NPN Open Collector 1

[Ruby12]

Hi, I have a Turbine oil flow meter installed and I need some assistance in relation to the pulse output signal. I am told that the meter provides a NPN Open Collector output pulse signal. There are three terminals on the meter (24V+,Pulse & Gnd). I have connected a 24VDC supply to 24V+ and Gnd. I would expect that the pulse signal would then be taken from Pulse and Gnd. However I am unable to determine if the meter is in fact working. Is there any way of checking (eg. using a multimeter) if the meter is in fact providing a pulse output. The logger I am using can support tansistor type/SO inputs but I am not registering any pulse even though there is flow through the meter. The manufacturer of this unit is no longer in business. Any guidance would be very much appreciated.

 

[fangas]

Try finding the signal between +24 & pulse. The NPN output is switching the Neg rail On-Off.

Ed

 

[fangas]

If your data logger uses Gnd as a common, you will need a pullup resistor connected between +24 and pulse. A 1k, 1/4 Watt should work OK.

Ed

 

[VEBill]

The recommended resistor isn't ideal. It will be fine if the pulses are short duty cycle. But if they're even as high as 50%, then there would be more than one-quarter watt. And if the output happens to stay on for any significant period, then there would be more than a half watt on the resistor.

1k, 0.25w resistors are common for 5v circuits, but a 24v signal probably calls for a higher value of resistance (maybe 4.7k or 5k), but more complete info is required to be certain.

Another option would be to use a lower pull up voltage. Typically the receiver circuit would provide this anyway leaving only the open collected at the source end.

If it's just for a short test, it's less important.

But I'd still use 5k vice 1k to eliminate that small risk.

 

[Ruby12]

Thanks fangas and VE1BLL. I have tried using a pull up resistor, but I am still unable to register anything at the data logger. The logger can accept reed switch or solid state pulse type inputs. Would there be any way of measuring if the pulse is actually being generated (eg. at the terminals of the oil meter)?

 

[BobM3]

Seems strange that it would say "24V+" on one connection terminal and also say it is open collector output. My guess is that the pullup resistor is already installed and sized for a 24 volt input. If so, then you would need no external resistor and could take the signal from between ground and pulse. But that's what you originally tried, right? Can you pull apart the meter to see this? How about a model number or schematic? Do you have access to a scope? A volt meter? You could use a 5k resistor as a dummy load and measure to see if the there is a changing voltage (if the oil flow meter is going slow) or a voltage other than 24 volts or near 0 volts (if the oil flow meter is turning fast).

 

[BobM3]

Or, does it say "24V+ Max" maybe? If so, you would need to supply a current limiting resistor between your 24 volt supply and the terminal labeled "24V+". If that's the case and you've already connected a 24 volt supply to it without a current limiting resistor then the NPN transistor may be fried.

 

[Ruby12]

BobM3, Thanks for your suggestions. I have already tried your initial method. I have a volt meter and will check the voltage readings.
The meter does not say "24V+Max"

 

[BobM3]

Yea, it dawned on me that the "24V+" is probably required from your external PS to power up the transducer or the base side of the transistor. There's no other power input to the meter, correct? Hopefully you didn't put +24 volts on the "Pulse" terminal.

You may learn something by checking ohm readings between different pairs of terminals with no power applied.

 

[fangas]

Good call on that 1K resistor. I agree, 5K probably much better. I must have been in TTL land too long.

Ed

 

[markcraig]

If all you are after is to check if it is working may I suggest that you leave it all connected and simply use a good quality digital multimeter on Hz between either the 24+ and pulse output or the Pulse output and ground. Depending on what exactly you are after you can get some moderately accurate readings this way. Not precise I know and not an exactly scientific method but it has work for me in the past.

 

[OperaHouse]

Assuming you have a cheap digital meter.... Wire the open collector to the + supply as suggested. Take a .47uF from the open collector to one input of the meter. The other lead to the - supply. Set the meter to AC Volts and look for any voltage when the wheel spins. Capacitor is needed because most cheap meters will read any DC present as AC.

 

[Ruby12]

@ BobM3
There is no other power supply to the oil meter. I put in a 10k resistor across Pulse and Gnd, but I am still reading zero Volts. I also checked ohm readings between different pairs of terminals with no power supplied. Still no wiser!

@ markcraig
Checked Hz across all terminals..... no deviation from zero

 

[Skogsgurra]

Wouldn't it be a lot more constructive to either tell us the make and type - or google it yourself? This guesswork leads nowhere.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[VEBill]

My guess: It's broken. (This assumes that you're providing suitable simulation of oil flow through the sensor while monitoring the output.)

 

[OperaHouse]

"I put in a 10k resistor across Pulse and Gnd," It may be bad, but the resistor should go from Pulse to +24 to get any result.