Pressuretrol 101 (2)

PMJ

So as not to hijack the original thread I start this one with the following comments to what was said there:

While I do agree fundamentally with the comments there, I will point out again that these systems can have no pressure at all until radiators begin to fill. The original designers intended that this never happen. With a continuous supply of steam from a coal fire I think we all understand that having all our radiation continuously full or even close would roast us. So the original design was to be continuously partially full, which by default mean pressures too low to realistically be used to control intermittent fire.

The problem with having the thermostat be the control is then the inherent time delay from when steam fills radiators to when the thermostat reacts. Anticipators do help with this (I did have a T87), but it has been quite a surprise to me that in all this time since the change to intermittent fire more effort has not gone into a control which uses data about actual fill amount of steam at a radiator. At the end of the day that is what we would all like to be more constant and more closely matched to the conditions. The time required to get steam to radiators from the start of fire varies of the order of 5 times depending on how cold the start. So running each and every burn until a known fill level is reached, and starting the next burn by determining when enough of the steam from the previous burn has condensed using the same temperature sensing device I have found to be a game changer. The time between burns and their length then adjust according to the conditions, with more frequent burns required to maintain the temperature at the radiator as it gets colder. The volume of steam condensing through the same filled volume of radiator per unit time is what is changing...automatically, and resulting in dramatically more consistent conditions at the radiators while satisfying the demand.

The truth is that if you really want even heat, neither a pressure device nor a thermostat can be the thing controlling the burns directly. A satisfied thermostat marks the end of a call for heat and max room temperature reached(or about to be with an anticipator). The one thing I know for sure about a control running one burn per call is that it is not very even or at least not what it could be. What is needed is longer calls with multiple burns. Even heat is actually defined by the length of a call and the longer the call the more even the heat. Obviously perfectly even heat would be an endless call because temperature at the thermostat would never change at all. Long calls require radiators whose temperatures oscillate a whole lot less than most are today. To achieve that you really need some direct input from what is actually happening in them. That equipment ( a temperature switch) costs very little.

delcrossv

That's been a question of mine too. Seems a lot of systems need a modulated flame to lengthen the call for heat. For gas, a Maxtrol could modulate the flame, but controlled by what? You'd need a proportional signal, yes?

Jamie Hall

One of the nice things about modern hot water heat using mod/con boilers with outdoor reset is that it is possible to come very close to @PMJ 's ideal: the system never turns off, and the boiler always runs at exactly the power level required to supply the amount of heat which the structure requires.

The primary control parameter is outdoor air temperature, which is a decent enough surrogate for structure heat loss in a modern tight house. Works fine.

Notably, the temperature of the emitters -- radiant floors, baseboards, radiators, what have you -- is varied inversely with the outdoor temperature in a smooth continuous fashion.

Now. Can we do this with steam -- or, perhaps more generally, with a phase change system. Yes, we can, provided only that we can find a phase change power source which modulates and can control the operating pressure of the system itself.

Phase change power source is a fancy term for boiler, in this case... and, sadly, I'm not aware of any residential or small scale commercial boiler fueled on either oil or gas which is capable of modulation, except pulse width modulation -- which is exactly what our steam boilers do do, whether that is controlled by a timer, a temperature switch out in the system, a vapourstat, or whatever.

The more serious problem, really is the pressure problem. Virtually all phase change transfer media -- which of necessity need to be fluid -- undergo their phase change at a temperature which is controlled by pressure (in the case of water, for instance, that change occurs at 212 F at standard atmospheric pressure). Now if we want to vary the temperature of our emitters so that they produce the exact amount of heat output needed at a given time, we need to control that pressure. We need to raise it to get more output, or lower it, to get less. If we do that, we can run the system continuously.

This is not quite as simple as it sounds, as anyone who has dealt with heat pumps or any refrigeration system can attest. The main condition is simple enough, however: the system must be sealed, and the only fluid in the system can be the heat transfer fluid. Even small amounts of non-condensable fluid (in our case, a fancy term for air) will harm operation; significant amounts will make it inoperative unless they can be removed easily (which is why there are air vents on a steam system...).

As I have said elsewhere, perfection is often the enemy of the good. In heating systems, the good is the simplest possible system which will deliver the desired output: namely, reasonably tight control of space temperature. I might add that any controlling device (such as a thermostat or pressure switch) should be measuring and responding to the parameter being controlled.

I'll stop here...

delcrossv

@Jamie Hall  But a pressure range from 0 to a few ounces doesn't change the temperature much. I do see some steam boiler controllers with an outside temp sensor, so that's a big plus.

Jamie Hall

delcrossv said:

@Jamie Hall  But a pressure range from 0 to a few ounces doesn't change the temperature much. I do see some steam boiler controllers with an outside temp sensor, so that's a big plus.

To change the temperature significantly in a steam system, you have to change the pressure a lot. The only feasible way to control a steam system is to turn the boiler on and off with a suitable balance between on time and off time to produce an average temperature of your emitters to produce the amount of heat you want. There are lots of ways to do this. By far the simplest is a thermostat in the space, provided you can tolerate the necessary differential between the temperature at which it switches on and the temp where it switches off, and provided it is calibrated to take into account both the lag in heat delivery on start up and the lag in the end of heating on shutdown. The old mercury T87 did that astonishingly well...

Do not confuse controlling the pressure within narrow limits -- which is needed on certain types of two pipe steam systems -- with attempting to control the emitter temperature. Two completely and totally different things. The pressure controller in this case is interested in only pressure for correct operation, and not in the temperature of the emitters -- nor in the heat provided. The heat provided control is the thermostat (or outdoor reset, if used).

Nor should you confuse controlling the pressure to avoid damage, a safety function, which is needed on any steam system. In this case the pressure controller is there entirely for the safe operation of the system.

PMJ

The assumption always seems to be that any change to a decades old control is too complicated and or costly, so reasonably tight temperature control has to be good enough. None that know me would describe me as a perfectionist. The complication and cost in what I have done is all in the capturing and viewing of data to be able to study the process. The actual control that has resulted from the effort is quite simple. I have resisted writing complicated code for this work, even though the PLC platform on which to do it has been there the whole time. Perhaps it is that this market has already rejected a very fine PLC based controI. And, I suppose I would just like to share the results of what I have found, perhaps slowing the obvious and unfolding demise of residential steam in the process. It really can be so much better.

The control I am thinking about might be $100. Wire a bypass switch with it so with one flick the whole thing isn't there at all. I run my system with no input of outside conditions at all, just reacting to changing demand on the piping - just as the original system did.

Jamie Hall

Please be aware -- not only @PMJ but others -- that I am not addressing how you or anyone else should control your steam system -- or any other device you happen to own. Nor am I advocating particularly for -- or against -- any control system you or anyone else should happen to fancy. That's your business. What I am trying to do is to address some of the more interesting aspects and implications of control theory -- in this forum, how they apply to heating systems.

As it happens, I have a strong philosophical aversion to using as a control a parameter which has no direct, definable, and systematic relationship to the desired outcome. Probably stemming from an aviation background, where there have been a number of outstanding fatal accidents resulting from doing so.

PMJ

@Jamie Hall , in my control the room temperature is still the final authority. I merely point out that temperature in the room cannot cycle an on/off boiler with the precision required to maintain even heat. I found really even heat meant operating entirely inside what would be the deadband of the thermostat making it of little use for what I needed to do.

The very clever dead men before me used the tiny variations in pressure inside the piping to modulate the fire through manipulation of the damper. By your definition I would assume that this internal pressure is also not directly related in any way to the temperature of the room. I, however, see how effective it would have been at matching an output to the need, and have merely copied their method of monitoring a condition of the piping that clearly varies with demand also. It turns out that the actual real time demand for steam may be the best indicator of what is needed. Outdoor temperature is from only one location around the structure and calculations based on it may vary a lot due to wind and sun on the other sides. That is why I never went there.

So what I have found is that the real time demand for the steam required to maintain a temperature is what needs to be monitored. The the current temperature in the room tells me nothing about that. It turns out that what is going on in the piping tells a lot about it.

delcrossv

PMJ said:

The assumption always seems to be that any change to a decades old control is too complicated and or costly, so reasonably tight temperature control has to be good enough. None that know me would describe me as a perfectionist. The complication and cost in what I have done is all in the capturing and viewing of data to be able to study the process. The actual control that has resulted from the effort is quite simple. I have resisted writing complicated code for this work, even though the PLC platform on which to do it has been there the whole time. Perhaps it is that this market has already rejected a very fine PLC based controI. And, I suppose I would just like to share the results of what I have found, perhaps slowing the obvious and unfolding demise of residential steam in the process. It really can be so much better.

The control I am thinking about might be $100. Wire a bypass switch with it so with one flick the whole thing isn't there at all. I run my system with no input of outside conditions at all, just reacting to changing demand on the piping - just as the original system did.

I'd be very interested in how you do that. Sounds very interesting.

Jamie Hall

Having missed the points of most of my attempt at discussion, I give up.

delcrossv

Jamie Hall said:

Having missed the points of most of my attempt at discussion, I give up.

The only feasible way to control a steam system is to turn the boiler on and off with a suitable balance between on time and off time to produce an average temperature of your emitters to produce the amount of heat you want.

I get where you're going, but , correct me if I'm wrong, but as installed the boiler was never "off" but rather went to very low input (dampers closed). A modulating / multistage gas set up could mimic that more closely. It seems (to my untrained eye) that the modulation could be controlled by pressure and the t-stat just serves as a final cutoff/ cut in. I do note that the Smith 19HE can be ordered with a two stage burner.

Always interested in learning something new

NYtimebomb

PMJ said:

The assumption always seems to be that any change to a decades old control is too complicated and or costly, so reasonably tight temperature control has to be good enough. None that know me would describe me as a perfectionist. The complication and cost in what I have done is all in the capturing and viewing of data to be able to study the process. The actual control that has resulted from the effort is quite simple. I have resisted writing complicated code for this work, even though the PLC platform on which to do it has been there the whole time. Perhaps it is that this market has already rejected a very fine PLC based controI. And, I suppose I would just like to share the results of what I have found, perhaps slowing the obvious and unfolding demise of residential steam in the process. It really can be so much better.

The control I am thinking about might be $100. Wire a bypass switch with it so with one flick the whole thing isn't there at all. I run my system with no input of outside conditions at all, just reacting to changing demand on the piping - just as the original system did.

Do share.

Adding complexity to any system will always have naysayers, but does its benefits outweigh its costs or will it add more risk to the system?

I'm always perturbed by the black and white of on/off mentality. With regards to steam, why not also modulate the amount of flame, not just if there is flame, to meet the desired outcome?

Jamie Hall

"why not also modulate the amount of flame, not just if there is flame"

not quite as simple as it sounds, @NYtimebomb , at least for oil -- and not really all that simple for gas, either.

EBEBRATT-Ed

@NYtimebomb

There is no modulating burner for oil in the typical residential size of equipment.

If there was a way to modulate a steam boiler it would have to be off room temperature.

There is not enough difference in steam pressure and steam temperature used in a residential boiler to allow for modulation unless the steam could be run as a vacuum system and even then the boiler would have to cycle at light load.

delcrossv

Jamie Hall said:

"why not also modulate the amount of flame, not just if there is flame"

not quite as simple as it sounds, @NYtimebomb , at least for oil -- and not really all that simple for gas, either.

Maybe not simple, but possible for gas. Hi flame to pressure setpoint A, low flame to end of call for heat? Or get input off condensate temp? Hmmm. I'm curious what control would go with Smith's 2 stage burner.

@EBEBRATT-Ed Pulling a vacuum just increases steaming time with the burner off, yes?

Jamie Hall

Folks. OI said I was going to quit on this thread, but I have to add this for my own sanity if no one else's. First, @EBEBRATT-Ed is correct -- of course! -- to say there are no modulating oil burners of residential size.

Second, "pulling a vacuum" on a steam system has two quite different meanings -- one is some systems use, or used, vacuum pumps to help remove air at the start of a cycle or even during a cycle. This does not affect the operating pressure and hence phase transition temperature. Others were arranged so that while the vents would open to release air, they would remain shut under a vacuum, at least for a time. These have some interesting properties and, if it were possible to truly seal such a system as one does a heat pump or air conditioner, some interesting potential.

Third, obtaining clean, high efficiency combustion even with gas is difficult; with liquid fuels its even more so. Read some of the work in that for gas turbines.

But last (I hope) -- be sure you clearly understand exactly how the system you are controlling works, what the target parameters for the system really are, what the controlling parameters for the system are, and how various components of the system interact. To be blunt, it is clear from some of the comments that this isn't always the case.

ChrisJ

You don't need to modulate steam temperature in order to increase comfort.

If you decrease the output of the boiler you decrease how much of the radiation is at 212F+-.

This decreases output into the rooms.
Yes, it does work when done correctly. As anyone that ever ran a coal fired steam system.


Me and @PMJ disagree on some things, be we also agree on a lot.
"Modern" controls for steam systems suck.

NYtimebomb

Keeping it simple goes a long way. I do not claim to know much about heating systems, simply asking questions to encourage the great minds here to share so I and others learn.

delcrossv said:

Maybe not simple, but possible for gas. Hi flame to pressure setpoint A, low flame to end of call for heat? Or get input off condensate temp? Hmmm. I'm curious what control would go with Smith's 2 stage burner.

That was essentially what I was thinking. Seems old school t-stats mentioned (T87?) and 'anticipators' weren't quite there yet.
Just seems a waste of so much fuel to fire the boiler at 'max' the entire time. I know they are two entirely different beasts, but the forced air system at my previous home would 'cut out' the flame but continue to blow the heated air once the set temp was reached. It just got me thinking about it with regards to the new heating system I am now owner of.

PMJ

There seems to be some interest and requests to share more.

Let's continue by defining what even heat is, and what getting it requires of the delivered steam.

Even comfortable heat by definition means spending lots of time between cut in and cut out of the thermostat(long calls for heat). You can't control the rate of fall from cut out but you can the rate of rise from cut in. Even heat means raising as slowly as possible, just never falling. Doing this will require more than one burn cycle per call. That is why the thermostat by itself is a poor control method for intermittently fired steam - anticipated or not.

So we are talking about a situation where we don't want the temperature of the room to raise quickly. You DO NOT want to go quickly from cut in to cut out! You want it to take a long time, the longer the better. This requires then radiator conditions that don't change a lot. They can't go back and forth from full and hot to room temperature. So you can't let your boiler run full on until the thermostat says stop no matter what its size. One burn per call for heat is the way it has been done for decades, and that is what needs to change to even out the heat.

Accomplishing the above with an intermittently fired boiler simply requires controlled digital pulses of steam. It can be done with any boiler - even one significantly oversized - it doesn't matter. It doesn't take many more cycles to dramatically improve things. It does not require flame modulation. I'm sure that would be interesting, but not realistic in this the dying business of residential steam heat. Steam pulses can be controlled a number of different ways, none of them complicated, and certainly not expensive. So the project becomes how to pulse fire the boiler so as to surely always raise the room temperature, but as slowly as possible.

If the above makes sense to those interested in pursuing this, then we can move to the details. If anything does not make sense, I think we should get on the same page with those things before going further.

delcrossv

NYtimebomb said:

Keeping it simple goes a long way. I do not claim to know much about heating systems, simply asking questions to encourage the great minds here to share so I and others learn.

delcrossv said:

Maybe not simple, but possible for gas. Hi flame to pressure setpoint A, low flame to end of call for heat? Or get input off condensate temp? Hmmm. I'm curious what control would go with Smith's 2 stage burner.

That was essentially what I was thinking. Seems old school t-stats mentioned (T87?) and 'anticipators' weren't quite there yet.
Just seems a waste of so much fuel to fire the boiler at 'max' the entire time. I know they are two entirely different beasts, but the forced air system at my previous home would 'cut out' the flame but continue to blow the heated air once the set temp was reached. It just got me thinking about it with regards to the new heating system I am now owner of.

Exactly. Sort of like the difference between full boil and "simmer". A coal fired boiler, which is what these systems were originally fired with, wouldn't go from cold off to full on but hi fire (dampers open) to low fire (dampers closed). Seems that could be mimicked with gas at least in part. I might be talking out my hat, but reheating all that iron and water from a cold start several times a day seems less efficient than having a longer cycle on low flame with occasional bursts of high fire to maintain temp.

Jamie Hall

Interesting. No wonder you and I find it hard to find common ground -- since even in the third paragraph I find I must disagree, in its entirety.

Carry on.

delcrossv

PMJ said:

There seems to be some interest and requests to share more.

Let's continue by defining what even heat is, and what getting it requires of the delivered steam.

Even comfortable heat by definition means spending lots of time between cut in and cut out of the thermostat(long calls for heat). You can't control the rate of fall from cut out but you can the rate of rise from cut in. Even heat means raising as slowly as possible, just never falling. Doing this will require more than one burn cycle per call. That is why the thermostat by itself is a poor control method for intermittently fired steam - anticipated or not.

With you so far, seems these systems want long cycle times.

So we are talking about a situation where we don't want the temperature of the room to raise quickly. You DO NOT want to go quickly from cut in to cut out! You want it to take a long time, the longer the better. This requires then radiator conditions that don't change a lot. They can't go back and forth from full and hot to room temperature. So you can't let your boiler run full on until the thermostat says stop no matter what its size. One burn per call for heat is the way it has been done for decades, and that is what needs to change to even out the heat.

Yes. Like solid fuel fired . Warm to hot- not cold to hot

Accomplishing the above with an intermittently fired boiler simply requires controlled digital pulses of steam. It can be done with any boiler - even one significantly oversized - it doesn't matter. It doesn't take many more cycles to dramatically improve things. It does not require flame modulation. I'm sure that would be interesting, but not realistic in this the dying business of residential steam heat. Steam pulses can be controlled a number of different ways, none of them complicated, and certainly not expensive. So the project becomes how to pulse fire the boiler so as to surely always raise the room temperature, but as slowly as possible.

Pulse or modulate the flame- 6 of one, half dozen of the other, yes?

If the above makes sense to those interested in pursuing this, then we can move to the details. If anything does not make sense, I think we should get on the same page with those things before going further.

I'm interested in hearing details.

ChrisJ

My system fires the burner when the thermostat calls for heat. It then shuts down early with the thermostat still calling. Then the thermostat is usually satisfied 5-10 minutes later.

Very rarely do I need multiple firings for a single call.

PMJ

ChrisJ said:

My system fires the burner when the thermostat calls for heat. It then shuts down early with the thermostat still calling. Then the thermostat is usually satisfied 5-10 minutes later. Very rarely do I need multiple firings for a single call.

No one "needs" more than one firing. I'm pointing out that the more even the heat is the longer the calls must be. Very long calls must have multiple firings. Perfectly even heat would mean the temperature never changed at all, the first call for heat would never be satisfied, and that one endless call for heat would have all the cycles for a whole season.

I'm just looking for agreement that these facts about even heat are true. Whether more even heat is worth any effort is a separate subject, a personal choice issue. I have no issue with anyone choosing to stay with one call one burn, none at all. 

PMJ

@delcrossv, I will post some control details soon.

PMJ

Jamie Hall said:

Interesting. No wonder you and I find it hard to find common ground -- since even in the third paragraph I find I must disagree, in its entirety. Carry on.

@Jamie Hall, as always I'm happy to discuss specifics when I see any to discuss. 

PMJ

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

ChrisJ

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

I wish I knew.
I've been confused ever since it happened.

PMJ

ChrisJ said:

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

I wish I knew.
I've been confused ever since it happened.

May I assume you have a backup unit on hand and a program copy?

ChrisJ

PMJ said:

ChrisJ said:

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

I wish I knew.
I've been confused ever since it happened.

May I assume you have a backup unit on hand and a program copy?

PMJ

ChrisJ said:

PMJ said:

ChrisJ said:

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

I wish I knew.
I've been confused ever since it happened.

May I assume you have a backup unit on hand and a program copy?

Me. Just curious as I know you have a great system that is no longer available and I'm sure you don't want to lose it. Perhaps Mark has agreed to support it anyway.

That it is none of my business works too.

KC_Jones

PMJ said:

ChrisJ said:

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

I wish I knew.
I've been confused ever since it happened.

May I assume you have a backup unit on hand and a program copy?

I can't answer for Chris, but I have no back up for mine. So far so good. Mark would probably assist in some way if we asked, but I don't think the actual business exists with insurance etc. so in some way it would be a big ask on our part.

I don't understand why it wouldn't take off either, IMHO it is on the list of greatest upgrades to my system.

I can vouch for how good it can work too. I have my thermostat set at 72, the ecosteam set for 70. Once we get to cold weather the thermostat will have a continuous call for heat all day and sit right at 70 degrees. It fires exactly what is needed and those radiators are warm/hot all day long, for as many sections are needed. Since my radiation is right on the edge, when it drops to design, they are all fully hot all day long. No hissing vents, no noise just warmth and it's delightful.

During the shoulder season it will go up to 72 and hit the thermostat. In my house this is preferred since the radiators won't be hot all day, a slightly higher overall temperature keeps it comfortable. This IMHO is due to less radiant, which equates to less comfort. If I'm being honest, this is where a hot water system starts winning out over steam. I'll shield myself for the rock throwing now.

ChrisJ

I bypassed the Ecosteam for a while, a month or so and the wife noticed and wasn't happy.

So, there you have it.

delcrossv

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

Sounds like a Tekmar279 with all the sensors (outdoor, room, condensate) wired in.(???) Similar?

KC_Jones

delcrossv said:

PMJ said:

Also @ChrisJ, so readers aren't confused, they should be told that you are running a very sophisticated PLC based control which calculates each burn based on structure and outside temp inputs. So what it can do in a single burn is way better the standard control. 

Sadly, this market rejected that very fine solution. You tell me why that was.

Sounds like a Tekmar279 with all the sensors (outdoor, room, condensate) wired in.(???) Similar?

It's a bit different and 1/3 the cost.

If I recall the tekmar doesn't have input for actual building heatloss it's some % deal. The EcoSteam you input actual building heatloss, outdoor design, boiler output, indoor design temperature and then calculates run time based on those inputs. It has a boost function (similar to Tekmar) which is based on an offset temperature. So if I setback 3 degrees I set the ecoSteam for 3 degrees below setpoint before boost kicks in.

He also added a sort of boost or curve offset for lower temperatures, basically allowing slightly longer run times when it gets really cold. What Chris and I found was all these systems have a shortcoming and that is when it gets windy and your infiltration actually increases (old drafty houses) the system would fall behind. Since everything is temperature based the best solution here was to add a temperature boost to adjust the curve.

Honestly for any reset system, a wind sensor, or ability to tie into a local weather station would be the final piece of the puzzle IMHO.

I believe the ecosteam is very close in concept to the Tekmar, but a bit more refined, and, most importantly, it was significantly cheaper. Also it still leverages the thermostat which I think the Tekmar does not, but again I am only vaguely familiar with the Tekmar product.

PMJ

I studied the Ecosteam and would have bought one if I could have.

All in to me a very fine solution. That you two are quite pleased comes as no surprise. If I were in your shoes I would do what it takes to get a spare unit loaded with the program on hand. These PLC's are really great until suddenly they aren't. Fortunately at this level they don't cost much. Cost is a non-issue considering what they provide.