style etc but a couple of places to start with are:

http://www.v-infinity.com/

VAWQ3-Q48-D12

http://www.powerstream.com/

PST-DG-48-24

 

[mick43]

To: Dave (dhwilliams),
Thanks for the tip. I installed a 24 volt Zener Diode, 5 watt, as you suggested, and it works! It does get hot to the touch, is that ok? Maybe I need a 10 watt Zener?

Since the Zener worked for the co2 sensor application, I got a second Zener to drop 48 vdc volts for a 5 vdc relay. But, that didn't work.

A 5.1 volt Zener, 5 watt, was connected in series with a 5 volt (100 ohm) relay coil, cathode to the positive voltage. The relay closed normally, but the meter measured only about 1 volt. Soon I could smell the relay coil getting hot.

Do I need a bigger wattage Zener? Or, maybe that's too much of a voltage drop and a Zener can't get rid of 43 volts.

Thanks for any suggestions.

mick43

 

[OperaHouse]

2W doesn't sound like a lot of heat until it is put in a small location. The power rating of the zener is based on it having sufficient radiating area attached to it with some air flow. The problems I see are this will be placed in an outlet box and it will have to be insulated. If you place 2, 3, or 4 zeners in series that add up to the same voltage, it may provide enough surface area to reduce the temperature even if the devices are placed in heat shrink tubing. Keep the same wattage zener for the larger body and thicker leads. Another option is soldering both zener leads to a large brass washer or crimp lug to act as a heat sink and then heat shrink over that.

 

[buzzp]

For 5V use the 24VDC bus and use a linear regulator for supplying 5V like an 7805. Less heat dissipation by far. Have not had the need but they likely have regulators that can operate with 48V input (all the ones I have dealt with generally have a 30V max input voltage).
I would use a zener to drop the voltage below 30V, if you cant find a regulator with more than a 30V input, then use a regulator/s to produce the voltage you want. It will be much better regulating using these regulators rather than a zener regulator. And, it wont get as hot as just a zener will.
A 2W 5.1V zener on a 48VDC bus is going to smoke. You need more in series or a larger wattage package like 10W.
Your confused on what a zener does in parallel and in series, I believe. In series, the zener will drop the amount of voltage equal to its zener voltage (5.1V in your case). If the cathode is hooked to the 48V bus and the anode hooked to the negative of this bus, it regulates the voltage to 5.1V for voltages across the zener (or anything hooked in parallel with it).

 

[dhwilliams]

A 5W Zener of this type in free air should easily handle the 2W without getting so hot it shortens its life - if you want to run cooler, use a stud-mounted zener on a couple of square inches of aluminium but bear in mind that the aluminium is at the same voltage as the stud terminal.

You should have used a 43V zener to drop the voltage from 48 to 5. Calculate the current through the coil which will also be flowing through the zener then multiply this by the 43V to give the Watts - double this to get the zener wattage rating and, if necessary mount it on a heatsink.

Dave

 

[mick43]

Thanks to everyone who sent me a response. I have learned something valuable from each post.

About Zeners in series, should I step the voltage down in steps, like first a 30 volt, followed by a 15 volt, followed by a 5 volt Zener?

About Zeners in parallel, do two or three of the same value in parallel share the heat load?

About Zener and linear regulator circuits, one general question:

When a Zener fails, what happens to the voltage? Can the full voltage ever flow through the circuit?

And about lineal regulators, the same question. What happens when the device fails? Can the full voltage ever flow through the circuit?

Mick43

 

[buzzp]

"About Zeners in series, should I step the voltage down in steps, like first a 30 volt, followed by a 15 volt, followed by a 5 volt Zener? "

The heat will be a little better dispersed doing this(not all in one package), although all will have the same current flow.

"About Zeners in parallel, do two or three of the same value in parallel share the heat load?"

Yes, this will equally share the heat load (all zeners are equal, especially zener voltage).

Zeners can fail either way, shorted or open. The ones I have seen failed were open circuited so the voltage is not regulated across the zener.

The other regulators generally have overload protection built in so if the current (load) is to large they shut down. I don't recall diagnosing many failures but when they fail, I believe, they don't generally apply any voltage to the load.

One note: voltage does not flow, current does. Voltage is just a potential difference giving the system the ability for current to flow if a path is made.

Are you working on a project at home?

 

[OperaHouse]

Zeners are not perfect and their voltage can vary 5-10%. The lowest one in parallel will see the most current. In non critical parallel applications, a small resistor in series with each one to develop a small voltage will help in current load sharing.

They should be the same voltage when you put them in series or else the highest voltage one will heat up the most.

I kinda zoned out and didn't comment about the 5V relay earlier. You can learn a lot by burning up stuff, but believe safety devices are not the place to learn. Damage to coils may not show up immediately. That 1V sounds like some damage was done. Be conservitive.

 

[mick43]

Again thanks for spending time to help me. I'm sure the level of my questions shows that I know very little about solid state electronics and I appreciate your patience to explain basic principles to me.

My only knowledge about electronics comes from fifty years ago when I was a teenager. Then I was an avid ham radio operator and I built a few transmitters and power supplies. I suppose that's what gave me confidence to try and do this project.

I worked in mushroom production business for 20 years and now I'm trying to put together a carbon dioxide control system for mushroom growers, using variable speed DC fans and blowers with a carbon dioxide sensor/controller.

My questions arise because I found at a very good price, several hundred brand new 48 vdc blowers with built-in 0-10 volt speed control circuits and I want to interface them with the co2 sensor/controller module that I am familiar with. I also found a supply of reasonably priced 48 vdc power supplies. So, the questions arose about how to simply connect these three things, the sensor/controller, the power supply, and the blower.

The co2 controller has two control circuits, an on/off relay, also a 0-10 vdc that can be set to span a selected range of co2.

In this system, the relay is used to activate the 48 volt supply to the blower at a preset level of co2. Then the blower begins to operate from the 0-10 volt analog signal, making the blower run faster and faster if the co2 continues to rise.

So, the Zener solved the problem of supplying 24 volts to run the sensor/controller module from the 48 volt supply. I can deal with sinking Zener's heat. Only thing I'm wondering is if Zener fails and sends 48 volts to the sensor/controller, several hundred dollars damage will likely result. How often/why do Zeners fail?

Now, about the relay circuit. The onboard relay on the co2 sensor/controller is too small (rated 24 vac 3 amp) to switch the 48 vdc 4 amp output from the power supply to the blower. So I need to add a relay.

Yesterday I bought another 24 volt Zener, a 5 vdc relay and a 5 volt linear regulator that outputs five volts. I think these components will do the job and on Saturday I'll hook it up for a test.

Again thanks to all of you! All I can offer is to help you if you ever want to know anything about growing mushrooms.

Mick43

 

[dhwilliams]

Like most modern semiconductors, Zeners are very reliable especially when they run cool! I don't know what the statistics are but I would assume that it could fail as a short-circuit so you need a way to protect your sensor. Protection circuits can get more complicated than the circuit you are trying to protect so I would suggest a simple circuit as follows:-
Put a fuse between the positive supply and the zener cathode (e.g. 500mA, fast acting) then connect a 30V Zener in parallel with the sensor - cathode to the positive of the sensor (i.e. the anode of the 24V zener), anode to 0V. This will not conduct under normal circumstances but if the 24V zener shorts, the voltage across the sensor will increase very rapidly. When it reaches 30V, the new zener will start to conduct and, as there is no current limiting, the fuse will blow - hopefully fast enough to protect the sensor.

Why bother with a 5V Regulator and 5V Relay? Couldn't you just buy a 48V relay (or a 24V one) and use the contacts in the onboard relay to operate it?

If the blower is 48V, how does the 0-10V analog signal change its speed?

 

[buzzp]

Test post

 

[buzzp]

Been having trouble posting replies for some reason.
Heat reduces the life of any electronic device, the cooler the better, generally. The hotter your zeners are the more likely they are to fail. If the zener fails open circuit, your sensor sees 48V. A zener, when conducting, is almost a short circuit already so a zener failing shorted would cause your upstream fuse/breaker to blow/open.

You might find a regulator with a 48V input but most have max of 30V input. It would definately be worth checking into because these will run cooler and you will have less chance of your sensor going south due to regulation problems (zener will allow 48 V to sensor when fails). The other regulators have short circuit protection so they shut down if the load is to high. I have never seen one fail in a mode that applied the input voltage (48) to the regulators output. Could happen, I suppose.

 

[dhwilliams]

buzzp, I agree that in a "normal" zener circuit, failing open-circuit would allow the full voltage to be applied to the load but when the zener is in series with the load, as I described earlier, it would actually disconnect the voltage. It is the short-circuit failure that would cause the full voltage to be applied to Mick43's sensor.

 

[mick43]

Dave asked, "Why use 5vdc for the relay instead of 48 or 24 vdc, which is already present?" I thought the 5vdc would be safer, but my judgement about that might be incorrect.

I tought like this: Since the onboard sensor relay's contacts are rated 24 vac 3 amp, I thought it was not a good idea to send 48 vdc through them. Maybe 48 vdc is really ok, but I thought I should drop the voltage.

Why not 24 volts? I thought 5 vdc would be safer and the relay cost is the same. The sensor/controller module is connected to the power supply by a 100' cable with 6-pin din connectors. I worry about short-circuit if someone trips over the cord or steps on the din plug/socket by accident, because if 24 volts shorts to the 0-10 volt circuit, it would burn out the sensor's output stage. By making the relay control 5 vdc I thought it would lessen the chance of damage by reducing the number of +24 volt wires in the cable from two to only one.

I searched for an economical 4-5 amp solid state relay that could switch 48 vdc, but couldn't find one. Red color dc/dc SSRs are rated 3 amps, but the next size I could find was 10 amps and it was bulky and a bit expensive. It there something in between?

Buzzp -- Highest value linear regulator I could find was for 45 volts. So, I got a Zener to drop 48 vdc before the lineal regulator.

Dave -- Thanks for telling me how to add protection to the sensor supply circuit. I will do it.

Dave asked how the analog signal changes the 48 vdc motor speed. Answer is that I don't know, it is built into the motor. Motor has two wires for 0-10 volt control signal and when about one volt is present, then after a few seconds, the blower begins to run at half speed (200 cubic feet per minute). As volts increase, motor spins faster until it produces 400 cfm at 10 volts.

I have been curious to see the components, they must be small, because the motor doesn't appear to be oversized (compared to typical dc motors for propeller fans, such as those on computers) but the motor is centered inside the squirrel cage and to open the motor I'd have to remove the bearing and probably destroy one blower just to find out what's in there. And then, I probably wouldn't understand what I was looking at anyway. These fine 48 vdc blowers were so darn cheap, I couldn't resist trying to figure out how to use them. I think they were manufactured for commercial aircraft ventiulation systems.

My long range focus is to develop a co2 control system using variable-speed 24 vdc fans, like those manufactured for cooling computers and telecommunications equipment. During tha past couple of months, I made some prototype variable-speed 24 volt fan systems using this co2 sensor/controller with a dc to pwm modulator and they work very well.

The dc to pwm modulator I use is rated 6.5 amps (8-35 vdc), which is a bit small. I sure would like to find a more powerful dc to pwm module that accepts a 0-10 volt control signal, one rated at 12 amps instead of 6.5 amps. Does anyone know about such a device?

Mick 43

 

[OperaHouse]

Next lesson. You are on the edge of having real problems switching 48V @4A. Above 30V DC, a contact begins to have the ability to maintain an arc when it is opened. If you look at the manufacturers spec you will see that the relay is not rated for 48V DC. Some people cheat a little and put the contacts in series to increase the efective gap.

You can build your own power supplis by using a cheap universal PWM controller IC like the SG3524 or TL494 and a couple of pwer FETs. Many small DC fans are now brushless and wil not accept a variable supply. This is because they are basically poly phase AC motors operating at a fixed frequency. I seriously doubt the need for variable speed anyway. A lower cost system would be for a microprocessor to operate the fixed speed fans for so many minues each hour based on the amount of CO2. This can be done by a simple 8 pin micro and only about six extra parts. A few years back I worked for a small electronics company that designed controllers for small mom and pop operations. Built a small 120V AC pump controller that had a 24V power supply for the pressure transducer, relay and a couple LEDs. It did some pretty complex stuff and was only $42 (less than they paid for the pressure transducer) in quantities of 300. In a "one of" design it is ok to just piece things together, but if you get serious, this needs a lot of rethinking.

 

[mick43]

Thanks OpreaHouse,
I'll take another look and try to find a SSR (instead of a relay with contacts) to switch the 48 volt load. When the relay closes there is very little load because the blower doesn't begin to turn until about 5 seconds after that. I didn't think about the arc being a problem, but based on what you told me, I guess it might be anyway.

In mushroom growing, the co2 produced varies a lot (2x-5x) from day depending on the stage of growth and how many mushrooms are present. So, a time based system wouild have to be reset practically every day. Usually we use an electronic damper actuator that varies the proportion of fresh and recirculated air in the production area based on the signal from a co2 sensor. In the system I'm developing here, the blower exhausts air to the outside and fresh air enters passively from the opposite side of the production area to replace the displaced air.

Again, thanks to you all, brianr, dhwilliams, buzzup and operahouse, for taking time to teach a novice some basics. I'm going to stop checking this thread after three more days.

Mick43



Mick43