EMR Class 222’s - declining reliability

[Railperf]

Looks like only 18 222's starting the day for 20 diagrams. But with an 810 out, possibly a 222 covering that diagram. So only 17 /20 covered?

 

[222001]

Looks like only 18 222's starting the day for 20 diagrams. But with an 810 out, possibly a 222 covering that diagram. So only 17 /20 covered?

Two sets (011 and 103) were unfortunately involved in fatalities on Sunday/Monday, and they haven't returned to service since. So that will definitely be having a short term impact.

 

[tram21]

023 has was stuck just before Kettering for 75 ish minutes, and has now been terminated at Leicester following a fault.

 

[43074]

1B22 0812 Nottingham to London (222014) had a fault at Loughborough for around 30 minutes, which has caused quite a bit of disruption on EMR Intercity this morning.

Meanwhile, 1H16 0910 Corby to London seems to be having issues between Bedford and Flitwick, in both causes trains were able to divert to the slow lines to get around them.

Has to be said things do seem to have reached a bit of a low point on the MML recently.

 

[Railperf]

1B22 0812 Nottingham to London (222014) had a fault at Loughborough for around 30 minutes, which has caused quite a bit of disruption on EMR Intercity this morning.

Meanwhile, 1H16 0910 Corby to London seems to be having issues between Bedford and Flitwick, in both causes trains were able to divert to the slow lines to get around them.

Has to be said things do seem to have reached a bit of a low point on the MML recently.

222014 still performing well according to the berth timings south of Loughborough. Units 9, 13 and 19 also good runners today.
The slowest units today so far seem to be 15, 21, 23 and 101/103 pair. All losing time on their schedules!

 

[Dave61]

1B22 0812 Nottingham to London (222014) had a fault at Loughborough for around 30 minutes, which has caused quite a bit of disruption on EMR Intercity this morning.

Meanwhile, 1H16 0910 Corby to London seems to be having issues between Bedford and Flitwick, in both causes trains were able to divert to the slow lines to get around them.

Has to be said things do seem to have reached a bit of a low point on the MML recently.

Wife and I were thinking of heading to London for a day out later in April but have decided to give them a bit more time to sort things out

 

[Railperf]

222015 has only four engines running but only three providing traction.

 

[1D54]

1F37 - STP-SHF was 49 late departing with a problem on 006.

 

[tram21]

1F37 - STP-SHF was 49 late departing with a problem on 006.

And caused quite a few knock-on delays to services too.

 

[Railperf]

222015 has only four engines running but only three providing traction.

It is quite weird being above the engine that is running but not providing traction. It idles normally, but seems to respond to changes in power selection with a very slight rise of the revs but falling again to idle as the train moves off from rest and gathers pace.
I'm surprised there doesn't appear to be a travelling fitter joining en route able to resolve the issue- like in the days of HST's.

 

[43096]

I'm surprised there doesn't appear to be a travelling fitter joining en route able to resolve the issue- like in the days of HST's.

It's a bit easier for a TRI to work on an HST power car whilst in motion as the engine, alternator, control cubicle etc are all above floor. How does a TRI work on the engine or control system of a 222 on the move when it is under the floor?!

 

[LowLevel]

It's a bit easier for a TRI to work on an HST power car whilst in motion as the engine, alternator, control cubicle etc are all above floor. How does a TRI work on the engine or control system of a 222 on the move when it is under the floor?!

Usually by plugging their laptop in

 

[43074]

All losing time on their schedules!

Worth pointing out the xx:00 Sheffield to London and xx:12 Nottingham to London are consistently poor performing services, and there are a few short sectional running times in the mix as well, so losing time on EMR Intercity isn't necessarily a reflection of whether a unit has all working engines or not.

 

[londonmidland]

Worth pointing out the xx:00 Sheffield to London and xx:12 Nottingham to London are consistently poor performing services, and there are a few short sectional running times in the mix as well, so losing time on EMR Intercity isn't necessarily a reflection of whether a unit has all working engines or not.

I was going to say, unless you're actually driving said service(s), you can't actually determine how well or poorly a service is performing, by simply looking at RTT or seeing how many engines are out. There are so many factors which can affect punctuality.

 

[liamf656]

1F37 - STP-SHF was 49 late departing with a problem on 006.

The issue was with 222008 that caused the delay

== Doublepost prevention - post automatically merged: 21 Mar 2026 ==

Worth pointing out the xx:00 Sheffield to London and xx:12 Nottingham to London are consistently poor performing services, and there are a few short sectional running times in the mix as well, so losing time on EMR Intercity isn't necessarily a reflection of whether a unit has all working engines or not.

Exactly this. There are far too many assumptions in this thread and the 810 one about the performance of certain trains

 

[Railperf]

Worth pointing out the xx:00 Sheffield to London and xx:12 Nottingham to London are consistently poor performing services, and there are a few short sectional running times in the mix as well, so losing time on EMR Intercity isn't necessarily a reflection of whether a unit has all working engines or not.

Which sectional running times do you consider short?

 

[43074]

Which sectional running times do you consider short?

Most services lose time against the schedules for these sections, regardless of whether 222 or 810, or running on all engines or with one out

Sheffield to Dore Station Junction
Chesterfield South to Clay Cross Jns
Spondon to Long Eaton/Sheet Stores
Beeston South/Beeston to Trent East
Kilby Bridge Jn to Market Harborough (passing both)
Radlett Jn to Hendon

Dwell times at Leicester of 90 seconds are rarely achieved too.

In some cases these are balanced out by mandatory engineering allowances (e.g. approaching Derby). But basically there are elements of the current timetable which contribute to time loss overall, and this is reflected by the number of EMR services which, even on a good day, arrive at St Pancras 7+ minutes late having missed their paths at Harpenden.

Would be interesting to compare with stopwatch timings you must have gathered from hundreds of timing runs in the last few months to establish which sets are running with engines out?

 

[Amos390]

1F26 cancelled from St Pancras due to M8 - Problems with the train. The inbound service was 1C20 operated by 222011 and 222102. Not sure whether one or both of the units have failed, but they will return to Derby under 5Z26.

 

[Railperf]

Most services lose time against the schedules for these sections, regardless of whether 222 or 810, or running on all engines or with one out

Sheffield to Dore Station Junction
Chesterfield South to Clay Cross Jns
Spondon to Long Eaton/Sheet Stores
Beeston South/Beeston to Trent East
Kilby Bridge Jn to Market Harborough (passing both)
Radlett Jn to Hendon

Dwell times at Leicester of 90 seconds are rarely achieved too.

In some cases these are balanced out by mandatory engineering allowances (e.g. approaching Derby). But basically there are elements of the current timetable which contribute to time loss overall, and this is reflected by the number of EMR services which, even on a good day, arrive at St Pancras 7+ minutes late having missed their paths at Harpenden.

Lots to unpack here.

Does anyone know if the basis of the current EMR Intercity timetable was built around Class 222 sets with engines rated at the original 750bhp settings or the later 700bhp derate? And if it os designed around 700bhp settings, why are the timings short? And why haven't they been updated / reworked to reflect actual performance?
Regarding the tight sections above, how much time do you estimate needs to be added to each section to allow a reliable on time service?

Assuming that no TSr's are in force, would it be true on sections with tight timings that it would be preferable to have a healthy set running on all five engines or does the evidence suggest that performance equivalent to four engines is sufficient?

Would be interesting to compare with stopwatch timings you must have gathered from hundreds of timing runs in the last few months to establish which sets are running with engines out?

Network rail berth timing data helps to establish the performance of each set. No stopwatches required.

Take Chesterfield as an example. To compare the performance of trains starting southbound, look for the publicly available berth timings at signal berths TN4888 and TN 4844.

What you will find is that sets with 5 engines providing traction power will achieve a time of around 4 mins 15 seconds. A four engined set take around 4 min 35 seconds, and sets on three engines will be struggling to break the 5 minute barrier.

Carry out the same exercise using different signal berth timings at other station stops en route and on the train's diagram, and you start to build up a picture of how the train is performing.
On Saturday I established 222015 was setting berth to berth times equivalent to a set with only three engines providing traction power. Indeed when i joined it later that day, i was able to confirm that it had one engine out and another not providing traction.
Needless to say it was a very slow set, losing time on each section.
The berth timings do not lie!

== Doublepost prevention - post automatically merged: 22 Mar 2026 ==

Exactly this. There are far too many assumptions in this thread and the 810 one about the performance of certain trains

Taking Chesterfield as an example, if set 006 consistently produces times in the 4 min 12 second to 4 minute 18 bracket , while set 020 is consistently struggling to acheieve timings of 5 minutes, would it be true to conclude that 006 has five engines proividing traction, and 020 only having three engines providing traction? Or would that be an assumption?

And if you think they are simply assumptions, please explain why you think that?

 

[liamf656]

Taking Chesterfield as an example, if set 006 consistently produces times in the 4 min 12 second to 4 minute 18 bracket , while set 020 is consistently struggling to acheieve timings of 5 minutes, would it be true to conclude that 006 has five engines proividing traction, and 020 only having three engines providing traction? Or would that be an assumption?

And if you think they are simply assumptions, please explain why you think that?

I've seen many of your posts about the 222/810s over the past few months and they have been heavily critical without much to back your points up until now. It's a fair way to work it out but please be aware it is not completely accurate. I spend a lot of time on the 222s and some units you mention that are running on 3/4 engines have actually been running perfectly on 5 and were delayed for other reasons, and others which you have said were good performers, have failed on 2/3 engines and various other traction issues. Unless you stand on a platform and hear every single 222 with your own ears every day, you won't know for sure

 

[Railperf]

I've seen many of your posts about the 222/810s over the past few months and they have been heavily critical without much to back your points up until now. It's a fair way to work it out but please be aware it is not completely accurate. I spend a lot of time on the 222s and some units you mention that are running on 3/4 engines have actually been running perfectly on 5 and were delayed for other reasons, and others which you have said were good performers, have failed on 2/3 engines and various other traction issues. Unless you stand on a platform and hear every single 222 with your own ears every day, you won't know for sure

Hey, the forum is for sharing information. This thread is about declining reliability, and what I am reporting is what the signalling data plus my own observations tell us.
There's no criticism of staff here. so please try not to take this personally. At the end of the day, fare paying passengers are seeing more delays and cancellations as a result of the fleet changeover. That isn't something to be praised.

This thread started before Lumo's 222's went off-lease and was down to concern over the state of the 222 fleet - which has slowly gone downhill.
If you think any of the observations are inaccurate, you are welcome to message me or clarify the information here.

As I said before, the data is based on the diagrams being linked to the reported set numbers for those diagrams. If RTT states 006 is on 1B10, but RTT has missed a set swap somewhere, then it is possible the data for that diagram is being attributed to the wrong set. But if you notice that, please help clarify.

I get out on the MML quite a bit, so i'm usually able to verify the situation visually.
The berth timings are checked for as many stations as possible along a set's diagram. Therefore a slow start from one station doesn't mean the set is bad and we can correlate that with the starts from other stations.

If a set is producing slow timings along the whole diagram, and then its return diagram - we can say - with a reasonable degree of certainty - that it has a traction fault - even if it supposedly running with all 5 engines running!

As has been mentioned earlier, the timings are tight, and no driver I have ever spoken to says they can afford to dilly dally around. There tends to be a wider difference in braking points to stations, sometimes due to signal aspects, weather etc.

The 222's are pretty good accelerating in wet weather, but it is possible a 'good' set has some issues with wheelspin meaning a driver has to reduce power. But generally we don't see that affecting a good set consistently throughout the day from every station.

If you are aware a 'good' unit has other issues preventing the driver from using full power - feel free to share that info privately or here.
A set that appears to have all five engines running, but having to be nursed along at low power levels can't be classified as a healthy unit if it isn't able to put its full power to the rail.

222015 on Saturday was delayed departing St Pancras due to crewing issues - train manager on a late inbound service needing to have his mandatory break time.
Leaving around 10 mins later than booked, the 12:35 was now out of path meaning It then got caught behind Thameslink services at St Albans. It also experienced a slight signal check approaching Wigston. Leaving Leicester it experienced adverse signals before Syston Jn interrupting its acceleration to line speed. Arrival in Nottingham 19 minutes late!

On the return journey it was impeded approaching East Midlands Parkway by a late running Sheffield service until Leicester. And then being out of path experienced adverse signals until Wigston. Only after Wigston did it get a chance to run with a clear road ahead. But with only three vehicles proividing traction, acceleration was a lot slower than usual, and it it struggled up the hills at Kibworth, Desborough and Sharnbrook. Losing time on those sections. Though downhill it did reach 124mph before Bedford.

Actually it had such a clear run from Kettering south that it blitzed the 50 minute schedule with a sub 46 minute timing - although there are quite a few pathing allowances. A Healthy set with five engines would have been a couple of minutes quicker still. So a 14 minute late start fro Kettering turned into a 9 min late arrival at St Pancras - which I can't complain about.

I hope that with more 810's covering diagrams that the remaining 222's can get some more down time for remedial works. At the moment. It seems every single service I use experiences delay due to a combination of traction issues craeating delay - combined with other infrastructure issues.

 

[TT-ONR-NRN]

Hey, the forum is for sharing information. This thread is about declining reliability, and what I am reporting is what the signalling data plus my own observations tell us.
There's no criticism of staff here. so please try not to take this personally. At the end of the day, fare paying passengers are seeing more delays and cancellations as a result of the fleet changeover. That isn't something to be praised.

Exactly. Some people do give the impression that if they work in a specific area, the performance of that area therefore cannot be criticised as it's somehow offensive to them and their colleagues who are working hard.

No one doubts that those maintaining the 222s are working hard, however if the 222 reliability continues to be rubbish, consumers still have the right to criticise that reliability, especially on a public discussion platform. We cannot be expected to withhold rightful criticism of a currently unreliable stock type just in case it offends people who maintain them, that would be ridiculous.

 

[liamf656]

Hey, the forum is for sharing information. This thread is about declining reliability, and what I am reporting is what the signalling data plus my own observations tell us.
There's no criticism of staff here. so please try not to take this personally. At the end of the day, fare paying passengers are seeing more delays and cancellations as a result of the fleet changeover. That isn't something to be praised.

This thread started before Lumo's 222's went off-lease and was down to concern over the state of the 222 fleet - which has slowly gone downhill.
If you think any of the observations are inaccurate, you are welcome to message me or clarify the information here.

As I said before, the data is based on the diagrams being linked to the reported set numbers for those diagrams. If RTT states 006 is on 1B10, but RTT has missed a set swap somewhere, then it is possible the data for that diagram is being attributed to the wrong set. But if you notice that, please help clarify.

I get out on the MML quite a bit, so i'm usually able to verify the situation visually.
The berth timings are checked for as many stations as possible along a set's diagram. Therefore a slow start from one station doesn't mean the set is bad and we can correlate that with the starts from other stations.

If a set is producing slow timings along the whole diagram, and then its return diagram - we can say - with a reasonable degree of certainty - that it has a traction fault - even if it supposedly running with all 5 engines running!

As has been mentioned earlier, the timings are tight, and no driver I have ever spoken to says they can afford to dilly dally around. There tends to be a wider difference in braking points to stations, sometimes due to signal aspects, weather etc.

The 222's are pretty good accelerating in wet weather, but it is possible a 'good' set has some issues with wheelspin meaning a driver has to reduce power. But generally we don't see that affecting a good set consistently throughout the day from every station.

If you are aware a 'good' unit has other issues preventing the driver from using full power - feel free to share that info privately or here.
A set that appears to have all five engines running, but having to be nursed along at low power levels can't be classified as a healthy unit if it isn't able to put its full power to the rail.

222015 on Saturday was delayed departing St Pancras due to crewing issues - train manager on a late inbound service needing to have his mandatory break time.
Leaving around 10 mins later than booked, the 12:35 was now out of path meaning It then got caught behind Thameslink services at St Albans. It also experienced a slight signal check approaching Wigston. Leaving Leicester it experienced adverse signals before Syston Jn interrupting its acceleration to line speed. Arrival in Nottingham 19 minutes late!

On the return journey it was impeded approaching East Midlands Parkway by a late running Sheffield service until Leicester. And then being out of path experienced adverse signals until Wigston. Only after Wigston did it get a chance to run with a clear road ahead. But with only three vehicles proividing traction, acceleration was a lot slower than usual, and it it struggled up the hills at Kibworth, Desborough and Sharnbrook. Losing time on those sections. Though downhill it did reach 124mph before Bedford.

Actually it had such a clear run from Kettering south that it blitzed the 50 minute schedule with a sub 46 minute timing - although there are quite a few pathing allowances. A Healthy set with five engines would have been a couple of minutes quicker still. So a 14 minute late start fro Kettering turned into a 9 min late arrival at St Pancras - which I can't complain about.

I hope that with more 810's covering diagrams that the remaining 222's can get some more down time for remedial works. At the moment. It seems every single service I use experiences delay due to a combination of traction issues craeating delay - combined with other infrastructure issues.

Just to clarify, my intention wasn't to criticise your critique, but to wonder how you accumulated your information (and the dig was more at other posters who have made sweeping statements). You make some good points in your explanation

Of course maintenance teams are working to the bone to ensure all fleets have a good enough availability and have faced various challenges (many unforseen and aren't related to the engines). I do agree, diagram coverage has been poor but the past few days it's picked up again so hopefully it stays up, especially when we receive more 810s and the teams get more used to their quirks

 

[eastwestdivide]

At the risk of opening a can of worms, would looking at berth timings via the internet be subject to variable latency in getting the data from the track circuit via various systems to computer? Probably only really a problem if you were looking at one or two in isolation, and things would even out over the course of a journey? And then you're also looking at the reported timings as another cross-reference.
We're spoilt for data these days compared to 10-20 years ago.

 

[Merle Haggard]

At the risk of opening a can of worms, would looking at berth timings via the internet be subject to variable latency in getting the data from the track circuit via various systems to computer? Probably only really a problem if you were looking at one or two in isolation, and things would even out over the course of a journey? And then you're also looking at the reported timings as another cross-reference.
We're spoilt for data these days compared to 10-20 years ago.

Slightly differently but I've been on a train and looking at OTT- Maps to see that it's already berthed at a station when I know it hasn't, and waiting for an Up EMR Connect train at Wellingborough where the platform screen display shows 'Arrived' before it's come into sight.
But, as you suggest, it it's a constant time lag (or the opposite) it shouldn't be material.

 

[Railperf]

At the risk of opening a can of worms, would looking at berth timings via the internet be subject to variable latency in getting the data from the track circuit via various systems to computer? Probably only really a problem if you were looking at one or two in isolation, and things would even out over the course of a journey? And then you're also looking at the reported timings as another cross-reference.
We're spoilt for data these days compared to 10-20 years ago.

Network rail record the berth timings and these are accessible via the internet. Granted if you were to watch the open train times map, and to measure the times via a stopwatch, then you might experience some latency. By why do the hard work when it is done for you?

Here is an example of the sort of data you can find. Column 3 provides the timings, column 4 shows the berth data.

https://trackit.uppyjc.co.uk/TrackI...e&Estimate=False&DisplayMode=TD&Detailed=True

The timings are listed top to bottom from latest to earliest.

So to find Chesterfield, go down to the bottom of the page and then look in column four for TN4888 to 4878, note the time in column 3.

Then check the line which contains TN 4844 to 4842, and note the time along the same row. Now you can calculate the time taken to travel between the two berths to come up with a number.
There are other sites which also provide this data in other formats.

 

[eastwestdivide]

Network rail record the berth timings and these are accessible via the internet. Granted if you were to watch the open train times map, and to measure the times via a stopwatch, then you might experience some latency. By why do the hard work when it is done for you?

Here is an example of the sort of data you can find. Column 3 provides the timings, column 4 shows the berth data.

https://trackit.uppyjc.co.uk/TrackI...e&Estimate=False&DisplayMode=TD&Detailed=True

The timings are listed top to bottom from latest to earliest.

So to find Chesterfield, go down to the bottom of the page and then look in column four for TN4888 to 4878, note the time in column 3.

Then check the line which contains TN 4844 to 4842, and note the time along the same row. Now you can calculate the time taken to travel between the two berths to come up with a number.
There are other sites which also provide this data in other formats.

Aha that makes more sense: I had the mad idea that you were sitting watching one of the map sites! Didn't realise that level of detail was available. Thanks

 

[louis97]

Slightly differently but I've been on a train and looking at OTT- Maps to see that it's already berthed at a station when I know it hasn't, and waiting for an Up EMR Connect train at Wellingborough where the platform screen display shows 'Arrived' before it's come into sight.

The data doesn't show a train is "berthed at a station", it shows the train is somewhere between the previous signal it passed and the signal the berth is representing. The interpretation of the data via a map on the other hand may have a platform annotated next to a particular berth, but that is typically illustrating the platform relationship to the signal. The berth represents a wider area than just the platform.

 

[Merle Haggard]

The data doesn't show a train is "berthed at a station", it shows the train is somewhere between the previous signal it passed and the signal the berth is representing. The interpretation of the data via a map on the other hand may have a platform annotated next to a particular berth, but that is typically illustrating the platform relationship to the signal. The berth represents a wider area than just the platform.

Thanks, I understand that, but my point related to timing trains using berths and to illustrate the problem.
As it turns out, the train performance data being posted relies on more specific information.

 

[Railperf]

For context, the time penalty for a set on three engines on a northbound service leaving both Kettering and Market Harborough is 1.5 minutes on each section due to gradients.

On flatter sections of line such as Leicester and Loughborough northbound where the set has to accelerate to a line speed of 110/120/125mph, the time penalty is lower but no less than a minute for each station start.

So time loss will build up rapidly. A four engined set will lose slightly less than half that time.