Proportional Electric Heat - Leaking SCRs?

[fixitman]

I'm anticipating using some Viconics 'stats to provide proportional control of some electric duct heaters. The heaters in question use pressure switches to enable them when air flow is present. As currently configured, when the fan starts, the safety contactors pull in and stay in during the entire time the fan runs. So they close once a day.

A controls consultant is telling me that SCRs (brand new, also Viconics, in this case) can sometimes leak and waste energy. He advocates using a DO in the 'stat to enable the heater in addition to the fan switch. This means that the contactors might close many times a day, rather than just once. Unfortunately, Viconics tech support says the DO in question would close after the proportional signal is applied, meaning that the contactor will sometimes close when the SCR is turned on. This will shorten contactor life, in addition to the noise-nuisance of the contactors closing in occupied spaces.

I've seen a few leaking SCRs in my time - but wonder just how common it is these days, given the present state of the technology. The devices I've found leaking were built 20 or more years ago. It seems like today's devices would be expected to perform a good deal better.

Comments? TIA ...

[DaveCR]

From what I've read all solid state relays have a bit of leakage. The Viconics, at 5 mA, are on the low side. Depending upon the supply voltage that's 1-2 watts being wasted when heat is not required. I'm guessing the safety contactors draw 5-10 watts when on. It starts to add up. Does the Viconics thermostat have an "O" terminal (energized when in heat mode) that could be wired to the safety contactors?

[s2sam]

Good day,
I think the term leakage may be inaccurate... All devices (relays, semiconductors, etc) when used as a switch have some form of resistance... which will dissipate energy in the form of heat when a current is passed through it (i.e. Power = Current^2 * Resistance). Are there other and newer technologies that can be used? You bet, but some of these have other caveats (i.e. sensitivity to overvoltage, static discharge, etc) that must be considered and/or protected from (i.e. with other circuitry).
Cheers,

Sam

[fixitman]

Good day,
I think the term leakage may be inaccurate...
Sam

Sam, Thanks for your reply.

If the proportional control voltage to an SCR is at 0V, and that SCR is delivering, say, 200 Watts to the heating element, I'd say it's leaking.

I've seen supposedly "turned off" SCR outputs on DDC controllers allow enough juice to flow that they turned on the input of a circuit board they were connected to. (Yes, the board was designed badly - was willing to turn on its function at 3 V when it "should" have taken more like 20 V at a minimum.) This is one reason why many controllers that have solid-state outputs need isolation relays. This sort of spurious electrical flow through SCRs or triacs or ERs seems like leakage, in my book.

[fixitman]

Dave CR, thanks for your reply.

I'm talking about something that goes beyond the "rated" leakage current that the shiny, new SCR allows. I wouldn't fuss about the 5 mA if it stayed there ... and it well might. I've dealt with some admittedly old equipment that might be dumping say, 300 Watts into a 2 or 3 KW heater that's supposed to be turned off stone cold. It's not enough to make anyone uncomfortable, or even notice. But take a handful doing this in the same small clinic building, and also making the air conditioning pick up the extra load ... it turns into some real $$

[s2sam]

Sam, Thanks for your reply.

If the proportional control voltage to an SCR is at 0V, and that SCR is delivering, say, 200 Watts to the heating element, I'd say it's leaking.

I've seen supposedly "turned off" SCR outputs on DDC controllers allow enough juice to flow that they turned on the input of a circuit board they were connected to. (Yes, the board was designed badly - was willing to turn on its function at 3 V when it "should" have taken more like 20 V at a minimum.) This is one reason why many controllers that have solid-state outputs need isolation relays. This sort of spurious electrical flow through SCRs or triacs or ERs seems like leakage, in my book.

Good day FixItMan,

I understand your point and I guess I am /was looking at it differently. Given that to be the case, you are referring to "Off" state leakage current... which I always assume there is, as all semiconductor switches will allow some current to flow...thus any current flowing whether it be in the "On" or "Off" state will generate heat via the device's resistance... which is how I was looking at it.

Anyway, there are devices that are much better in both of these parameters (devices such as FETs, IGBTs, etc), but these have issues too and can have different drive needs (i.e. how you turn them off/on). So, I guess to answer your original question... yes, there are superior devices out there, but they most likely are not drop in replacements to a standard SCR. Can some of the newer perhaps better components be used? Yes, but my feeling is that additional circuitry would be needed to achieve what you want and in fact there may be control "modules" that already do this. Remember that in most DDC controllers cost, physical size, and generic needs tend to influence their I/O design... does a manufacturer support a BO that can drive 1A, 5A, 20A, or 100A?... and at what voltages? the larger the drive the more heat is dissipated and so the controller needs to be physically larger, maybe needing a heat sink... and with the higher currents/voltages different compliance standards need to be met... it can go on and on. In most controller designs I have seen the manufacturer chooses the least expensive approach that satisfies the most needs. In your case, as I mentioned earlier, I would assume that there is a controllable module that you would connect to a controller's BO/DO and achieve what you are looking for. Speaking of which what are the tech specs on your needs anyway (voltage, current, etc)?

Cheers,

Sam

[fixitman]

Good day FixItMan,

I understand your point and I guess I am /was looking at it differently. Given that to be the case, you are referring to "Off" state leakage current... which I always assume there is, as all semiconductor switches will allow some current to flow...thus any current flowing whether it be in the "On" or "Off" state will generate heat via the device's resistance... which is how I was looking at it.

Anyway, there are devices that are much better in both of these parameters (devices such as FETs, IGBTs, etc), but these have issues too and can have different drive needs (i.e. how you turn them off/on). So, I guess to answer your original question... yes, there are superior devices out there, but they most likely are not drop in replacements to a standard SCR. Can some of the newer perhaps better components be used? Yes, but my feeling is that additional circuitry would be needed to achieve what you want and in fact there may be control "modules" that already do this. Remember that in most DDC controllers cost, physical size, and generic needs tend to influence their I/O design... does a manufacturer support a BO that can drive 1A, 5A, 20A, or 100A?... and at what voltages? the larger the drive the more heat is dissipated and so the controller needs to be physically larger, maybe needing a heat sink... and with the higher currents/voltages different compliance standards need to be met... it can go on and on. In most controller designs I have seen the manufacturer chooses the least expensive approach that satisfies the most needs. In your case, as I mentioned earlier, I would assume that there is a controllable module that you would connect to a controller's BO/DO and achieve what you are looking for. Speaking of which what are the tech specs on your needs anyway (voltage, current, etc)?

Cheers,

Sam

Sam,
Thanks for your reply.

The thrust of my original question was about whether significant leakage was likely enough, with modern, proprtionally controlled (pulsed) SCRs, that I should consider disabling the heaters with the safety contactors when they are not actively heating. In other words, disconnect the power from them so that, if an SCR is leaking significantly, it won't have any power when it's not supposed to be running anyway. I already have the 'stats and the SCRs. So I'm looking for peoples' experience with significant leakage in recently-manufactured SCRs, to decide whether or not to leave them powered up all the time.

I'm inclined to leave them powered up all the time the fan is running, so reducing the degradation of the safety contactors, and reducing the nuisance noise caused by lots of contactor cycling.

A consultant is telling me that, if I do this, I will "probably" have significant leakage that I won't be likely to catch. I would have guessed that this is quite a rare occurrence these days, but have no personal experience or knowledge to back that up. I only note that, in my experience, the heater manufacturers don't normally use their safety contacts to intermittently kill power to their SCRs. They leave them connected to the power supply for the entire time the fan runs.

The SCR Manufacturer tells me that such leakage problems are extremely unusual. The consultant tells me that they are actually fairly common. I have noted problems like this in SCRs that were manufactured 30 years ago. Does anyone have field experience to refute the consultant's claim, or back it up?

Thanks in advance ...

[s2sam]

Sam,
Thanks for your reply.

The thrust of my original question was about whether significant leakage was likely enough, with modern, proprtionally controlled (pulsed) SCRs, that I should consider disabling the heaters with the safety contactors when they are not actively heating. In other words, disconnect the power from them so that, if an SCR is leaking significantly, it won't have any power when it's not supposed to be running anyway. I already have the 'stats and the SCRs. So I'm looking for peoples' experience with significant leakage in recently-manufactured SCRs, to decide whether or not to leave them powered up all the time.

I'm inclined to leave them powered up all the time the fan is running, so reducing the degradation of the safety contactors, and reducing the nuisance noise caused by lots of contactor cycling.

A consultant is telling me that, if I do this, I will "probably" have significant leakage that I won't be likely to catch. I would have guessed that this is quite a rare occurrence these days, but have no personal experience or knowledge to back that up. I only note that, in my experience, the heater manufacturers don't normally use their safety contacts to intermittently kill power to their SCRs. They leave them connected to the power supply for the entire time the fan runs.

The SCR Manufacturer tells me that such leakage problems are extremely unusual. The consultant tells me that they are actually fairly common. I have noted problems like this in SCRs that were manufactured 30 years ago. Does anyone have field experience to refute the consultant's claim, or back it up?

Thanks in advance ...

Good day FixItMan,

I understand. That being said, my comments from the sidelines are as follows:

1. I would be inclined to believe the actual SCR manufacturer than the consultant. No one is going to know their own parts better than the people who make them. If you are still in doubt, perhaps select a sample batch and test.

2. Remember that not all SCR manufacturers have the same specs despite being a 2nd source or an "identical" replacement. This may not be an issue for the initial install because you have all the parts now, but in the future if replacements are needed.

3. Contactors (and Relays) usually have a spec for the maximum number of on/off cycles... Depending upon the on/off frequency you could reach those maximums and thus reducing their lifetimes...

In regards to question about the SCRs of today and how their leakage current has been minimized, I do not think you will find a generalized statement that will apply to all SCRs made today simply because of the variety of SCR's available... There are some which are manufactured as they were from their inception and others are newer because they are targeted towards specific applications. For the most accurate info I would suggest you send an e-mail to a few of the SCR manufacturers along with your typical part numbers along with your typical electrical specs (switching voltage, current, control circuit, etc). No one is going to give you better data than the people who actually manufacture the parts.

Cheers,

Sam

[mechdorn]

Good day FixItMan,

I understand. That being said, my comments from the sidelines are as follows:

1. I would be inclined to believe the actual SCR manufacturer than the consultant. No one is going to know their own parts better than the people who make them. If you are still in doubt, perhaps select a sample batch and test.

2. Remember that not all SCR manufacturers have the same specs despite being a 2nd source or an "identical" replacement. This may not be an issue for the initial install because you have all the parts now, but in the future if replacements are needed.

3. Contactors (and Relays) usually have a spec for the maximum number of on/off cycles... Depending upon the on/off frequency you could reach those maximums and thus reducing their lifetimes...

In regards to question about the SCRs of today and how their leakage current has been minimized, I do not think you will find a generalized statement that will apply to all SCRs made today simply because of the variety of SCR's available... There are some which are manufactured as they were from their inception and others are newer because they are targeted towards specific applications. For the most accurate info I would suggest you send an e-mail to a few of the SCR manufacturers along with your typical part numbers along with your typical electrical specs (switching voltage, current, control circuit, etc). No one is going to give you better data than the people who actually manufacture the parts.

Cheers,

Sam

It's easy, the on board safety is exactly what it is, a fail safe to not energize the electric heating without air flow present. The controls should be setup to only enable the electric reheats when the air flow is applied. Otherwise, if relying on the airflow safety, you are controlling off a safety.....wrong. I see this stuff frequently. I'm not saying you're gonna set the place on fire, usually the reheat just burns out. It could be as simple as a master time clock or a master air proving switch, or advanced controllers to stop the PID function without airflow proof.

[fixitman]

It's easy, the on board safety is exactly what it is, a fail safe to not energize the electric heating without air flow present. The controls should be setup to only enable the electric reheats when the air flow is applied. Otherwise, if relying on the airflow safety, you are controlling off a safety.....wrong. I see this stuff frequently. I'm not saying you're gonna set the place on fire, usually the reheat just burns out. It could be as simple as a master time clock or a master air proving switch, or advanced controllers to stop the PID function without airflow proof.

Mechdorn, thanks for your reply.
I wasn't suggesting using any limits as controls. There are three limits - the air pressure switch, a "klixon" disc-type overheat switch, and a fatal overheat limit of the sort that self-destructs if neither of the other two manage to stop the heater. The third of these limits is directly in series with the heater elements themselves. The other two limits are at 24 Volts, in series, and operate a contactor. Having a non-limit control device in a series circuit with these two limit devices isn't going to reduce the safety of the heater at all. If either of those limits opens, the contactor still opens and shuts down the heater - just as they would if the control device was not there.

[joey791]

The last 20 or so heaters I've done have Viconics SCR controllers and I send them a proportional signal only, I haven't really seen any issues with leakage.

[fixitman]

The last 20 or so heaters I've done have Viconics SCR controllers and I send them a proportional signal only, I haven't really seen any issues with leakage.

Joey,
Great!
Thanks for your reply!
I'm planning to let then stay connected when the fan is running. In theory, we will be checking these periodically, and will have the opportunity to catch a larger leakage problem if it develops. I'm painfully aware that duct heaters usually get looked at only when they stop working. Fried wiring because connections never get re-tightened, and who knows what-all else. Hopefully, the experience will be different this time. And we'll have the opportunity to catch a seriously-leaking SCR if one should develop.

[joey791]

Joey,
Great!
Thanks for your reply!
I'm planning to let then stay connected when the fan is running. In theory, we will be checking these periodically, and will have the opportunity to catch a larger leakage problem if it develops. I'm painfully aware that duct heaters usually get looked at only when they stop working. Fried wiring because connections never get re-tightened, and who knows what-all else. Hopefully, the experience will be different this time. And we'll have the opportunity to catch a seriously-leaking SCR if one should develop.

Its funny you bring up the fried wiring, I had some issues several years ago with some CHWAHUs that had Indeeco preheats and reheats. The Allen Bradley disconnect switches were rated for 80 amps, only problem with that was the heater fla was 82 Amps, needless to say I had a few blow up from loose connections.

[bob_scheel]

I'm anticipating using some Viconics 'stats to provide proportional control of some electric duct heaters. The heaters in question use pressure switches to enable them when air flow is present. As currently configured, when the fan starts, the safety contactors pull in and stay in during the entire time the fan runs. So they close once a day.

A controls consultant is telling me that SCRs (brand new, also Viconics, in this case) can sometimes leak and waste energy. He advocates using a DO in the 'stat to enable the heater in addition to the fan switch. This means that the contactors might close many times a day, rather than just once. Unfortunately, Viconics tech support says the DO in question would close after the proportional signal is applied, meaning that the contactor will sometimes close when the SCR is turned on. This will shorten contactor life, in addition to the noise-nuisance of the contactors closing in occupied spaces.

I've seen a few leaking SCRs in my time - but wonder just how common it is these days, given the present state of the technology. The devices I've found leaking were built 20 or more years ago. It seems like today's devices would be expected to perform a good deal better.

Comments? TIA ...

Leakage current during the heating season is not a serious issue because the waste heat does go towards heating the building and is unlikely to drive temps beyond their control setpoints. During the cooling season open the safety contactor to eliminate the loss of energy.