Compressor Control Circuit

Another one of my spaghetti boxes for anyone that is interested. Once again the usual warnings and dangers of working with mains voltage apply. This project also uses mains voltage activated relays purely because I already had these and not many activations are needed

The background to this one is as follows.
I have a small compressor that sits outside the shed in a semi-sounprooof enclosure. As most of you will know when you turn a compressor on, it run until the pressure reaches a certain level and then turns itself off. The problem is I forget it's on and because of other noise I don't hear it for the rest of the day and then late in the evening, often while in bed and things are much quieter I sometimes hear the compressor start itself up again. No one has complained but I reckon the neighbours don't need this at night and worse is when I have gone away for a couple of weeks and it has run all this time.

To help me remember to turn the compressor of I even set up a red light near the door that is wired into the Compressor circuit but I still managed to usually walk right past it and not turn off the compressor.

So the aim of the new circuit was to set up a box that would automatically turn the compressor off after a certain hour every day. There are plenty of 10A timers that will automatically stop and start an appliance but there is nothing I'd trust at the 15A level AND I don't want the timer to automatically turn back on again - I want to have that control. Plus while I was at it I thought I would implement an automated Tank vent - Currently I have an irrigation solenoid connected to it and can trigger the venting from inside but I often forget to do it for weeks until the water starts coming out of the air lines.

This is what the box looks like
https://www.woodworkforums.com/attach...d=314459&stc=1
The enclosure is home made from some ally channel, white PVC and clear perspex scavenged from a Skip at work
The orange power wires underneath are from a 15A extension cord - LHS is power in and the RHS is Power out.
The black plug on the RHS of the box is a 24V~ supply to the venting solenoid on the compressor tank.

The no volt switch (NVS) will start and stop the compressor manually provided the timer is on
The 24 Hour clockwork timer is scavenged from an old timer that has not seen use for ~15 years.
https://www.woodworkforums.com/attach...d=314458&stc=1

FWIW here is the inside with the front door open.
https://www.woodworkforums.com/attach...d=314457&stc=1

And here is a sort of circuit schematic.
https://www.woodworkforums.com/attach...d=314460&stc=1

This is how it works.

Incoming AC power is supplied to the timer, a 30A Power Relay (PR) and a Venting relay (VR2).

The timer is in series with the NVS so both have to be on before the PR will allow current to reach the output to run the compressor.
The Timer is set to cut the power off every night at say 9pm which breaks the power to the compressor but triggers VR2 to switch off allowing 240V to reach the transformer (T) which will open the tank vent valve. (VR2 is wired to deliberately switch the power to the transformer when it is switched off)
When the timer turns the power back on VR2 also turns on and now cuts the power to the transformer so the tank vent valve will close. but no current reaches the compressor until the NVS is triggered manually. This prevents the compressor operating if I am away etc.

The PR ($3.50), VR2 ($4) and the NVS ($13) were purchased on ebay.
It's not necessary to use a 16A NVS but I got these so cheaply it would have been about the same price to purchase a lower current rated switch.
I had a 24V transformer from an irrigation controller but it was too big for this enclosure so I bought this one from Altronics ($13)
All up it probably cost $60 worth of components.

I admire your ingenuity Bob.
I have also forgotten to turn off my compressor when leaving the shed a number of times. My solution was to put a switch for the compressor right beside the light switches beside the access door, I turn them all on on the way in, and turn them all off on the way out.

Quote:

---

Originally Posted by Ironwood

I admire your ingenuity Bob.
I have also forgotten to turn off my compressor when leaving the shed a number of times. My solution was to put a switch for the compressor right beside the light switches beside the access door, I turn them all on on the way in, and turn them all off on the way out.

---

That's the solution I was chasing with the red indicator light by the door but it doesn't seem to work. I even thought about just shutting down all the breakers as the breaker box is by the door but I always seem to have something cooking, like batteries charging, or some electrolysis going, or motor testing or, . . . . . Actually it's nothing that can't wait either.

Cheers

Thanks for this BobL

This is a great idea and one which you should probably have sold to an electronics magazine or made into kits for sale through Jaycar.

Beautifully described and drawn. Even an idiot could put this together which is very lucky for me. :D

I will start to gather the parts immediately.

Quote:

---

Originally Posted by Bob Willson

Thanks for this BobL
This is a great idea and one which you should probably have sold to an electronics magazine or made into kits for sale through Jaycar.
Beautifully described and drawn. Even an idiot could put this together which is very lucky for me. :D
I will start to gather the parts immediately.

---

No worries and stay safe.

OK another similar type of circuit as the others so the usual warnings about working with mains AC apply.

I have my compressor in an enclosure outside my shed.
The enclosure is cooled by a fan that draws warm air out of the top of the enclosure. The fan is currently switched on via the compressor pressure switch so that when the compressor is pumping the fan runs but one problem with that is that it would be helpful to continue to cool the compressor for some time after the compressor has finished pumping. The need for additional cooling is because I recently purchased a 20 CFM compressor and it is running a 5HP 3phase motor that draws almost 3 times the current that my current 2.5HP compressor draws.

In the new version of the cooling control the cooling fan can be switched to OFF, ON, or a DELAYED OFF, all from inside the shed.

Here is a semi schematic circuit diagram.
https://www.woodworkforums.com/attach...d=316258&stc=1
There are two relays R1 and R1.
R1 switches on when the pressure switch is closed (i.e. compressor starts running) and allows 240V AC to get to the fan in the compressor enclosure.
The R2 coil also gets its coil power via the pressure switch but this relay is set to be OFF when power is applied to the coil.
When the pressure switch opens (i.e. compressor stops) R2 shuts off and this allows power through to the Omron timer which continues to transfer power to the fan for a preset time.

For testig purposes I mocked up the pressure switch with a simple dummy switch (DPS)
https://www.woodworkforums.com/attach...d=316257&stc=1

Here's what the guts looks like.
R1 & R2 are the two 240V relays I bought for very little from ebay
The Onron timer (T) comes from the large air con controller I found during a verge rubbish collection.
It's the same type of timer as the one I used for the DELAYED POWER OFF circuit I use for my DC - which BTW is working very well.
The solid looking enclosure comes from the same source but the cover was damaged so I have replaced it with some clear PVC which I found in a skip at work.

https://www.woodworkforums.com/attach...d=316256&stc=1

The timer cam be set to run from 1/10th of secs to hours. In the picture the "delay fan off time" is set on 10 seconds for testing purposes but I'll probably set this for at a couple of minutes after the compressor has turned off in winter and maybe 10 minutes in summer. This is done by turning the dial on the front of the timer.

Couldn't you have set that timer to mode "D" (off delay) and just used one relay?

Quote:

---

Originally Posted by RustyArc

Couldn't you have set that timer to mode "D" (off delay) and just used one relay?

---

Good point.
Except my timer is an A8 model and this doesn't have a "D" mode.
The A, AS and AP models can do the "D".

I installed a digital current meter onto the compressor control box (PM2) and immediately noticed that the currents it was reading were significantly less than the inline meter (PM1) I had been using using.

So I got the inline meter and connected it in series before the compressor control box. For good measure I also got out my current clamp meter and DVM and connected it after the control box. And also set the VFD to read out current and the photo shows what they were all reading at the same time. The DVM reading should be multiplied by 100 to get Amps.

DVM 16.8A
PM1 15.4A
PM2 12.2A
VFD 2.8A

The one I trust the most is the DVM with the clamp meter because when I first got my current clamp I did test it against a quality meter but that was 6? years ago.
Anyone have any ideas?

https://www.woodworkforums.com/attach...d=316384&stc=1

Re the metering differences, not sure about the 2.8 A reading from the VFD, but the variation in the DVM and the LCD panel meters is probably due to how they handle non truly sinusoidal waveforms.

Most AC DVM's and panel meters are calibrated to indicate RMS values for true sinusoidal waveforms, but do not include a True RMS capability. I.e. someone researches an algorithm that they can program into a microchip that relates peak or peak to peak values to an equivalent RMS indication, and if they get it right, it will work quite well for a sinusoidal waveform. However, add a little DC offset, or use a different waveform, (squarewave, triangular, clipped sinusoidal), introduce waveform distortion, spikes etc and the results go everywhere.

Some classy meters do incorporate True RMS modules and actually do calculate the root of mean of the square of instantaneous values over a given time interval, and if used within their device specs and ratings will indicate a true RMS value regardless of the waveform type being measured. However the truly true RMS units are way more expensive than the generic units that will indicate 'RMS' for a sinusoidal waveform.

Another consideration is the sampling bandwidth of the meter, whether it is really True RMS, or RMS indicating. The output from the VFD would be a form of additive square waves, with significant harmonic components likely well beyond 2KHz when outputting a 50Hz signal. If you used meters with the same algorithm, but different bandwidths of say 250Hz and 1000Hz, the unit with the lower bandwidth would be expected to read lower, because it cannot detect the higher harmonics that the higher bandwidth unit could see.

Current draw by a switching device like a VFD can be tricky to measure as it's generally non-sinusoidal, so you need a good true RMS current meter to measure it - thus I'd be inclined to believe the DMM.

Thanks for the info guys. Much appreciated.

Just to be clear on what I'm doing. All these meters (except the VFD) are in series sensing mains power before the VFD, not after it. The DVM and PM2 are also sensing on the same conductor.

Quote:

---

Originally Posted by BobL

Just to be clear on what I'm doing. All these meters (except the VFD) are in series sensing mains power before the VFD, not after it. The DVM and PM2 are also sensing on the same conductor.

---

My comments above were based on you sensing on the motor side of the VSD, and I initially thought "that's half an hour of my life wasted". Ten seconds later, I recalled some of the stuff from 40 years ago when I designed high power (2-5KVA) power supplies for a living. Current only flows through the rectifier for periods when the peak voltage is greater than the sum of the storage cap voltage and forward junction voltage of the rectifier diodes, so the current waveform is a series of very high current narrow pulses starting just before peak mains voltage, and finishing soon after peak mains. As such, it is very spiky with extreme harmonic content, and the size of the spikes could possibly overload the input stages of meters intended for mainly sinusoidal waveforms.

Here is a link that illustrates the issue, but in your case, the load is potentially 16ohm at 4.5KVA. Not sure what size the cap bank is in the VSD, but it cannot be way oversized, which is the traditional way to smooth the current peaks

Under the circumstances, it fails both the sinusoidal and harmonics criteria mentioned previously for getting consistent current readings from a non true RMS meter.

Sounds good Malb. :2tsup:

I connected both panel meters up in series to my 3HP Dust extractor (no VFD) and the both read within 0.1A of each other at 9.2A.

I forgot to add that PM1 has the current sensor coil onboard (i.e. embedded in the meter panel) and is in an enclosure with no other electrical gear whereas PM2 is in a box with lots of other stuff and the coil is outboard at the end of connector cable.

I brought home an old CRO from work and and had quick play around with it to see if I could look at some waveforms.

I didn't want to stick mains V into this thing so I used the current clamp.
It is supposed to generate 100mV~ per 10A~ so I could also determine crude current outputs.

Here is the waveform of the input power to my DC.

Assuming this is sinusoidal then using Vp/SQRT(2) that works out to be about 8.1A while the RMS value (square root of the average of the values across the wave form squared) of this wave form is closer to 9.1A
The 2 digital panel meters both gave readings of 8.2 A while the DVM with the current clamp gave 7.8A
So the cheap panel meters seem closer to the RMS value than the DVM with the clamp meter
https://www.woodworkforums.com/attach...d=316477&stc=1


Now here is the waveform of the input power to the VFD and compressor.
It's not as bad as I thought it would be
The RMS current across the waveform works out to be 17.5A
The DVM with the clamp reading was 16.7A, while PM1 and PM2 read 13.5 and 15.2A respectively.
So now the DVM with the current clamp reading is closer to the RMS value than the Panel meters.


https://www.woodworkforums.com/attach...d=316478&stc=1

I will try to measure the 3phase currents with this and see if it makes any sense.