Vaillant S53 compared to other brands

[fezster]

Hi All.

For those who read my LLH thread, this question is because I'm weighing up whether my problems are just down to it being a Vaillant boiler.

Quick background - I have 3 heating zones (u/s, d/s and hw). Any individual zone works fine. Multiple zones work fine when running together. The problems occur when the boiler is firing and individual zones open and close, because the return temp to the boiler suddenly cools, and it hits S53.

If you imagine a typical heating system, zones open and close as the stat calls for heat, so depending on timings, sometimes the boiler can sit in S53 for hours, and the heating runs only luke warm. Other times the rads are at full temp.

My question is this - is this a Vaillant problem? (specifically the 438). Or do all boilers have this issue? I just cannot believe in the modern day with people having multiple underfloor heating zones etc that this problem does not occur more often and for more people. Unless everyone fits a Low Loss Header when running multiple zones?

Thanks for reading and appreciate any input.

 

[Last Plumber]

I think there is more likely an issue with the system.
You can't rule out the Boiler obviously as Vaillant Boilers do develop faults like any other.
I can only speak from my own experience here, I have been installing Vaillant for 15 years ish, Commercial and domestic and I can't say I have ever known one sit in S53 for as long as yours seems to.
I still think your best bet is to get a Heating Engineer to look at the whole picture.

 

[fezster]

Thanks. I do have an engineer coming next week with regards to a LLH installation.

Have you fitted a Vaillant on a heating system with multiple zones? I think the S53 in the scenario I described is inevitable. Please correct me if Im wrong?

What's frustrating is the amount of time it sits in S53, which seems to be unpredictable. But the ideal aim is for it to not hit S53 in the first place. What would work is if an opening of any zone first turned the boiler off (allowing the return temperature to stabilise between the zone which is hot and the zone which is cold) and then to fire up again so that the delta between flow and temp never exceeds 30 degrees. Maybe a relay arrangement of some sort.

 

[Last Plumber]

Thanks. I do have an engineer coming next week with regards to a LLH installation.

Have you fitted a Vaillant on a heating system with multiple zones? I think the S53 in the scenario I described is inevitable. Please correct me if Im wrong?

What's frustrating is the amount of time it sits in S53, which seems to be unpredictable. But the ideal aim is for it to not hit S53 in the first place. What would work is if an opening of any zone first turned the boiler off (allowing the return temperature to stabilise between the zone which is hot and the zone which is cold) and then to fire up again so that the delta between flow and temp never exceeds 30 degrees. Maybe a relay arrangement of some sort.

Yes I install them in all sorts of buildings, all sizes, multiple zoned too.
I have one in my own home with 3 zones. No UFH (Under floor heating).
I have not come across your problem. That's why I think it needs looking at.
Have you checked the flow and return temps at the Boiler ?
You can check on D40 and D41.
Have the UFH zones been set up properly?

 

[fezster]

No UFH heating zones. Only rads - upstairs zone, donwstairs zone, HW zone.

I've monitored D40 and D41 for individual zones and zones in combination. The delta is always roughly 20 degrees now (I say now, because it took some fine tuning to achieve). It may creep up to 22 degrees, but no more.

It's *only* when a cold zone (say upstairs heating) opens whilst another zone (downstairs heating) is fully hot. So D40 will be 70, D41 at 50 - then the upstairs zone opens and D41 shoots down to 30. Boiler sees the large differential and goes into S53. It then waits an arbitrary time (must be some logic to it?) before powering to full power again. At which point, the delta is again 20.

However, what happens far too often is that whilst in S53, the u/s room stat becomes satisfied and closes the zone. When the boiler comes out of S53, it's only the d/s zone open and so it fires back up to 70/50. Then the u/s zone opens again and the cycle repeats. I should note this sequence doesn't happen all the time, but far too often. The result is that the d/s zone always remains luke warm and the stat temp is never satisfied. This may be because during the day (on weekends) we have the d/s quite hot with young children in the house, and leave u/s relatively cool.

I also suspect (not proven) that the boiler increases the S53 duration every time it occurs, exacerbating the situation.

Apologies for the long post. Appreciate you reading and taking the time.

 

[Last Plumber]

No UFH heating zones. Only rads - upstairs zone, donwstairs zone, HW zone.

I've monitored D40 and D41 for individual zones and zones in combination. The delta is always roughly 20 degrees now (I say now, because it took some fine tuning to achieve). It may creep up to 22 degrees, but no more.

It's *only* when a cold zone (say upstairs heating) opens whilst another zone (downstairs heating) is fully hot. So D40 will be 70, D41 at 50 - then the upstairs zone opens and D41 shoots down to 30. Boiler sees the large differential and goes into S53. It then waits an arbitrary time (must be some logic to it?) before powering to full power again. At which point, the delta is again 20.

However, what happens far too often is that whilst in S53, the u/s room stat becomes satisfied and closes the zone. When the boiler comes out of S53, it's only the d/s zone open and so it fires back up to 70/50. Then the u/s zone opens again and the cycle repeats. I should note this sequence doesn't happen all the time, but far too often. The result is that the d/s zone always remains luke warm and the stat temp is never satisfied. This may be because during the day (on weekends) we have the d/s quite hot with young children in the house, and leave u/s relatively cool.

I also suspect (not proven) that the boiler increases the S53 duration every time it occurs, exacerbating the situation.

Apologies for the long post. Appreciate you reading and taking the time.

I misunderstood your post. I thought you were saying that You had underfloor heating. Sorry about that.

Why have you set the temperature differential to 20 degrees ?

 

[Chalked]

In my experience, that boiler suffers from poor circulation when the pump is fitted ( as it usually is) on the flow outside of the boiler.
Vaillant system boilers have the pump fitted to the boiler return, when fitted integrally.
I would go with a low loss header and fit a 15/60 pump to the return leg.

 

[fezster]

Why have you set the temperature differential to 20 degrees ?

As I understood it, this is the temperature differential condensing boilers run optimally at. My rads are sized according to this differential. Although getting the differential any less will be quite difficult with my current setup regardless - the flow rates become much larger at 11 degrees.

In my experience, that boiler suffers from poor circulation when the pump is fitted ( as it usually is) on the flow outside of the boiler.
Vaillant system boilers have the pump fitted to the boiler return, when fitted integrally.
I would go with a low loss header and fit a 15/60 pump to the return leg.

Appreciate that tip. Most of the LLH diagrams I've seen show the pump on the return.

 

[tolly]

Whos sized the boiler...do you need 38kw? What do you have D0 set to in the diagnostics menu?

 

[tolly]

Auto bypass fitted aswell?

 

[fezster]

Whos sized the boiler...do you need 38kw? What do you have D0 set to in the diagnostics menu?

The installer recommended it and I did a total house heat loss calc, which came to 34KW (this was 5 years ago). And it said to allow an additional 3KW for the hot water cylinder. D0 is set to 34.

Auto bypass fitted aswell?

Yes. It's a Honeywell DU144 I believe. Set according to the charts to 0.1. Without it, it was impossible to achieve a 20 degree delta at the boiler.

 

[doitmyself]

It's a Honeywell DU144 I believe. Set according to the charts to 0.1. Without it, it was impossible to achieve a 20 degree delta at the boiler.

A 20C delta is just the recommended maximum; it doesn't matter if it is less. What sort of delta were you getting with the DU144 on a higher setting?

Let's say that the calculated head is 2 metres, the flow rate is 500 litres/hour, and we have a 15-50 pump. The attached pdf shows what happens in real life: the actual flow rate is 573 litres/hr and the head is 2.63 metres. This is because the working point has to be on one of the fixed curves (in this case speed 1). At 573 litres/hr instead of 500, the temperature differential will be 20x500/573 = 17.45C. So you need to balance for that temperature, not 20C.

When demand reduces the boiler will automatically reduce output (modulate down). If the pump (assuming constant speed, e.g UPS15-50) has been set to give a 20C differential at 34kW, then at 17kW the differential will be 10C, and 5C at 8.5kW. It's only when the boiler is unable to modulate any lower (approx 6.5kW for the 438), that the ABV needs to come into operation to prevent the differential rising above 20C.

 

[fezster]

I used the chart in the manual:

http://www.honeywelluk.com/documents/Full-Specification/pdf/801.pdf

The theory - So at 1462 l/h (34KW) and at 650 mbar (6.5m total resistance) differential pressure, the setting is 0.1. My index circuit is 2.4m downstairs and 1.6m upstairs. The 438 has a hex resistance of 3.5m at 1462 l/h. I've taken worst case of 4.05m at 38KW, hence arrived at a total resistance of 6.5m.

The reality - anything above 0.1 on the ABV and the delta at the boiler rises. At 0.3 for example, the delta is closer to 26 degrees. At 0.1, it just about manages 20 degrees, usually 22/23.

The grundfos 25-80 can overcome 6.7m head at 0.4 l/s (1462 l/h):

https://www.pumpsalesdirect.co.uk/m...s-light-commercial-circulators-data-sheet.pdf

Even if just the u/s zone is open, total resistance 4.05+1.6 = 5.65m (far below what the 25-80 is capable of), I still see a delta T of around 22.

I think I understand what you say about the boiler modulating down but Im not sure how that ties into what I am seeing above. I know the 438 fires at max (or 60%?) output for the first minute or so before modulating down. It's possible the actual index circuit resistance is higher than that which has been calculated. In any case, it varies depending on the zone which is open and the flow through the boiler (given the biggest head loss is in the boiler).

 

[doitmyself]

The reality - anything above 0.1 on the ABV and the delta at the boiler rises. At 0.3 for example, the delta is closer to 26 degrees.

You are using the ABV to compensate for too low a flow rate. If the delta is 26C for an output of 34kW, the flow rate is 18.77 litres/min, whereas it should be 24.4 litres/min to give a delta of 20C. You're just feeding hot water straight back to the boiler, which won't help efficiency.

You said, in another topic, that you are running at 75C flow, 55C return. The output of a rad is not constant, it varies with flow, return and room temperatures. Assuming a 20C room temperature, the total output of your rads at 75C/55C is 38.8kW. But your boiler is only giving of 34kW, so the the output from the rads can only be 34kW. As the flow temperature is fixed, the return temp automatically reduces to balance the rad output to the boiler output. This is achieved when the return is about 49C, which gives a delta of 26C. Where have we seen that before?

Lowering the flow temperature to 71C, with a 20C delta, will reduce the rad output to about 34kW, matching the boiler output.

I know the 438 fires at max (or 60%?) output for the first minute or so before modulating down.

Have you confirmed this by observation? The fan speed is the parameter to monitor as it controls the boiler output; d.33 is the target, d.34 the actual. If you checked this with the boiler output (d.0) set to different values, you would soon see if changing d.0 has any effect during the first minute.

 

[fezster]

Lowering the flow temperature to 71C, with a 20C delta, will reduce the rad output to about 34kW, matching the boiler output.

So how do you achieve this? I'm assuming you don't mean just reduce the dial on the front of the boiler - the boiler obviously ramps up the heat from 30 to 75 flow, so simply limiting it's max temperature can't be the answer, as the 26 degree delta persists throughout.

Assuming a 20C room temperature, the total output of your rads at 75C/55C is 38.8kW

Sorry, how did you work this out? Or did you mean the output required

In fact, the rads are oversized intentionally and total quoted output is close to 46KW.

You are using the ABV to compensate for too low a flow rate.

I'm not necessarily disagreeing with you, but do want to understand why you think this is the case. As per my previous post, the ABV has been set exactly as the MIs instruct (i.e. total head loss around 6.5m, setting of 0.1). The purpose of the ABV is to maintain minimum flow rate through the boiler as valves close. So either:

1. It's possible the head loss calc is greater than calculated. In which case the ABV setting would have to be even lower.

2. The pump is not delivering the head the manual indicates - maybe it needs to be on the return, as suggested by someone else?

 

[fezster]

Btw - before anyone says it. I am going down the Low Loss Header route, so all of this is hopefully academic. I'm just intrigued as to exactly how this all works.

 

[doitmyself]

I'm assuming you don't mean just reduce the dial on the front of the boiler - the boiler obviously ramps up the heat from 30 to 75 flow, so simply limiting it's max temperature can't be the answer, as the 26 degree delta persists throughout.

Parameter d.71 allows you to set the temperature when the knob is on max - the default is 75. If you change d.71 to 70 then"max" will be 70. Of course you can't stop some one turning the knob to a lower temperature!

Don't understand what you mean by "the 26C delta persists throughout". It can't be 26C when the flow is at 30C.

Sorry, how did you work this out? Or did you mean the output required

Most rad manufacturers publish a table showing the correction factor to use when the difference between room temperature and mean water temperature ([flow+return]/2) varies from the default of 50 [([75+65]/2) - 20]. Here's one example (there is very little difference between manufacturers):

5 = 0.050
10 = 0.123
15 = 0.209
20 = 0.304
25 = 0.406
30 = 0.515
35 = 0.629
40 = 0.748
45 = 0.872
50 = 1.000
55 = 1.132
60 = 1.267
65 = 1.406
70 = 1.549
75 = 1.694

If the flow is 75 and return 55, the mean water temp is 65. So the difference is 65-20=45C, which gives a correction factor of 0.872. So rads totalling a nominal 40kW (at 75/65/20) only produce 40 x 0.872= 34.88kW at 75/55/20.

As per my previous post, the ABV has been set exactly as the MIs instruct (i.e. total head , around 6.5m, setting of 0.1). The purpose of the ABV is to maintain minimum flow rate through the boiler as valves close.

Assuming that the head loss of 6.5m is correct, what flow rate did you use to arrive at the ABV setting of 0.1?

 

[fezster]

Parameter d.71

I actually just increased d.71 to 80 when the cold weather set in.

Don't understand what you mean by "the 26C delta persists throughout". It can't be 26C when the flow is at 30C.

Obviously at 30 the boiler is just ramping up. But as soon as it stabilises (say 50), the return will have a 26 delta (or 22/23 delta now the abv is at 0.1). This 26 delta will persist regardless of the flow going up. So if flow is 51, return is 25. 52, then 26. 70 then 44. You get the point. So surely flow temperature has nothing to do with the delta?

Most rad manufacturers publish a table showing the correction factor to use

Yes, this is precisely why the rads were oversized. One thing my original installer did get right. My whole house heat loss was calculated to be 34kw.

Assuming that the head loss of 6.5m is correct, what flow rate did you use to arrive at the ABV setting of 0.1?

1462 l/h (34kw at 20 delta T)

 

[doitmyself]

I actually just increased d.71 to 80 when the cold weather set in.

Why? The whole house calculator you used assumes a worst case, i.e. an outside temperature of -1C or lower. The rads are oversized so they produce the correct amount of heat with a 20C differential. There should therefore be no need to increase the flow temperature to 80C.

Obviously at 30 the boiler is just ramping up. But as soon as it stabilises (say 50), the return will have a 26 delta (or 22/23 delta now the abv is at 0.1). This 26 delta will persist regardless of the flow going up. So if flow is 51, return is 25. 52, then 26. 70 then 44. You get the point. So surely flow temperature has nothing to do with the delta?

It is a dynamic system. The boiler's output varies, as does the radiators', the flow rate and the room temperature. The only thing which is constant is (in your case) the rotation speed of the pump. Assuming constant water temperatures of 51C and 25C, when the room is cold (0C) your rads will be producing 30kW; at 10C, 19kW; and at 20C, 8.6kW. The boiler's controller must adjust the output to provide the correct amount of heat. Presumably it does this by ensuring that the delta across the hex does not a specified value, possibly 26C.

I wasn't only suggesting that you should drop the flow temperature to 71; I also said that you should then rebalance for a 20C delta. Your current rads add up to 45.44kW (according to the floor plans), but the boiler is set to 34kW. This gives a required rad factor of 0.748. From the table I posted, this is equivalent to a MWT-Room delta of 40C, giving a MWT of 40+20C = 60C. This equates to 70C flow and 50C return. So you set the flow to 70C and rebalance so the return is 50C (or as near as you can get with a fixed speed pump).

1462 l/h (34kw at 20 delta T)

I thought that was what you did. Unfortunately it's not the correct method. The attached Honeywell pdf explains how to set an ABV. In their example the required flow rate is 300 litres/hr. This gives a boiler output of 7kW when delta is 20C. I think this represents the minimum boiler output, not the maximum. As I said before, the modulation mechanism takes care of the delta when the output is between minimum and maximum output. Your boiler has a minimum output of 6.3kW, so the flow rate for a 20C delta is 4.5 litres/min (270 per hour) Assuming the 25/80 pump on speed 2, that flow rate means a head of 6.75 metres. From the DU144 graph, a flow of 270 litres and head of 6.75 metres requires a setting of 0.6. With the current setting of 0.1 there is a permanent flow of hot water straight back to the boiler, which you ca verify by feeling the pipe after the ABV. This pipe should normally be cold and only get hot when the valve opens, e.g when a zone valve closes.

 

[fezster]

I thought that was what you did. Unfortunately it's not the correct method. The attached Honeywell pdf explains how to set an ABV. In their example the required flow rate is 300 litres/hr. This gives a boiler output of 7kW when delta is 20C. I think this represents the minimum boiler output, not the maximum. As I said before, the modulation mechanism takes care of the delta when the output is between minimum and maximum output. Your boiler has a minimum output of 6.3kW, so the flow rate for a 20C delta is 4.5 litres/min (270 per hour)

Hi doitmyself.

I think this makes a lot of sense, and I read some previous posts by you on this topic. Eg.

Two questions about a bypass valve

Which made all of this clearer to myself (so thanks!).

However, the problem with taking this approach is that Vaillant boilers (and perhaps others?) - especially the ecotec plus range - fire at max output (or some %age of max output) for the first minute or so. This means that unless you set the minimum flow rate based on max output (i.e. that which is specified in the manual - approx 1600 l/h for a 438 @ 38KW and delta T of 20), then during that initial firing stage, you will likely hit S53.

 

[MoniczkaMoniczka]

For me vaillant is the best