RS232/RS485 COMMUNICATION SOURCE

[fifty60]

Looks like I am going to have to use the TTL. The RS232 protocol in not ModbusRTU. They have their own software implementation. Terrible. Why wouldn't they use Modbus RTU? I believe their RS232 is supposed to be used with a desktop computer.

For TTL, I need to use the ground from their DB-25 input to get my 0 signal right? I cannot just use the open contacts from my relay?

 

[GoldDigger]

Yes, you must use the ground from the DB-25.
An RS-232 connection to a computer (desktop or laptop) will be interfaced through a serial port to a program or even just a terminal emulator to allow you to type the commands.
RS-232 defines point to point communications with separate receive and transmit data wires, not a bus structure of any kind.

Tapatalk!

 

[fifty60]

It is common practice to install small components like resistors, or even active circuitry, inside a three piece connector like the DB-25 you will need to make up.
From there you can run a shielded or twisted pair wire to your dry contact relay.

Tapatalk!

Do have any pictures of approximately what this should look like? I'm not sure where exactly to connect the resistors. Would they go directly on the 12VDC output pin 1? Or would I run the 12VDC back to my sub panel and divide the voltage there. My relay is going to be located at my sub panel which is located inside my lab equipment, and not inside the actual power supply that I am trying to shut down.

 

[fifty60]

Since I can add an RS232 card to my PLC, I believe I will look into getting an RS232 to TTL converter. This may be the most straight forward way of accomplishing what I am trying to do. The problem I have with this is that there are 2 seperate power supplies that I need to shut down with the TTL signal. Does taking the RS232 signal from my PLC to terminal blocks and then branching out to two seperate RS232 to TTL converters sound like it would work?

I also have RS485 available. The RS485 would be more reliable over the transmission cable. Does the where exactly does the signal get converted? Is it at the DB25 that connects to the remote equipment? Would I have RS485 all the way to the equipment, and then it gets converted to a TLL signal?

 

[fifty60]

https://www.phoenixcontact.com/online/portal/us?uri=pxc-oc-itemdetail:pid=2982728&library=usen&tab=1

Do these optocouplers actually give me a 5V output, or do I have to supply the 5V for the output? Would I need to use this and a DC to DC converter?

 

[gar]

140417-2353 EDT

fifty60:

You are jumping all over the problem area without understanding the signals with which you are working and the problems of connecting together disassociated pieces of equipment that likely have different EGC points.

My judgement is that you should have galvanic isolation between all your pieces of equipment to build a rock solid reliable system. There are many ways to do this. Also you want to minimize noise coupling into your input circuits.

Phoenix basically makes good terminal blocks. I have never used any of their other products. The datasheet is very poor, confusing, and is possibly incomplete. The referenced device is essentially an optical coupler with some protective diodes. This is essentially a device that apparently thresholds at very low voltages relative to 24 V and the diagram shows nothing for current limiting at the input to the optical coupler. Unless there is internal amplification there is unlikely the ability to switch a 50 mA load at the frequencies implied. On the surface there are inconsistencies. This does not supply any output source voltage. It is simply an output switch. I do not think that the circuit as drawn would correspond with the datasheet specifications.

If you want to work with TTL, CMOS, and RS232 circuits, then get some reference material and study the characteristics of each of these types of devices. You need to understand them to properly apply them.

See http://www.renesas.eu/products/opto/technology/architecture/index.jsp for discussion on opto-coupled FET.
A device like this http://documentation.renesas.com/doc/YOUSYS/document/PN10269EJ02V0DS.pdf
could switch moderate current at low voltage drop, but has a much higher input threshold than a TTL gate. I doubt Phoenix is using an FET output because the drawing shows a bipolar transistor.

You do not need to switch a large current at a TTL input.

.

 

[mgookin]

Let's start at the beginning.
What is the voltage and amperage of the alarm circuit? (the event trigger)
What is the voltage and amperage of the circuit we want to shunt?

 

[fifty60]

As much as I would like to build up my own electronic solution, using discrete components and integrated circuits, the scope of what I am doing does not allow it. I am going to need completed electronic components, if possible.

At my equipment I have 24VDC. The remote power supply I am trying to shut down accepts a TTL signal for the purpose of remotely turning off the power supply.

I have two options, and RS232/RS485 option and the TTL option. I do not necessarily like either option, and would prefer to shut the power supplies (I may have forgot to mention that there are 2 remote power supplies) off directly by removing the electrical supply to them. They unfortunately draw more than 30A a piece so I will not be able to do that.

I think the TTL is the most reliable, since an error condition will produce the off signal (0Volts).

If my solution of using a DC to DC converter for my 24VDC to 5VDC, and then using an optocoupler seems viable, I believe that is the solution will pursue. I know that galvanic isolation is a major factor here with the 5VDC signal, but If I can feel confident that this solution will work, I can focus on the grounding after I purchase my materials.

I would use the DC to DC converter to give me the 5V signal that I would feed into terminal blocks. I would connect shielded 18/2 wire to the terminal blocks and feed them to DB-9's at the rear of my equipment. From the DB-9's I will link them directly to the power supplies via a cable and DB-25 at the remote supplies. I will use a solid state relay to switch the 5V signal from 0-5V inside of my equipment.

Does this seem like a correct solution?

 

[gar]

140418-1000 EDT

fifty60:

Some of the characteristics of one TTL gate, a single inverter (NOT function), of a standard power 7404 are:

Plus 5 V on pin 14, Vcc, and 5 V common (ground) on pin 7. Pin 1 of one of the six inverters is floating at about 1.6 V. High enough to be a logic 1. Pin 2 the inverter output is about 0.09 V. Low enough to be a logic 0.

Some of the versions are low power (L series), somewhat low power (LS series), standard power (no letter), and higher power (H series). The input and output characteristics vary with the versions and whether 54 or 74 series. Also there are CMOS versions.

Unless there is shunting circuitry inside the power supply an open input to the TTL input should cause the power supply to turn on based on your statement that 0 input turns the supply off. Thus, no voltage source is required at the input. But this input is then quite susceptible to noise changing the input to a logic 0. Thus, one wants a low impedance source (a TTL output) or a pullup resistor at the TTL input. The smaller the pullup resistor the less sensitive the gate input is to noise. Further shunt capacitance can help reduce noise problems.

If the supply voltage to the 74 series package is +5 V and the pullup is also supplied from the same +5 V, then the pullup can be any resistance including 0. If the pullup is supplied from +12 V, then a voltage clamp (such as a Zener diode) must be used to limit the peak input voltage to the TTL gate.

Also a negative voltage should not be applied to the TTL gate input.

On a sample of one 7404 at room temperature the output is a solid high output (3.7 V no load) up to about 0.6 V input, and a solid low output for an input above 1.6 V. The logical output of the TTL gate is indeterminate for inputs between 0.6 and 1.6 V.

You have not made it clear where the TTL input is on the power supply. That is what connector. Is it the 25 pin connector? Are +12 and common available on the same connector as the TTL input? If this connector is the 25 pin RS232 connector, then why is a TTL input mixed up into an RS232 connector?

.

 

[fifty60]

There is a separate DB-9 for RS232 and then a separate DB-25 that is an analog input connector.

I think I have been looking at this wrong, and several of the guys here on the board have been trying to clarify it for me. The pin in question, PIN 22, is the External Off pin that I need to pull high or low in order to remotely shut the power supply off.

I have PIN 22 and I have a ground on Pin 13. I believe through a contact closure between PIN 22 and PIN 13, I will pull the input low (0V). Then, if I open the contacts then internally the PIN 22 will be pulled high (5V). I don't think I actually need to supply the 5V, just a contact closure.

So open circuit (no contact closure between Pin 22 and Ground) would be 5V, and closed circuit (contact closure between Pin 22 and Ground) would be 0V.

Is this typically how this kind of input operates?

 

[GoldDigger]

Yes, although greater noise immunity is gained by adding a low current pull up resistor to +5V.
If no resistor, I would at least put a ceramic cap to ground to improve rejection of noise pickup in the wires.

Tapatalk!

 

[gar]

140418-1525 EDT

fifty60:

This started out as an RS232 problem. No initial mention of TTL. Has turned into possibly a TTL problem.

In the original days of integrated circuits there was RTL and then TTL. RTL used a resistor as pull-up for the output transistor load. When the output transistor was off, a 1 state, the impedance was moderately high. Along came TTL with the pull-up resistor replaced with a transistor. Now a lower output impedance in the 1 state. A big improvement in noise immunity.

Before the first post you should have made a contact with the manufacturer. Do that now and really find out your options, and exactly what are the input characteristics of whatever on-off input you plan to use.

No you can not assume that the so called TTL input is already pulled up to +5 V.

I would not expect a general standard for interfacing.

.

 

[fifty60]

I am using pin 22 and the ground pin at the remote power supply. I am running a cable from the db-25 at the customer power supply to a DB-9 on my equipment. I am opening and closing the circuit with a very low resistance bifurcated relay.

Can I ground the cable shields from this circuit on the grounds of my equipment, or would I be better off leaving the cable ungrounded. The ground on the remote DB-25 pin is certainly different from the ground I have in my equipment.

 

[fifty60]

I am going to purchase the DB9 to DB25 cable. The communication wires will be grounded to the inside of one of the connectors right? This will then ground to the frame of my equipment when it is plugged in? Does this mean that the db25 (connected to the power supply) will be grounded to the power supply side and the db9 (connected to my equipment) will be functionally grounded to my equipment?

Also, the total run will be about 15 feet between the equipment and the remote power supply. From what I can find online, it does not look like this will be a problem for the 5V ttl signal. It is an on/off 5V signal, no serial. Anyone see a problem running this type of signal 15 feet? Should I use a cable designated as a "transfer cable" or "modem cable" or "serial cable" or does it matter?

 

[GoldDigger]

No standard DB9 to DB25 cable will connect to the pins you need on the DB25 side. Make your own cable instead.

Tapatalk!

 

[fifty60]

I see that now. The 22 pin lines up on the DB9 but the 13 pin does not. The modem (not null modem) cable means DTE to DCE so straight through, and that is what I want. Will need to use two DB25's or make my own as Goldigger suggested.

 

[gar]

140421-1318 EDT

fifty60:

Your discussion is still confusing.

From post #19 and the drawing therein and some other comments I conclude:
1. The customer supplied power supply has at least one DB25 for which a partial schematic is shown in post #19.
2. The sex is not identified.
3. This connector is clearly not a standard RS232 and does not appear to have any RS232 signals.
4. There is a pin 22 not shown that is apparently your point for ON-OFF control of the power supply.
5. Nothing is shown to indicate the type of input that is connected to pin 22.
6. Pin 14 is identified as API GND. What does that mean? Is it connected to the power supply chassis and in turn to the power supply EGC?

In my opinion you should not connect the chassis of your equipment to the chassis of the customer power supply with some small wimpy wire that you might find in a communication cable. This will simply parallel the power supply EGC. If I short the power supply chassis to a hot 120 V wire your wimpy common connection will burn up.

What circuitry, if any, that you place at the customer power supply associated with pins 22, 14, and possible pin 1 for your relay output to control will depend upon what is the input inside the power supply at pin 22.

.

 

[n1ist]

Looking at the spec sheet, the remote on/off is listed as ttl active high or low, programmable. You may be able to set it active low and just tie it to ground on the Chroma to control it (if there's an on-board pullup). You definitely will need to make a custom cable. I would definitely not tie a common ground between your equipment and the supply; just provide a dry contact if possible.

Why don't you give Chroma a call? They may be willing to tell you. I can't believe that they charge for web access to their pdf manuals online...

 

[gar]

140429-1310 EDT

fifty60:

What have you decided to do and how well does it work?

.