View Full Version : flow switch connection

04-12-2005, 11:06 AM
hi to all
this is my question:
where shoud the wires of flow switch be connected(central a/c)?
1-to the compressors,so if the flow in pipes not moving the flow switch will stop the compressor
2-to the chilled water pump,so if flow is not running the flow switch will stop the chilled water pump
thanks and regards:) [/SIZE][/RIGHT][/FONT]

04-12-2005, 12:23 PM
the chilled water pump is interlocked with the compressor cercuit ,, so if the flow switch shutt-off the chilled water pump ,,the compressor will stop also
there is a reason for connecting the flow switch with chilled water pump ,,,is that:when the impeller is lost from shaft ,the water will not move ,then the flow switch stop the chilled water pump

04-12-2005, 01:09 PM
yes a flow switch should be linked with the compressor control circuit as a pump failure would lead to the evap. freezing up and possibly bursting. i have repaired many chillers where the evap has burst and then filled the system with water, a problem we can all do without.

US Iceman
04-12-2005, 09:23 PM

I like to use the flow switch as a secondary method. I prefer to have a set of normally open auxiliary contacts on the pump starter. When the pump is energized the interlock is closed.

The interlock on the pump starter is wired into the compressor safety circuit. If the pump(s) are off the auxiliary contact prevents the compressor(s) from starting.

The flow switch is also wired into the compressor safety circuit. Pump starter interlock first, then the flow switch.

One thing I would caution you about is the flow switch itself. The normal switch enclosure is usually a simple metal cover. After some years of operation with chilled water the flow switch mechanism will rust in the closed position. When this happens, the flow switch is useless.

I like to use a watertight switch cover, and seal the conduit with silicone caulking. The sealant prevents moist air from traveling through the conduit and condensing in the switch mechanism.

I have never seen a pump loose the impeller, but I suppose there is a first time for everything.:confused:

Another safety control for any water/glycol chiller is a freeze-stat to shut down the compressor(s) if the fluid reaches it's freezing point (or close to it).

chillin out
04-12-2005, 09:34 PM
I had a flow switch that was in an iron pipe that was so rusted, that it stuck the lever and the evap burst which got flooded with water.:mad:


US Iceman
04-12-2005, 09:46 PM
I had a flow switch that was in an iron pipe that was so rusted, that it stuck the lever and the evap burst which got flooded with water. :mad:

I agree with you. I have seen this before too. Sometimes the hole in the pipe (where the paddle is installed) is not smooth or completely round also.

I have seen more problems with flow switches than almost any other control switch on chillers.

US Iceman
05-12-2005, 12:15 AM
I completely agree with the differential pressure switch usage. Good point to add to the discussion.

If you have the manufacturers flow curves for the chiller barrel you can use the resulting pressure difference at "flow" to dial in the differential pressure switch setting.

This is more precise than trimming the paddles on the old flow switches.

05-12-2005, 12:39 AM
I like to see both a pump starter inerlock as well as a flow switch, either paddle type or differential switch across the pump or the chiler. You should insure you have positive flow across the chiler before the compressor is allowed to start. In addition you need a good positive freeze protection on the chiller. A low temperature switch in the line provides no protection to the chiller if you do not have a flow for it to sense. So I like to have a pressure freeze stat in with the low pressure control.

Use the chiller flow and pump starter interlock as a permisive and the chiller will not run unless you have both made.

I have seen way too many chillers frozen due to low flow or low refrigerant pressure. I personally like to protect on three fronts for the chiller low flow and pressure/temperture. Many chiller manufacturers like to go low $$$ and skimp on the controls.

05-12-2005, 01:18 AM
binder points either side of the BPHE would give a more accurate indication of chilled water flow rate than would a conventional orifice plate.

Alco bi-flow expansion valve's in reverse

I have never heard the term "binder points" before. Also, I am ignorant of the reverse flow expansion valves. Care to elucidate, Marc?

05-12-2005, 09:03 AM
i do agree that the flow switch must be connected to the chilled water pump ,
but : how the chilled water pump be started at biginning if there is no flow of chilled water?

US Iceman
05-12-2005, 01:17 PM
You can push pressure and temperature probes through them to the inner side.

This sounds like an item we use here in the US. The one I am familiar with is called a "Pete's Plug".

Here is a link to the manufacturer you can review:


how the chilled water pump be started at beginning if there is no flow of chilled water?

The systems I used to work on had the pumps started manually in the spring when the weather started to warm up. Since the pumps were started manually, the flow was created in this manner. Then the flow switch is "active" for the compressor safety.

The response from TXiceman is one I find to be the safest measure of control use. Multiple safety switches like this provide several layers of protection for the most commonly seen faults; Low temperature, low pressure, and low flow.

05-12-2005, 05:58 PM
Hi afeef!
I have not quite caught relation flow switch and central a/c.
So I will apply a word "if"(if you dont mind):)

If you speak about chiller system, flow switch is put in with the main goal not allow run chiller at lack of a flow through the evaporator.
You must to connect normally open contacts of the flow switch to terminals specially intended for this purpose and available in any control system of the chiller.
Then you should on the pump and be convinced that contacts become closed. Otherwise the control system of chillers will give out a signal of alarm. As a rule, a little adjustment of the flow switch would be required that it worked correctly. More reliable stuff, of course, is relay of the differential pressure connected to the evaporator. These two devices can work simultaneously.

If you speak about control of the pump system, then you can use relay of the differential pressure
connected to an input and an output of the pump, and, of course, normally open auxiliary contacts on the pump contactor.Pay attention that contactor can be on, while a power will be switched off It will give a false signal to a control system.
Both signals are used by a control system (for example PLC or BMS)for the proven good status of the pump.
At start of system both signals are imitated by relay with delay function which in 15-20 seconds is disconnected.

Hope this help.

06-12-2005, 07:51 PM
Indeed, paddle flow switches, when installed and commissioned to manufacturer spec provide the best protection against system damage by low flow. They sense velocity pressure which for a given fluid density and pipe diameter is a direct first hand indication of fluid flow. However, they are also probably "the" safety requiring most attention during maintenance and should even be "the" safety given a routine replacement schedule, perhaps even biannually.

At the risk of diverting, quite far, off from the central topic almost to the point of being irrelevant concerning the question "where shoud the wires of flow switch be connected(central a/c)?" I want to say that an evaporator fluid side pressure differential switch that cuts-out when either an indicative increase or decrease in evaporator pressure drop occurs is fast becoming a popular alternative to the paddle switch.

On the contrary this is highly relevant to the subject. if an evap. were to foul or freeze up then the fluid pressure drop would increase and if the pump stopped it would drop to zero. this problem would certainly endanger any plant with a faulty or badly wired flow switch. experience tells me that changing flow switches twice a year may be a little excessive, especially the johnson penn type as i have seen them in the field 10/15 years old and still going strong.

US Iceman
06-12-2005, 10:32 PM
experience tells me that changing flow switches twice a year may be a little excessive, especially the johnson penn type as i have seen them in the field 10/15 years old and still going strong.

I have seen them last this long too.

I have also seen some that rusted shut in the closed position after 2 years or less.

The take home message is not the time frame, but that they should be periodically checked for proper operation and condition.

It is the ones that fail in a short time that make for happy clients.:(

07-12-2005, 04:58 AM
Why couldn't they use a spinning magnetic wheel and a sensor coil so if the flow stops or the wheel jams for any reason, it fails open rather than closed?

07-12-2005, 08:55 AM
I saw something strange the other day when I was on site, when I took the cover of this flow switch it was full of water, have any of you guys seen them burst internally or something like that

US Iceman
07-12-2005, 01:37 PM
Hi tonto,

The flow switches I have seen like the one you describe could be a leaking body. After the cover is removed, does water still drain or drip from the switch at a consistent rate? Not condensation from being cold...

Several of the flow switches where I have seen this I attributed the water to condensation build-up in the switch cover. After the cover is removed, no major dripping from the valve paddle seal.

Even if the switch cover is water-tight, water can condense in the switch because moist air can travel through the conduit.

08-12-2005, 03:08 PM
Preferred a pressure differential switch than a water flow switch, the tongue of the flow switch doesn't last last.

23-12-2009, 06:55 PM
We use three forms of protection for our chillers.
(1) Temperature
(2) Differential pressure
(3) flow
However we no longer use paddle type flow switches, for some time we have been using a flow sensor which works on the principal of heat dissapation, brand name Turck. Check them out if you are interseted in getting away from conventional flow switches which we also had many failure issues with. We have been using this other type of flow switch for about 8-10 years without issue or incident.