Replacement boiler sizing in SE MA

[Dana]

A common rule of thumb for short cycling are burns shorter than 5 minutes, more than 5 burns per hour. At 5 minutes into the burn the average efficiency will usually be within 1% of the steady state efficiency, after discounting for flue purge and ignition cycle losses. At 10 minutes it's within a tiny fraction of a percent.

With higher frequency of ignition cycles the wear & tear goes up, ignition components fail, etc. There's no magic to the number 5- fewer is always better, but 5 burns in an hour with an average length of 6 minutes is a heluva lot easier the equipment and more efficient than 10 burns averaging 3 minutes, which is the same overall duty-cycle.

A differential swing of 30F below 145F isn't necessarily a problem, but can be depending on how long it takes the return water to get north of 130F. This boiler can take the thermal shock of cooler return water, but if it's chronically spending 75%+ of the burn times at temps below 130F it's likely to be a problem. A system bypass branch with a thermostatic mixing valve set to mix direct boiler output into the return water stream whenever the entering water temp drops below 130-135 F can help avoid that. A cheaper approach is to put a ball valve on the branch and hand-tweak it, but with an outdoor reset control that isn't going to work as well as a thermostatic mixing valve.

One approach to a system bypass looks like this:

The tee at the bottom just ahead of the pump would be where the thermostatic mixer goes, and the balance valve goes away. With the thermostatic control when ever the entering water temp at the pump is below the valve's setpoint there it reduces the return flow from the system, takes more flow from the bypass branch directly from the boiler's output. When the return water is eventually above the valve's setpoint it closes off the bypass branch completely.

Take a look at (or even a picture of) the near boiler plumbing, see what if any bypass measures were installed, just in case it needs to be tweaked.

 

[PeterDux]

Thanks for the clarification on short cycling. My old boiler would do that sometimes, sometimes running for only a few seconds. And burn times well under 10 minutes were very common, even when it was very cold out.

One more question about settings: You write:

Either that, or set the water temp at +11F to where the 163' of baseboard is emitting 51,000 BTU/hr, which would be about (51,000/163'=) ~ 315 BTU/hr per foot. For typical fin-tube baseboard that happens at an AWT of about 140F (145F out, 135F back, or thereabouts.)

You mention "setting the water temp" and also an "output temp" elsewhere in the post. Is this the Hb, or High Boiler Temp? If I want to go with this scenario, would I change that to 145, and set the Low Outdoor temp to +11 instead of -13? And if I do that, would the boiler bump the output temperature up when the outdoor temperature falls below 11F?

And in describing water returning to the boiler, when you say "how long it takes return water to get above 130F", does that mean that when a zone is calling for heat the return water may initially be cooler, but will get warmer as time goes on? So if it's only, say, 110 at first, but then climbs to 130 or more, then there won't be any issues with thermal shock or condensation? I'm thinking of the zone where my mom is (36 ft.) which is also the farthest from the boiler (probably about 35 ft horizontal in the basement and then up a floor before it hits the baseboards). All the space is heated or heavily insulated (the vertical run is in an insulated chase in the insulated garage). That zone runs a lot, as she is often calling for 76 or 78 degrees. I would guess that during the day the return water would get warm enough without a mixing valve, if I'm understand it correctly. And if I was concerned about it, wouldn't another alternative be to raise the boiler output temperature, or change the differential temperature setting to 20 or even the default 15 degrees?

Here are a couple photos.

 

[Sponsor]

 

[Dana]

You mention "setting the water temp" and also an "output temp" elsewhere in the post. Is this the Hb, or High Boiler Temp? If I want to go with this scenario, would I change that to 145, and set the Low Outdoor temp to +11 instead of -13? And if I do that, would the boiler bump the output temperature up when the outdoor temperature falls below 11F?

I have not read the full manual on the reset card (but you should). Until just now I had only been interpreting the short sheet you attached, so much of this is guessing at the interpretation. Hopefully you have a printed current version of the manual for the IQ Reset Option Card (?) to refer to?

To the extent possible you want to tune the curve so that the heat emitted by the radiation matches the load, which will make for long steady calls for heat, and very steady room temperatures.

Take a look at the graph on p13 of the reset card manual. Lo is the outdoor temperature at which the water needs to be hottest, and the radiation is at it's maximum output. Hb is the temperature you want the boiler to deliver at the programmed Lo temp. If you're leaving Hb at the default 180F, the idea is to set Lo to the temperature at which the heat loss from the house matches the heat being delivered by the radiation when the entering water temp (EWT) is 180F. Based on the 943BTU/hr per degree-F number from your fuel-use heat load calculation the heat load of the house will roughly match the output of the baseboard radiation (not counting the toe-kick) when the EWT is 180F somewhere around -13F or so.

The heat loss of each zone isn't exactly proportional to the radiation, so there may need to be some adjustment if some zone or another doesn't keep up when it's really cold out, but 180F water at -13F is a pretty good starting point.

Ho is the outdoor temp at the other end of the reset curve, which corresponds to the Lb temperature parameter. According the graphic on p.13 the default Lb is 70F, but your short-sheet called out 110F as the default, and that is verified in the decription of the Lb temperature parameter further down on p.10.

Based on verbiage from the main boiler manual for the series I believe 110F the lowest non-abusive temperature to start firing the boiler. Below that there is some amount of thermal shock/mechanical stress to the boiler, so limiting the number of times it has to truly cold start below that temp every year promotes longevity. But it should fine to let the controls purge heat out of the boiler down to 110F when the heat loads are low, but not all the way down to 70F.

The Lt parameter (bottom of p14) is the lower limit for when the boiler is actively firing during a long call for heat, which needs to be above the safe condensation temperature. With low mass radiation like fin-tube it's fine to leave it at the default. In fact it's probably better/safer to program Ho to be the temperature at which the heat load of the house matches the heat emitted by the radiation with Lt set to 140F, so that the return water entering the boiler will always be 130F or higher when cycling during extended calls for heat during warmer weather. This should be measured and verified early in the heating season in case it needs to be tweaked.

At an EWT of 140F or an AWT of 135F the baseboard is emitting about 300 BTU/hr per foot, so x 163' of baseboard you're looking at about 49,000 BTU/hr. Using your 943 BTU/hr per degree constant derived by fuel use the heat load of the house will match that that somewhere around 65F -( 49,000/943)= +13F.

But you can't set Ho to +13F, the lowest it will allow is +35F. The approximate heat load at +35F is 943 x (65F - 35F) = 28,290 BTU/hr. So the most the 163' of baseboard needs to emit at that outdoor temp is 28,290 / 163'= ~175 BTU/hr, which is about what you get at an AWT of 115F. So it looks like a reasonable first cut is to set Ho to +35F and Lb to 120F.

I suspect with those settings it won't short cycle or have condensing issues, but if it does we can take a closer look as the season gets rolling.

And in describing water returning to the boiler, when you say "how long it takes return water to get above 130F", does that mean that when a zone is calling for heat the return water may initially be cooler, but will get warmer as time goes on?

If you let the boiler cool down to 110F and the water in the radiation & distribution plumbing is starting out at 70F after being off for awhile it will always take some amount of time for the burner to raise the output temp high enough that the returning water temp entering the boiler is above 130F. There isn't a huge amount of thermal mass in the boiler but some, and with 3/4" plumbing baseboard there isn't a lot of thermal mass in the distribution & baseboard to worry about either. I expect heat up times to be short. If you had high volume cast iron rads it would be worth estimating the thermal mass and calculating the time, but you don't, so just measuring/timing it early on in the season when calls for heat are happening when the radiation is starting up cold and the boiler is near is coolest temp would be enough to know if we need to make any adjustments (to either the parameters or the plumbing.)

 

[Dana]

BTW: What's on the other side of the pink fluff in the studwall?

For the fiberglass to perform to spec it needs air barriers on all 6 sizes- a layer of drywall would take care of that. But depending on the whole stackup of that assembly it may be worth making some adjustments to keep from turning it into a mold-farm when you close it in.

 

[PeterDux]

Thanks for your response and the link to the reset card manual. I hadn't located that. I will read it. I find it interesting that boilers now have these sophisticated controls, but installers (or my installer at least) doesn't make an effort to use them to optimize system operation. I guess they just leave the default settings because of the old "people only complain when there's not enough heat" ethic.

Anyway, I'll study this and make the starting settings. And as you say, we'll see what happens as fall comes.

There's a finished room on the other side of the fluff: that's the playroom I don't use these days. I could drywall it in. It really hasn't been a consideration up until now, as with the old boiler the basement area was always warmer than the playroom.

 

[Dana]

Thanks for your response and the link to the reset card manual. I hadn't located that. I will read it. I find it interesting that boilers now have these sophisticated controls, but installers (or my installer at least) doesn't make an effort to use them to optimize system operation. I guess they just leave the default settings because of the old "people only complain when there's not enough heat" ethic.

Anyway, I'll study this and make the starting settings. And as you say, we'll see what happens as fall comes.

There's a finished room on the other side of the fluff: that's the playroom I don't use these days. I could drywall it in. It really hasn't been a consideration up until now, as with the old boiler the basement area was always warmer than the playroom.

Some installers even take it a step further, setting them up to always run at 180F, basically defeating the outdoor reset, out of concern that the default settings might leave the customer cold! (With condensing gas boilers that's a pretty serious efficiency hit, not just a comfort hit, forcing it to run at 85% - 87% efficiency when it could be averaging in the mid-90s while providing better comfort.)

 

[PeterDux]

I haven't made any changes to the boiler yet, but I have had an opportunity to read the manual. It's interesting because it seems some set points will override others, and I'm not sure how to best adjust others.

• There's a separate domestic hot water set point. This temp is requested when DHW is needed, regardless of the boiler temp for heating. Seems this should be at 180, the default.
• There's a setting for domestic hot water priority time. If it's at zero, then DHW isn't a priority ever. I can set it for between 0 and 101 minutes. Since my system typically replenishes hot water (after a shower, for example) for 5-10 min., I was thinking I'd set it for 10 min. I also want to make sure the system is wired to run the circulator to the boilermate when this is activated.
• There's also a "boost time" parameter. If home heat is not satisfied within the time selected, the reset card will increase the operating setpoint by 10F. This will repeat if the demand isn't satisfied within the next 20 min (I think, it doesn't specify) until demand is satisfied or the boiler reaches its high limit setting. Default is 20 min, range is 0-30 min. I was thinking of setting this to 30 min as I'd prefer it not increase temp to baseboards when running for longer periods of time, such as after a nighttime setback.
• I wonder why I'd use -13F as the low outdoor temperature. I realize that's what the heat loss (or demand) calculation indicates, but I've lived in this area pretty much all my life and never seen lower than -5 or -8F. We may get one day a year (some years none) that are 0F. But perhaps that doesn't matter?
• I'm a little confused about the High Boiler Water Temperature setting. They also call this the Boiler Water Design Temperature, which we calculated as 145F. But the manual says for convection baseboard fin tubes, high boiler water temperature should be 160-190F. Not sure where to set this one. 160F, perhaps? I also wonder what takes precedence, the High Boiler Water Temperature in the outdoor reset card, or the High Limit Setting on the IQ Control System? Does the outdoor reset card override the control system?

The more I dig into this the more complex it seems. But it also seems that correct settings should have a significant benefit in comfort and fuel consumption.

 

[Dana]

• I wonder why I'd use -13F as the low outdoor temperature. I realize that's what the heat loss (or demand) calculation indicates, but I've lived in this area pretty much all my life and never seen lower than -5 or -8F. We may get one day a year (some years none) that are 0F. But perhaps that doesn't matter?

The way we got to -13F for the cold end of the curve was: "Based on the 943BTU/hr per degree-F number from your fuel-use heat load calculation the heat load of the house will roughly match the output of the baseboard radiation (not counting the toe-kick) when the EWT is 180F somewhere around -13F or so."

... as explained in response #23.

It doesn't matter that it never hits -13F in your neighborhood, but only need the boiler output to hit 180F when it's that cold outside. So when you set up the curve to hit 180F @ -13F it'll still do the right thing at warmer temps, say when it's "only" -5F or 0F outside, since it automatically adjusts the target temperature downward as temps rise. We could just as easily have kept the 0F default and estimated the (lower than 180F) water temp needed to heat the place at 0F.

• I'm a little confused about the High Boiler Water Temperature setting. They also call this the Boiler Water Design Temperature, which we calculated as 145F. But the manual says for convection baseboard fin tubes, high boiler water temperature should be 160-190F. Not sure where to set this one. 160F, perhaps? I also wonder what takes precedence, the High Boiler Water Temperature in the outdoor reset card, or the High Limit Setting on the IQ Control System? Does the outdoor reset card override the control system?

The "High Boiler Water Temperature" or "Boiler Water Design Temperature" is the max temp you set for the boiler at the cold end of the curve (the 180F @ -13F that I estimated.) See the graph on page 13 of the manual- notice that the default curve is to have the boiler's target temp of 180F when the outdoor temp is 0F. But your load isn't high enough at 0F to warrant 180F water, so I'm recommending you move that coordinate to 180F @ -13F.

A temp of 145F is the lowest the burner can be operated continuously without high risk of destructive condensation, but letting the controls heat-purge to 140F is still safe. I estimated that you have enough baseboard to fully heat the place at +13F outdoors with 140F water, so there isn't really much of a reset curve to speak of- most of the time you'll be running at the lower boiler temp, which is fine.


The high-limit on the IQ Control should be set high, and should be overridden by the outdoor reset control until/unless it's cold enough outside that the curve would actually exceed that limit, or the Boost function is asking it to exceed that limit. So if you set the IQ control's high-limit to 190F, the reset curve won't reach 190F until it's -20F outside or something (or if the Boost time interval is set super short and it ratchets up fast.)

Don't sweat the Boost interval too much. Most of the time even at the lowest boiler temp on the curve the thermostats would be satisfied well within 20 minutes when just maintaining the setback temp, but when recovering from setback the Boost function guarantees that you'll get the house back up to temp in a reasonable amount of time even when it's cooler outside. If in practice it takes too long to recover from deep setbacks, shorten the Boost interval to 15 minutes or 10 minutes, see what happens.

If it's working fine, not cycling on/off during or other misbehavior during calls for heat from the indirect, don't mess with the domestic hot water set point on the controller, since the aquastat on the indirect (or the pumping rate) would have to be tweaked to keep it all running well without cycling or excessive temperature differences. It's fine to adjust the priority interval, but as you point out, recovery times are quite fast when getting the full output of the boiler dedicated to the hot water.

 

[PeterDux]

I got cold feet (pun intended) about changing settings on the boiler myself, so I had my contractor stop by to adjust the settings. They also have a Burnham hand held that makes setting it up much easier. The tech they sent does all the setups for them, and after some initial resistance I was able to have him describe what they typically do and how, and it resulted in setting some things differently than I originally intended.

Here's what we did:

• First, the tech said that the high limit setting on the main control overrides all other settings, and suggested we keep it where he set it on install, at 195. He said it's more of a fail safe than anything else, in case the boiler goes over the 180 it's set at elsewhere.
• We set pre-purge at 5 min instead of 10: He said they've measured boiler temps during pre-purge and if a zone calls for heat the boiler temp drops very rapidly, and is often at 100 or so within 2-3 min. We set it at 5 min, since the zone that it will usually be serving is small.

For the Reset Option Card:

• Set point stayed at 180, he wanted to set the domestic set point at 190. Seemed OK to me. We left Priority Time a 0, since I suspect getting hot water won't be an issue. As he pointed out, it never has been before.
• We left boost time at 20 min.
• After much discussion, we left low and high outdoor temps at 0 and 70. He showed me the heating curve and he said that with the low temp at 0 we'll get full output of the boiler at that temp. Since I have a lot of glass in my living room (8 windows, a set of french doors, and a 15 window single door) I've found that when the outside temp drops to 0 that room is only just able to maintain temperature. And if we set it at -13 he said it would never run at 180, since it never gets to -13 where I live. And he insisted that if we set the high temp at 35, then the boiler would not fire to deliver heat above that temperature. Circulators would run if the thermostat calls for heat, but the burner would not fire. That's obviously not what I wanted, so we left it at 70.
• And he said that we we set the high boiler temp at 140 or 145, the boiler will never exceed that temperature, except during boost time. The reset curve and the two ends of the scale (0 and 70) determine the boiler output temperature, but this setting is a not to exceed number. If he's right it makes sense.

We'll see how it works. I haven't run the heat in the main house yet, although I was tempted to this morning. But my mom has run her zone and it seems to be working fine.

 

[PeterDux]

UPDATE: We've had a few cold days, finally, morning temps in the teens. System is working beautifully. This morning with 4 of 5 zones calling for heat (5th zone is rarely, if ever, used) the boiler was still cycling, but run times were pretty long. Honestly it's so quiet I don't really notice when it stops. Last weekend with temps in the 40s and thermostats maintaining daytime temps (68 for the house, 76(!) for my mom's suite), it would cycle, but run times were usually 8-12 mins. Seems pretty ideal. The heat purging when a zone first calls for heat seems to be a great feature: I notice that happening often. Baseboard temps in cool (not frigid) weather are lower, and the house is more comfortable. Temps don't over-shoot the requested temperature. Hot water heating is zero problem, even without hot water as a priority zone. The basement isn't super cold, but the ground isn't really frozen yet. I may work on improving the circulating pipe insulation in the basement to reduce waste heat.

Overall, I'm very happy with this system, and even happier that I held out for the smaller boiler. I'll track fuel deliveries this winter to see if we're using less, but regardless we have a more comfortable house!

 

[Dana]

Sounds like it's working alright then!

It hasn't been super-cold yet but it's been within ~5F of the 99% design temperatures at least once this week in most of eastern MA. So if it's still cycling you have more than enough capacity, but the fact that the cycles are long is a big part of the increased capacity. The real test will be on the next Polar Vortex disturbance cold snap, but from what you've observed so far it should do just fine.

I'll bet the boiler room is no longer the warmest place in the house too!

 

[PeterDux]

It is working great. I will be interested in how it performs when it (or if) it gets really cold, but I'm not concerned. I think it will do fine.

Basement isn't as cool as I expected. However, the ground is still pretty warm. Dana, I've been reading your other posts about insulating distribution and return pipes properly, as I have some pretty long runs across the basement ceiling, and the pipes now have the black thermal wrap on them. I wonder if going to 1" fiberglass would make a significant difference.

 

[Dana]

It is working great. I will be interested in how it performs when it (or if) it gets really cold, but I'm not concerned. I think it will do fine.

Basement isn't as cool as I expected. However, the ground is still pretty warm. Dana, I've been reading your other posts about insulating distribution and return pipes properly, as I have some pretty long runs across the basement ceiling, and the pipes now have the black thermal wrap on them. I wonder if going to 1" fiberglass would make a significant difference.

If the basement walls are reasonably tight and have some insulation it'll probably stay above 60F even if the distribution pipes are insulated. R3 is current code minimum for any pipes that are warmer than 105F. I'm not sure what the "thermal wrap" is, but even foam pipe insulation would have at least a half-inch wall thickness to meet that spec, as would 1" fiberglass.

If you have an infra-red thermometer you should be able to find the hot-spots pretty easily. The boiler jacket and flue will of course be pretty warm if it's running any duty cycle. Bare copper & bronze, or even galvanized iron have very low emissivity and would yield false low temperatures, but a wrap of hockey tape or a spot of spray paint (any non-metallic color) would yield realistic temperature readings.

 

[PeterDux]

I meant foam pipe insulation. And the contractor didn't put it on the new plumbing when they installed the boiler. I'll do that first. Basement temps are in the 60s, and will probably drop some when we get some consistent cold weather. Definitely cooler than before.

 

[Dana]

I meant foam pipe insulation. And the contractor didn't put it on the new plumbing when they installed the boiler. I'll do that first. Basement temps are in the 60s, and will probably drop some when we get some consistent cold weather. Definitely cooler than before.

Be sure to check the operating temperature limits of any foam pipe insulation you install- some can handle 180F plumbing, others can't. (Fiberglass pipe insulation can take 220F steam and higher.)

 

[PeterDux]

Start of 2020/21 heating season update: After a full season with the MPOIQ84 I've gone back and calculated my home's fuel load using Dana's heating size calculator. I used different dates because they're based on when the oil tank is filled, but actually covered a longer time period: in 2018/9 I used from 2 January to 12 March, in 2019/20 I used 20 November to 25 February. I realize some of this time may not be considered peak heating season, but we had a very mild March in Massachusetts, so I didn't have an oil fill up after 25 February until 03 April.

Two interesting pieces of information came out of the calculation:

• The balance point calculation for boiler size went down, from 50,992 BTU at 65 to 42876. I imagine this is because of (a) milder temps during the different timeframe perhaps; and (b) that my old boiler was less efficient than its label stated. Not surprising after 24 years.
• Overall fuel use for the season dropped drematically, from 1430 gallons for the 2018/19 season to 899 gallons for 2019/20. I was surprised enough by this to check degree days for same date ranges in the two years, and found that 2019/20 was about 7% warmer. That's probably a pretty significant number. But I was also using slightly higher thermostat settings in the more recent season.

My monthly fuel budget number has dropped from $417/mo (!) to $269/mo. Of course lower oil prices are a big help, but still I'm pretty happy about that. And the house was very comfortable. The only thing I noticed is during the one cold snap we had (low single digits for a couple nights) my window intensive living room was slow to warm up. I increased the temp where the system generates 180 degree water to the baseboards from 0F to 10F to help out with the relative shortage of baseboards in that space. And I plan to use smaller overnight temperature setbacks this winter.

Overall this boiler selection seems to be a win. I'm really happy I didn't give in to the contractor's persistently suggesting a larger size boiler.

Thanks, Dana!

 

[Dana]

Start of 2020/21 heating season update: After a full season with the MPOIQ84 I've gone back and calculated my home's fuel load using Dana's heating size calculator. I used different dates because they're based on when the oil tank is filled, but actually covered a longer time period: in 2018/9 I used from 2 January to 12 March, in 2019/20 I used 20 November to 25 February. I realize some of this time may not be considered peak heating season, but we had a very mild March in Massachusetts, so I didn't have an oil fill up after 25 February until 03 April.

Two interesting pieces of information came out of the calculation:

• The balance point calculation for boiler size went down, from 50,992 BTU at 65 to 42876. I imagine this is because of (a) milder temps during the different timeframe perhaps; and (b) that my old boiler was less efficient than its label stated. Not surprising after 24 years.

The oversizing leading to ultra-low duty cycle (=excessive standby loss) was probably a bigger factor than mere age of the boiler. The difference in measured heat load with the new vs. old boiler would indicate an 18% difference in efficiency, and that's real. But it's very unlikely that it's the raw combustion efficiency that was 18% worse, rather the "as-used AFUE", which is largely due to oversizing.

• Overall fuel use for the season dropped drematically, from 1430 gallons for the 2018/19 season to 899 gallons for 2019/20. I was surprised enough by this to check degree days for same date ranges in the two years, and found that 2019/20 was about 7% warmer. That's probably a pretty significant number. But I was also using slightly higher thermostat settings in the more recent season.

My monthly fuel budget number has dropped from $417/mo (!) to $269/mo. Of course lower oil prices are a big help, but still I'm pretty happy about that. And the house was very comfortable. The only thing I noticed is during the one cold snap we had (low single digits for a couple nights) my window intensive living room was slow to warm up. I increased the temp where the system generates 180 degree water to the baseboards from 0F to 10F to help out with the relative shortage of baseboards in that space. And I plan to use smaller overnight temperature setbacks this winter.

Overall this boiler selection seems to be a win. I'm really happy I didn't give in to the contractor's persistently suggesting a larger size boiler.

Thanks, Dana!

It looks like you won the "right-sizing" game! (You're welcome!)

It's unfortunate that many or even most installers don't really understand how oversizing eats into efficiency and comfort. If they can tune the burner at 85% combustion efficiency they assume it's going to perform at the same efficiency as any other boiler that tests at 85% combustion efficiency, which simply isn't the case.

 

[PeterDux]

Eighteen months after install we finally got some cold weather here in Southeastern MA! Single digit F temps for three nights, daytime highs in the teens. Also very windy the first couple cold days, which provided an opportunity to test whether or not a much smaller boiler than originally spec'd for this house (84K BTU vs 151K BTU) would do the job.

Short answer is, no problem! With three or four of the five zones calling for heat (I rarely use the basement playroom) run times were 20-30 min with 10-15 min between runs. I didn't use night time setbacks on the first floor as we've always been short of baseboard in the main living space (few walls, lots of windows), but the system had no difficulty maintaining 68 or 70 degrees. And my mom was comfortable in her space with the thermostat at 76 or higher. I'm also happy that the basement stayed relatively cool, unlike it was with the old boiler.

Without the data Dana's fuel use calculations provided I probably would have given in to my contractor's pressure to put in a boiler one size up from the MPOIQ84, or worse. Thanks, Dana!

 

[Dana]

Once again "your welcome"- I'm happy to know that arithmetic still works!