Soft starting an inverter with a series resistor.

[MarkGallimore]

Hi,

I need to apply / remove power to a servo controller/drive combination several times in the space of 5 to 10 minutes before using it for a test which might last between 1 minute and 1 hr. this cycle is repeated frequectly. I'm afraid that the rapid current surges will damage the caps / psu circuitry.

I'm thinking of using a 50W resistor in series with the drive input to limit the current and then short the resistor with a contactor once the caps are charged.

Can anyone confirm that this will help reduce stress on the controller.

During a brown out the relay might open and hence current would pass through the resistor for an extended duration leading to overheat of the resistor and any associated damage. Can a thyristor be used in place of the relay so that even during a brown out the thyristor remains conductive until all power is completely removed?

Do resistor / switch / thyrister combinations exist just for this purpose - we're talking about a 10amp sp 230 controller.

Many thanks for anyone taking the time to reply.

 

[sreid]

My personal opinion based on long experience; cycling power to a VFD at, say, once per minute should have very little to no effect on the life of the B+ Bus capacitors.

 

[ozmosis]

I would agree with sreid.
It is the spacing between power cycles that is important. If several starts in 10 mins are all based on a period of (say) 20 secs, then the inrush would potentially create a problem with the drive.
Fitting a resistor into the AC incomer would, inmho, create more problems than solve.
I guess the key question is, why do you wish to power up/down the drive in such a short space of time? What is the test you are doing that requires this?

 

[MarkGallimore]

In reply to SED2DEVELOPER

Thanks for reply. The machine we are building has a guard that the operator must open / close several times during the period when he initially sets up a test piece onto a madrel. This involves an adjustment (guard open) a quick test run (guard closed) reopen, retest etc unti the set up is correct. The test is then initiated for the required test period. I am using 1 safety relay which is activated by either an estop switch or the safety guard (they are in series)and the relay then operates a contactor (with an instantaneous and a delayed output) which cuts power to the two drives in the system. The instantaneous output commands the drives to do a powered stop and then the delayed output cuts power to the drives. They still retain the 24v logic power. This approach was taken to save the complexity of using two other motor contactors between the drive and the motors and the possible need to use a second estop relay.

 

[Skogsgurra]

Many inverters do have a "Safe Stop" input that is regarded just as safe as cutting power to the inverter.

One make (Siemens) has a 24 V DC circuit that, when opened, removes gate drive to the IGBTs. That is a very safe method to make the motor stand still. Even a catastrophic failure in the IGBTs does not make the motor turn. Using such a feature would make your machine simpler.

Gunnar Englund

http://www.gke.org

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

 

[ozmosis]

The nicest solution would be the one described by Gunnar. The "Safe Stop" system is one now adopted by many companies, especially in the servo drive systems.
In the abscence of this, I would look to utilising the timer delay you already have within your system to ensure the ON time is such that you will always have sufficient delay. This should not compromise any safety aspects as it would only be operational for the ON command not the OFF.
I would contact the supplier of your servo drive to determine the minimum number of times you can cycle the drive per minute or whatever they use, and then use this for your timer.

 

[MarkGallimore]

My drives are made by Baldor (Mint II). They require 1)a hardwired "drive enable" input as well as 2) seeing either a software enabled command or alteranively a hardwired signal on one of the other digital inputs. Unless I missunderstood him, my Baldor rep was very reticent about suggesting the reliance upon using the drive enable as the sole means of disconnecting power from the motor. Indeed, their literature calls the input a drive enable and suggests that the line can be switch contolled but it makes no suggestion that it can be used as a safety stop mechanism. I'd hoped to use it initially and would like to use it now if only I hadn't been put off by the comments I received.

Another aspect that swayed me from the route was that my undestanding (limited) of safety / estop systems was that (and I can't remember where I read this) the use of semiconductor devices in estop systems is OK provided that the final leg of the system (ie the power disconnect) is a mechanical contactor. This viewpoint on the compulsory use of a mechanical contactor makes me wonder about the purpose of the hardwired enable input but I guess there are plent of applications where safety would not be an issue and hece it could then be useful. I'd be interested to hear your opinion on this.

 

[Skogsgurra]

Re "electronic safety stop"

It has been proven that an inverter, that has its gate drive removed in a safe (electromechanical) way cannot spin the motor. Not even if the power electronics fails (one or several shorted transistors). So the requirement that the motor doesn't spin is fulfilled.

But, it does not guarantee that there are no lethal voltages on the motor terminals. Many sales reps do not understand this difference and therefore say that safe stops are unsafe. You will have to dig deeper in the manuals to see which kind of drive enable you have.

An ordinary drive enable signal shall never be used for safety purposes.

Gunnar Englund

http://www.gke.org

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

 

[MarkGallimore]

Hi Skogsgurra,

In your reply, the phrase "ordinary drive enable signal" - does this mean as apposed to a non-electromechanical manner? Is the electromechanical means you talk about a relay inside the drive (that's what I'm assuming you mean) as opposed to the physical wire link which you'd wire to the enable input and open / close with a switch.

To late to ask the controller folk now but will def do so tomorrow. Cheers.

 

[Skogsgurra]

As you indicatated, there are two types of signals being used in these situations: An electrically floating (isolated) mechanical contact and an electric control signal, that is often coming from a PLC.

There are also, of course, commands coming via field busses. But they are not supposed to be safe if the equipmentis not meeting a certain SIL. So, I leave bus signals out.

Standard inverters usually need an isolated electromechanical contact for the "Safe Off". I have never seen any that accepts a PLC signal for that, be it fail-safe or not.

Inside the inverter, there is a direct path from the positive gate supply, via the external electromechanical contact and then back to the gate drive system in the inverter. No electronic components in that path, but it may be possible to use a relay also inside the inverter. The relay being controlled by the external electromechanical contact.

This is safety stuff. So you should make sure that you do the right thing. What I say is only indicative, although based on some experience and good faith.

Gunnar Englund

http://www.gke.org

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

 

[ozmosis]

Actually, I think we do need a separate forum somewhere to talk about specific safety issues as this is cropping up time and time again.
However, MarkGallimore, there is a detailed book published by Siemens (my employer btw)that explains in detail issues concerning Safety in the industrial process environment. It covers product related issues as well as standards, more European related but still, safety is sort of safety wherever you are.
Chapter 9 covers "Fail-Safe Drives" and explains this relatively new direction for drives by integrated safety functions within the drive itself. It is designed specifically for applications you have already described in your site. The hardest part is taking the traditional perception of a lump of electronics being safe when "power" is still apparently applied.
It's a large download but a worthy read. Ok, it's a Siemens document but the information provided is useful for most drives that provide a form of "Safety Integrated". There is a new addition also that includes more up-to-date info on safety related issues.

http://support.automation.siemens.com/WW/view/en/17711888

 

[MarkGallimore]

To skogsgurra, i'm just grateful for your opinion. Of course it's my responsibility at the end of the day but these sort of conversations are useful for highlighting what I need to get confirmed by the equipment manufacturers.

To sed2developper, thanks for the link - clicked on it and went to the web page but couldn't see how to actually download it - Siemens UK can hopefully help me tomorrow.

 

[ScottyUK]

Mark,

If you click the blue word 'Display' to the right of the 'evil e' icon for Internet Explorer a subwindow should open up. Here you have the options to open or download the files.

Good link sed2 - you've posted that one before sometime.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[Drivesrock]

Hi Mark,

To answer the in-series resistor question: In my experience it is a drives internal pre-charge resistor that is not rated for frequent use. The resistor(s) will probably fail if the drive is power cycled very regularly. They do not have much of a heat sink - if any - or are not even fitted in the forced cooling air flow.
If your drive doesnot have the 'safe e-stop' would it be possible to fit a contactor between the drive and the motor? This way you can leave the drive powered on. Also, you will save the energization time where you are first pre-charging the drive and then the microprocessor self-checks etc. This could be valuable lost time in your process.
If you can use an output contactor I would recommend that you use an auxiliary contact to stop/trip the drive so that the main contacts open without load so no sparking consequences will be 'seen' by the output IGBTs which can damage them or their suppression circuits.

 

[MarkGallimore]

Scottyuk - got the document - thanks for help.

Drivesrock - I could fit a contactor between the drive and motor. I could use a safety relay with instantaneous and a delayed output to 1) stop / disable drive with relay's instantaneous output and then drop out the contactor with the delayed output. When the safety relay is reset I assume I need to ensure that the contactor is closed again before re-enabling the drive. If I used one of the contactors n/o auxilliary contacts I could wire that between the safety relay instantaneous output and the drive enable. This would hopefully ensure that the contactor had to be closed before the drive is re-enabled. Anyone done it that way?

Sed2developer - thanks - a very interesting document. It describes the process of using pulse disable and power circuit disable for the safe estop function quite well.

It also mentions that this can remove the need for contactors to physically remove power. Is this just for Europe however? On page 28 there is a section of NFPA79 and it says that "contrary to EN 60204-1, NFPA 79 specifies that for emergency stop functions the electrical power must be disconnected using electromechanical devices"

If I also understand the same section it's suggesting that a controller with a built in formal estop capability (from a reliability point of view) ought to conform to iec 61508. Is this the key point I should be asking my controller supplier?