DC power supply

have recieved from machines from germany that have 2 seperate 24 volt dc power suppplies. They have one that used for the plc inputs,outputs etc. and then there is another that is used on the analog input and high speed counter input cards.
Is this a good pratice. From the schematics the DC common of each power supply are not connected.
The reason i ask is that we have had these machines for about a year and experienced multiple failure of the some of the flow meters that are powered off the one power supply used for the analog and counter cards. Basically what they did was take a 24vac tap on the main transformer and it goes to a circuit board with just 4 diodes and a capactior and that is how they are getting the 24vdc for the analog and counter cards. With AC inverters on these machines and with about 15 of them side by side with a scope you can see a lot of noise and spikes on the 24vdc out this what i consider a cheap power supply.

Hi,

The principle of what they have done is commonly used (over here anyway) and helps minimise noise etc. on your analogue circuits. The method they have used seems a little basic though!

Rob

Wow!
Is the value of your machine worth the investment of a 100$ regulated powersupply? OK the 120/24 give you some noise reduction but...!
We are using small and low cost Siemens logo powersupply here that work with any voltage from 85 to 240. My practice is also to always look for a 24vdc plc to be isolated from the line by the powersupply. This setup saved a lot of plc where in some plant they lose everthing else that was 120v powered. You just need a good reliable PS

having 2 separate power supply is ok as long as no device need to use + from one and the return from the other

The theory behind two separate power supplies is fine, but their implementation certainly is not.
The simplest solution would be to install two decent regulated 24vdc power supplies. Having two supplies isn't really necessary for most applications, but it can't hurt either.

thanks for advice. Actually just found out that the components used in the flow meter are rated to operate from 15-24vdc.
After telling the manufacturer of the equipment we was just going to rip there system out and replace it with our own they started be more cooperative, but still not much.
On a conference call I actually heard one of them in the background say that anything above 24vdc will cause the flow meters to fail. In my opinion, a bad design.
The machine has a Setex plc in it and i have found out after doing some research that it will see a signal as true as long as it is about 18 volts.
The machine has a very cheap power supply setup to control the analog components and high speed counter cards. Due to this the noise on the line will cause the 24vdc to spike and after a period of time the flow meters just fail. I am looking at replacing the diode setup with a pheonix contace Quit power suppply. Thanks

irondesk40 said:

Basically what they did was take a 24vac tap on the main transformer and it goes to a circuit board with just 4 diodes and a capactior and that is how they are getting the 24vdc for the analog and counter cards. With AC inverters on these machines and with about 15 of them side by side with a scope you can see a lot of noise and spikes on the 24vdc out this what i consider a cheap power supply.

Click to expand...

I have seen this on a number of machines from Germany, but mostly older stuff. Quality DC power supplies are inexpensive these days, so I wouldn't expect to see this on newer equipment.

If you can determine the load, I would just abandon the bridge rectifier/cap and use a regular DC power supply as others have suggested.



-Eric

thanks. we have 26 of these machines so going to replace the diode setup in all of them. The machine is 480 volt machine so looking at the 3 phase input Pheonix Quint power supply.
Any advantages or disadvatage using a 3 phase compared to a single phase. In almost all of the machines we have built in house we powered the machine with 220vac 3phase and i used a single phase 220vac power supply.

single phase ps would be less expensive and as better as a 3 phases unless you need huge amount of current

thanks

The Phoenix Quint family supplies are my standard.

Unless the machine has a step-down transformer for 240V or 120V AC, I use the 3-phase 480VAC input models. In your case I would probably use the cheaper 240V single-phase, though.

Your story reminded me of a customer who smoked the first two PanelView Plus terminals we shipped them, ten years ago. They hauled me and the sales manager to their shop so we could hook up a third terminal and apologize for our hardware.

Their shop power supply was a variac tapped for 24V AC and connected to four power diodes soldered into a bridge. All the components were secured to a square of plywood: the diode bridge was literally soldered into four wood screws.

They were insulted when I asked to measure the DC output: this had been their shop power supply for ten years !

It measured 50 volts DC. The autotransformer had slipped or shorted one winding, so it was nominally at 5:2 instead of at 5:1.

irondesk40 said:

have recieved from machines from germany that have 2 seperate 24 volt dc power suppplies. They have one that used for the plc inputs,outputs etc. and then there is another that is used on the analog input and high speed counter input cards.
Is this a good pratice. From the schematics the DC common of each power supply are not connected.
The reason i ask is that we have had these machines for about a year and experienced multiple failure of the some of the flow meters that are powered off the one power supply used for the analog and counter cards. Basically what they did was take a 24vac tap on the main transformer and it goes to a circuit board with just 4 diodes and a capactior and that is how they are getting the 24vdc for the analog and counter cards. With AC inverters on these machines and with about 15 of them side by side with a scope you can see a lot of noise and spikes on the 24vdc out this what i consider a cheap power supply.

Click to expand...

24 RMS VAC will be approximately 33 Peak VDC when using a capacitor only to filter the output under no-load conditions. To not use a 24VDC output regulator on this setup or use a regulated supply is just asking for excess voltage problems.

I use 1 power supply for analog and instrumentation and 1 power supply for I/O.
This helps prevent any transient surges or spikes (caused by valves or switches) passing noise to my analog signals or instrumentation.

But I use quality 24VDC regulated power supplies (Quint from Phoenix Contact), definitely not a home built DC with AC feeding a diode bridge. This is very "old school" and not what I would call "clean power".


..

thanks everyone

the setup with diodes and a cap will be max 24 volt, still having spikes etc.
please change it to two general PSU units.
use one for the outputs and another for the analog inputs (not for the counters.
Any input (analog and digital will work on 18 volts) no problem.

Ken Roach said:

Your story reminded me of a customer who smoked the first two PanelView Plus terminals we shipped them, ten years ago. They hauled me and the sales manager to their shop so we could hook up a third terminal and apologize for our hardware.

Their shop power supply was a variac tapped for 24V AC and connected to four power diodes soldered into a bridge. All the components were secured to a square of plywood: the diode bridge was literally soldered into four wood screws.

They were insulted when I asked to measure the DC output: this had been their shop power supply for ten years !

It measured 50 volts DC. The autotransformer had slipped or shorted one winding, so it was nominally at 5:2 instead of at 5:1.

Click to expand...

Laughing my butt off. Thanks for making my day, Ken.