Class 86/2s, TDM and DBSOs

[hexagon789]

This came up on another forum but we couldn't find an answer. When the 86/2s operated with DBSOs on the GEML how exactly was the driving done from the DBSO?

I have a reasonable idea of how TDM works with tap-changer locos and DVTs with their fully variable power handles but the DBSO retained their 4-notch EMU-style power controllers.

My question is how was each notch translated by the loco? Did it correspond to an approximate % of tractive effort and the loco would dial the taps up or down as appropriate or did each notch correspond to an exact "tap". Someone suggested taps 9, 18, 28 and 38 would be most likely but they weren't in a position to confirm.

It's something that's bothered for ages and any answer would be most welcome, thanks. It also means I can go over to the other forum and fill in the blanks there.

 

[rebmcr]

I can't remember where I learned it, but I believe the notches correspond to the 86's power handle's Run Up and Run Down positions, requiring a reasonably skilled driver to avoid tripping the breakers without the luxury of a visible ammeter.

 

[Railperf]

Any idea @dk1 ?

 

[hexagon789]

I can't remember where I learned it, but I believe the notches correspond to the 86's power handle's Run Up and Run Down positions, requiring a reasonably skilled driver to avoid tripping the breakers without the luxury of a visible ammeter.

Seen now that was how I imagined it, but the person who suggested the taps I listed above said that wasn't how it worked, each Notch gave produced a "demand" for power exactly as it did when the DBSOs were used with 47/7s, to me that meant a specific tap for each notch but that doesn't seem like enough control somehow.

If the notches do correspond to setting on the tap-changer handle, there's not enough for all the positions so then you have the question of which settings were available.

Complicated but hopefully people understand what I'm getting at.

Either way there seems only two possibilities -

1. Each notch gave a specific level of power
2. Each notch corresponded to certain of the run up/down, notch up/down and hold of the tap-changer on the loco and thus drivers would notch up and down in the DBSO, in contrast to DVTs where they just set a level of power and the TDM requested it off the loco

 

[captainbigun]

Short answer is the same as if a DVT had detents at third, half, 3/4 and full power. TDM manages TE demand, the loco side control then interprets that. In effect your TE demand is converted to current demand which will depend on a variety of things so it does not derive a specific notch. Indeed the control side (TET - tractive effort translator) has no comprehension of notch number. It merely controls the movement of the tap changer against the demand and the actual current feedback.

 

[hexagon789]

Short answer is the same as if a DVT had detents at third, half, 3/4 and full power. TDM manages TE demand, the loco side control then interprets that. In effect your TE demand is converted to current demand which will depend on a variety of things so it does not derive a specific notch. Indeed the control side (TET - tractive effort translator) has no comprehension of notch number. It merely controls the movement of the tap changer against the demand and the actual current feedback.

So the power demand corresponds to those fractions of continous tractive effort, with the TDM dialling up and down the taps as required to maintain that rather than locking to specific taps?

 

[captainbigun]

TE demand uses 8 TDM channels. This gives you 8 bits and therefore 256 intervals between 0 and 100%.

The TE demand is taken, converted to a crude current demand. This takes into account a max current limit that differs between 86 and 87. There’s a simple algorithm which then runs the tap changer up based upon the calculated current demand. All the while the actual current is being fed back. Within a tolerance, when actual meets demand the tap changer is held.

The same applies should demand be decreased or increased with the tap changer being run up or down as required.

TET controls the tap changer not the TDM. TET is only found on 86/87. 89/90/91 are all able to process the 8 bit demand.

 

[hexagon789]

TE demand uses 8 TDM channels. This gives you 8 bits and therefore 256 intervals between 0 and 100%.

The TE demand is taken, converted to a crude current demand. This takes into account a max current limit that differs between 86 and 87. There’s a simple algorithm which then runs the tap changer up based upon the calculated current demand. All the while the actual current is being fed back. Within a tolerance, when actual meets demand the tap changer is held.

The same applies should demand be decreased or increased with the tap changer being run up or down as required.

TET controls the tap changer not the TDM. TET is only found on 86/87. 89/90/91 are all able to process the 8 bit demand.

So the 4 notches generate a "demand" for motor current would it be correct to say? And they correspond to 1/3, 1/2, 3/4 and max TE?

 

[Wilts Wanderer]

This is a really interesting thread, I never considered the complexity of controlling a tap-changer loco from an effectively thyristor-Control driving vehicle.

Two questions - could you overload the traction motors on the loco via TDM, or was there a limiter to make up for the lack of ammeter displays? Secondly could a Class 90 multi-work with a class 86 or 87 and therefore did it work in a similar way if they could.

 

[liam456]

What an awesome thread; a question I had never asked myself, then when I saw it, immediately needed an answer to. Then a similarly satisfying answer!

 

[hexagon789]

This is a really interesting thread, I never considered the complexity of controlling a tap-changer loco from an effectively thyristor-Control driving vehicle.

Two questions - could you overload the traction motors on the loco via TDM, or was there a limiter to make up for the lack of ammeter displays? Secondly could a Class 90 multi-work with a class 86 or 87 and therefore did it work in a similar way if they could.

I'm sure more knowledgable people will correct me if im slightly wrong on the finer points but I believe that current was specifically limited to prevent overload on the 86/87s in some manner, with 90s the loco microprocessor did that itself.

The second bit I do know, tap-changer locos can multi with thyristor ones (86s have run freight with 90s) but it's preferred for the tap-changer loco to lead in such cases.

 

[captainbigun]

So the 4 notches generate a "demand" for motor current would it be correct to say? And they correspond to 1/3, 1/2, 3/4 and max TE?

When the slave is an 86 or 87 yes, equally it could be a 89/90/91 in which case they’d apply the requested TE.

 

[hexagon789]

What an awesome thread; a question I had never asked myself, then when I saw it, immediately needed an answer to. Then a similarly satisfying answer!

I once asked the same about Scottish 47/7s and DBSOs and the answer to how they worked was both interesting and satisfying to know! I have wondered about 86s and DBSOs as well, but forgot about it until prompted to recall by a thread on another forum.

 

[hexagon789]

When the slave is an 86 or 87 yes, equally it could be a 89/90/91 in which case they’d apply the requested TE.

Thank you, it finally makes sense now! It's actually very similar to how it works with DVTs then but just with only four power settings instead of a range.

 

[captainbigun]

This is a really interesting thread, I never considered the complexity of controlling a tap-changer loco from an effectively thyristor-Control driving vehicle.

Two questions - could you overload the traction motors on the loco via TDM, or was there a limiter to make up for the lack of ammeter displays? Secondly could a Class 90 multi-work with a class 86 or 87 and therefore did it work in a similar way if they could.

The feedback loop on 86/87 is slow so you could without too much difficult find that the tap changer is still being run up passed the required notch. In that case yes you’d overload the loco. As above the maximum current for the loco is applied in the calculation for demand. That’s loosely based around the motor overload limits for the respective loco classes. You don’t have that problem in 89/90/91.

You can couple as many locos as you like, and the supply will allow, with TDM.

 

[captainbigun]

Thank you, it finally makes sense now! It's actually very similar to how it works with DVTs then but just with only four power settings instead of a range.

Yes, a DBSO power controller is a pot with detents, take away the detents and it’s in effect a rotary version of the DVT’s controller.

 

[hexagon789]

Yes, a DBSO power controller is a pot with detents, take away the detents and it’s in effect a rotary version of the DVT’s controller.

Cheers, that makes complete sense of course but general information elsewhere seemed to suggest something more complicated based on mimicking the tap-changer handle in the loco was how it worked but nothing explained it conclusively. You've both explained it very clearly but also shown that it was really much simpler than it might at first have appeared.

 

[Wilts Wanderer]

The answers continue to be fascinating, thanks everyone!

As a timetable planner I can’t help wondering now if train performance with the loco on the rear would be inferior to having the loco on the front, due to the different driving style required.

 

[dk1]

Any idea @dk1 ?

Overloading was a problem as it was pretty basic from the DBSO. I'm racking my brain to remember now. Think it went in 3 or 4 stages & you just became used to how much to give her. There was only a traction/brake indicator but no ammeter & it was very easy to give it 'too many taps'.

 

[hexagon789]

Overloading was a problem as it was pretty basic from the DBSO. I'm racking my brain to remember now. Think it went in 3 or 4 stages & you just became used to how much to give her. There was only a traction/brake indicator but no ammeter & it was very easy to give it 'too many taps'.

So relatively crude in that respect

 

[dk1]

So relatively crude in that respect

Very much so. Everything about DBSOs was crude mate.

 

[hexagon789]

Very much so. Everything about DBSOs was crude mate.

One final thing, I assume going to OFF on the power controller immediately cut power rather than the loco dropping down through the taps first?

 

[Mag_seven]

Did the DBSO get a separate set of controls for AC Electrics when they were transferred to Anglia from Scotland or did they use the original 47/7 controls?

 

[hexagon789]

Did the DBSO get a separate set of controls for AC Electrics when they were transferred to Anglia from Scotland or did they use the original 47/7 controls?

They kept the four-notch EMU power controller but the Frequency division multiplex two-wire control system unique to 47701-17 was changed for the superior Time Division Multiplex which of course remains standard today. The 47/7s also worked rather differently with how the power demand from the DBSO was translated compared with how the TDM system worked as described up-thread.

 

[dk1]

One final thing, I assume going to OFF on the power controller immediately cut power rather than the loco dropping down through the taps first?

Oh my now that's asking. I seem to recall any sudden drop caused them to 'kick' so had to down it slowly but 1-off was ok.

 

[hexagon789]

Oh my now that's asking. I seem to recall any sudden drop caused them to 'kick' so had to down it slowly but 1-off was ok.

Thanks, I appreciate I've been testing peoples' brains with this but I've found the discussion and answers very interesting, it all makes far more sense than some of the vague descriptions dotted about the internet