Correlator Noise floor inflation by Legacy Signal Interfering with DCS

I dunno-

Your scope images show the composite track signal, I would assume (hope) that the DCS receiver/decoder has some frequency selectivity; the 455 kHz legacy signal is 4 octaves away! If the 455 kHz is getting through to the DCS decoder I can see the problem!

PLCProf posted:

I dunno-

Your scope images show the composite track signal, I would assume (hope) that the DCS receiver/decoder has some frequency selectivity; the 455 kHz legacy signal is 4 octaves away! If the 455 kHz is getting through to the DCS decoder I can see the problem!

 

When you look at the spread code MTH uses:  101100000110011010111101000110001, there's sections of 4&5 repeated symbols so you need to support bandwidth down to like 1/5 to 1/6 the bit rate to get clean eye windows, that's a pass-band starting at around 500-600 KHz. It's not a digital FIR filter so the stop-band rejection down at 450 KHz isn't going to be so great. I guess that's why it does come inside.

I wonder if a notch filter might help there, but it might need to actually be in the locomotive, probably a bit complicated.

FWIW, we've also noticed that at times the Legacy affects the DCS on the club layout.

gunrunnerjohn posted:

I wonder if a notch filter might help there, but it might need to actually be in the locomotive, probably a bit complicated.

FWIW, we've also noticed that at times the Legacy affects the DCS on the club layout.

Yup. It has to go in the engine. That's why it's hard to fix in a club with 100+ members.

I'm not sure where you'd have to put it either, a filter that handles full track power might be challenging.   Most of these have to pass the full signal, in our case that's track power.

I am think of taking the wire off the "U" terminal (I understand the signal is still there) and see if the signal degrades enough to add an MTH engine.  My biggest concern is someone turning off the Legacy signal while a Legacy engine is in operation and having that engine take off with 18 volts.

Some more quick testing to take a closer look at this. I blew the dust off the old FPGA decoder in the (old post) from last year and made the following setup:

So what we have is the FPGA decoder talking with the TIU exchanging packets and reporting the value of Cx (the packet correlation result) back to the PC. Then we have a signal generator injecting a 450 KHz tone with varying amplitude to see the correlation result Cx at different amplitude levels. Obviously doing only 1 packet is statistically meaningless so we compute the average Cx over 1000 exchanges at each voltage level. I've scaled the plot out of 16 instead of 31 (since 0 is actually 100% correlated but inverted) so the single axis can represent best to worst. This is with the actual MTH spreading code (obviously since I am talking to a real TIU)

I don't have control over the DCS excursion voltage since its set by the TIU and not adjustable but I did track it with the scope during the measurement (fixed at 6.17V in the test setup). As you can see from the above, once the injected 450 KHz tone gets to about 1.6V which is (about 25% (1.6 / 6.17) of the voltage of the DCS signal, the decoder correlation starts to drop off rapidly. I'm not sure if this scales linearly or not (like if we normalize the injection to DCS voltage will the drop at the same % each time).

It certainly doesn't fall apart gracefully, 1.2 to 1.8V (about a 0.6V change in injection) pretty much wipes out the correlator in my decoder. I'm sure the MTH one on the PS2/3 is pretty similar, at least according to their documentation.

I think the correlation confidence threshold Ct for MTH is somewhere around 13-14.

When Adrian came last night what he said is true and several other members that were their who contribute to this forum witnessed this strange event.

Any other thoughts?

Adrian, I wanted to thank you for spending the evening at the SD3R layout last night. A lot of the DCS users (me included) were going to leave the club over problems running DCS there. Finally, someone with real test equipment and an expert technical understanding DCS put the setup through a real diagnostic session.  The issues you discovered and that trick DCS panel you built gives SD3R's DCS users a light at the end of the tunnel--and I hope the light is an oncoming train...in O Gauge running DCS, that is. Thanks again!  

DJ's Trains posted:

Adrian, I wanted to thank you for spending the evening at the SD3R layout last night. A lot of the DCS users (me included) were going to leave the club over problems running DCS there. Finally, someone with real test equipment and an expert technical understanding DCS put the setup through a real diagnostic session.  The issues you discovered and that trick DCS panel you built gives SD3R's DCS users a light at the end of the tunnel--and I hope the light is an oncoming train...in O Gauge running DCS, that is. Thanks again!  

Thanks! I'm glad to help!

Immediately, I'm thinking it's probably best if you guys put in the Rev "M" TIU, get rid of the bulbs and other things that don't make sense on the track and just see how it works for a few weeks. Then I'll come back (probably mid June) and make sure the new TIU isn't degrading by repeating all the measurements we did last night. In parallel, I've got 2 or 3 ideas on new things we could build to manage the legacy interference (see above) that I'll be trying out prototypes for in our AGHR club the next few weeks or so...

After mid June I'll be a lot more free and I'm willing to come down to SD from LA as much as is needed to get your DCS running well. If there's more issues beyond what we already found, we'll figure it out together and share them with the community here!

We'll be beating on it in the coming weeks--trust me. I believe that Roger was going to install that DCS panel this or Tuesday evening.

Okay I have an interesting thought but I actually don't know that much about legacy and could use advice from someone like GRJ or someone else with legacy expertise... I actually only have DCS trains myself.

Thought Exercise:

I'm thinking the DCS packets are typically about 1.2 to 2.5ms in duration, and if a packet fails or gets a partial match, the TIU or PS2/3 will retry the exchange. If we just took a 555 timer and chopped a 6ms window into the legacy signal say every 25 ms the DCS should be able to find the holes through the retry process, and communicate in them provided they are wide enough, and frequent enough.

Something like this:

What would this do to the legacy signal?

1. Would it notice the windows (I would expect not since the current DCS packets themselves don't disrupt it).

2. I'm struggling to find good information about the Legacy signal format (modulation, duration, framing, error checking). If the command duration is longer than the window spacing (25ms) legacy will obviously fail. Is there a pulse width and duration that could satisfy both systems?

This may be a simple $4 fix...

Thoughts?

So just to be clear, nothing but MTH engines on layout with legacy active and MTH engines can not add?  Any other operator see this?  Lots of layouts run legacy and DCs so why just San Diego club.

Adrian! posted:

Immediately, I'm thinking it's probably best if you guys put in the Rev "M" TIU, get rid of the bulbs and other things that don't make sense on the track and just see how it works for a few weeks.

Rev "M" TIU?  What's that?

gunrunnerjohn posted:

Adrian! posted:

Immediately, I'm thinking it's probably best if you guys put in the Rev "M" TIU, get rid of the bulbs and other things that don't make sense on the track and just see how it works for a few weeks.

Rev "M" TIU?  What's that?

Rev "M" (with quotes) is the TIU with the 1n4148s inside and TVS. The one in my other post. I put one together for the SD3R club.

OK, that one threw me.

gunrunnerjohn posted:

OK, that one threw me.

I threw off a friend of mine using the "M" designation too on my Facebook post. 

GGG posted:

So just to be clear, nothing but MTH engines on layout with legacy active and MTH engines can not add?  Any other operator see this?  Lots of layouts run legacy and DCs so why just San Diego club.

 

I'm one of the DCS operators having to deal with this problem at SD3R.

"[N]othing but MTH engines on layout with legacy active and MTH engines can not add?"  Correct. And the same problem occurs with Legacy engines on the layout also. If no one is running Legacy the next time I'm operating trains down there, I'll do some experimenting with the issues I've been having with the Legacy transmitter base off and on.

GRJ, it sure does! Do you think any of Dale M's findings would help with this? Or is Adrian already aware of that info? I know you and PLCProf are and I know Dale was boosting the Legacy signal which seems to be somewhat opposite of this?  Just a thought here. It's all over my head...but I try to follow along anyway. 

rtr12 posted:

GRJ, it sure does! Do you think any of Dale M's findings would help with this? Or is Adrian already aware of that info? I know you and PLCProf are and I know Dale was boosting the Legacy signal which seems to be somewhat opposite of this?  Just a thought here. It's all over my head...but I try to follow along anyway. 

I think basically from the above measurements.... it's like:

IF) the Legacy signal is strong enough to run the layout at every spot on the track

AND IF) the DCS voltage at the same spot is greater than 4 times the legacy voltage,

THEN: it runs together seamlessly.

ELSE:  Trouble

Thanks, that even makes sense...used to dabble in Quick Basic many years ago.  The late Dale M did a lot of fiddling with the TMCC/Legacy stuff and I didn't know if that would help you or not. GRJ and PLCprof are much more capable of explaining Dale's research than I am, the reason I asked GRJ. Wasn't even sure it would apply?

Dale didn't look into the interaction between the two command types.

I was just thinking Dale may have found something in his research that could have been of interest here? Also figured if he had that you or PLCprof would have already mentioned it, but threw it out there anyway just in case. 

Adrian,

Thanks again for the DCS module!  I installed it on our layout last night (Monday night).  I didn't have a DCS engine with me to test it out last night, but I called down this morning and they said that they were able to find a DCS engine without a problem with the Legacy signal still turned on.  The difference is that they added the engine on a section of track on the main line near the back door.  The section that we were having problems with were the passing sidings on the Black Line.  We suspect that there still may be "track lamps" under that section that we couldn't get to Sunday night.  Maybe they caused enough problem in that section, combined with the Legacy signal, to block DCS in that section.

I will know more info when they finish their run today.  I'm sure there will be more posts tonight from the guys who ran.

Amazing how some details can shed light on a problem.  So with Legacy off, and MTH engines in these areas that they can not load when Legacy on, what is the Signal Strength with a TS test?  G

Roger L. posted:

Adrian,

Thanks again for the DCS module!  I installed it on our layout last night (Monday night).  I didn't have a DCS engine with me to test it out last night, but I called down this morning and they said that they were able to find a DCS engine without a problem with the Legacy signal still turned on.  The difference is that they added the engine on a section of track on the main line near the back door.  The section that we were having problems with were the passing sidings on the Black Line.  We suspect that there still may be "track lamps" under that section that we couldn't get to Sunday night.  Maybe they caused enough problem in that section, combined with the Legacy signal, to block DCS in that section.

I will know more info when they finish their run today.  I'm sure there will be more posts tonight from the guys who ran.

Hey there!

Wow you got all 4 channels hooked up already? That's pretty fast! I assume you are using the PH180 bricks as your source to the panel?

As long as you're getting rid of the lamps, do make sure you get rid of the wire and sockets that go with them. Any added capacitance to the track can hurt. I would go loop by loop and make sure literally nothing is connected to the track except the TIU. The smaller the electrical load it drives, the better the situation will be.

I'm working on other possibilities of digitally cutting some holes in the legacy signal for the DCS to talk through, but I'll need to test it in my club a bit and see if it causes undesired problems or not. As I said above I'm going to let you guys try it out a bit then come back in mid-june to remeasure everything and see the TIU "m" is holding up properly.

Once the summer starts and I'm not teaching I can come back as often as needed to get your DCS setup working perfectly.

Yes, I took your board home and spent the day (Monday) prepping some wires.  Then I went down at 5:15pm after the museum closed to install the board.  It still took me about five hours to install it because I had to connect a lot of wires.  Also, I had to mount the board with a hinge so that we could get to the electronics behind it. 

The power connected to the board is still switchable between the Z4000 and the bricks, but members of our club pretty much know to use the bricks for DCS (and TMCC).

I may not have a chance to go to the layout until Thursday to see if I can remove the "track lamps".   There is a lot of stuff in the way.

Showed up Tuesday morning (5-29) for my regular run day.  The new board was installed per above.  We cleaned track like normal and began setting up for our running.  Two members brought MTH engines.  One MTH engine has not been in command mode for over 5 years.  We set this engine up (ps2) on the top black line near the rear exit door, and with their passenger cars fully lighted (non led) they pushed the find engine button on the remote,  within 10 seconds the engine was found and loaded into the remote.  They pushed start and it started. 

The second MTH engine (ps3) was not placed on until later in the day on the lower level green line, with the same results.   Only this time a legacy engine was in operation, a conventional engine was operation and the first MTH with all those lighted cars were in operation.  We have never been able to add an MTH engine when other engines were operating.

Just for drill while the both MTH engines above were operating and the legacy engine was in operation, I got out my rail-king evo ps3 placed it on the blue track near mell's diner and got "out of range".  But the remote loaded the other two engines.  I then moved the evo to behind the roundhouse on the blue line and the evo loaded in seconds, and ran all over the layout.  I was not willing to unplug the legacy system while a legacy engine was in operation.

during the day we have removed what we believed are 100 percent of the bulbs and the bulb holders. 

The only other thing we need to do is remove the loose ground wires.

I believe we are 100 percent better with our MTH set up.

Of course we must get the members who have had issues to come and run their MTH engines and report any issues they have.

Members who have issues must report what issues they have and not just state "my engine does not work"...

We need members to report what issues they have like "out of range", "no engine on track", "engine not found" and the location that they have these issues plus any other thing the remote tells you so we can address these issues.  

We are working on a check list to report issues and hope to have this check list available soon.

bigdodgetrain posted:

Showed up Tuesday morning (5-29) for my regular run day.  The new board was installed per above.  We cleaned track like normal and began setting up for our running.  Two members brought MTH engines.  One MTH engine has not been in command mode for over 5 years.  We set this engine up (ps2) on the top black line near the rear exit door, and with their passenger cars fully lighted (non led) they pushed the find engine button on the remote,  within 10 seconds the engine was found and loaded into the remote.  They pushed start and it started. 

The second MTH engine (ps3) was not placed on until later in the day on the lower level green line, with the same results.   Only this time a legacy engine was in operation, a conventional engine was operation and the first MTH with all those lighted cars were in operation.  We have never been able to add an MTH engine when other engines were operating.

Just for drill while the both MTH engines above were operating and the legacy engine was in operation, I got out my rail-king evo ps3 placed it on the blue track near mell's diner and got "out of range".  But the remote loaded the other two engines.  I then moved the evo to behind the roundhouse on the blue line and the evo loaded in seconds, and ran all over the layout.  I was not willing to unplug the legacy system while a legacy engine was in operation.

 

during the day we have removed what we believed are 100 percent of the bulbs and the bulb holders. 

The only other thing we need to do is remove the loose ground wires.

I believe we are 100 percent better with our MTH set up.

Of course we must get the members who have had issues to come and run their MTH engines and report any issues they have.

Members who have issues must report what issues they have and not just state "my engine does not work"...

We need members to report what issues they have like "out of range", "no engine on track", "engine not found" and the location that they have these issues plus any other thing the remote tells you so we can address these issues.  

We are working on a check list to report issues and hope to have this check list available soon.

 

This is good news!

Your club already had a nice antenna in the ceiling but the connector is different (BNC) than the one on the TIU I gave you (SMA) so we couldn't hook it up yet and you're stuck with the short whip antenna I made out of a coax cable. The RF out of range will probably go away if you go back to the good antenna. I will order the SMA-to-BNC adapter and send it to you guys this week. Once you have good DCS in most places, we can use the telemetry train to look at the remaining bad spots and try to figure out why they are bad.

we have several 3 railer members running MTH today Friday June 1st.

hopefully they will give a complete report on their running, sometime in the near future.

it has been a very long time since we could lash up 3 MTH loco's

with this video it appears we have success.

bigdodgetrain posted:

it has been a very long time since we could lash up 3 MTH loco's

with this video it appears we have success.

 

Glad to see things are improving! Once you get the external antenna wired in, things should be quite responsive.

Adrian! posted:

bigdodgetrain posted:

Showed up Tuesday morning (5-29) for my regular run day.  The new board was installed per above.  We cleaned track like normal and began setting up for our running.  Two members brought MTH engines.  One MTH engine has not been in command mode for over 5 years.  We set this engine up (ps2) on the top black line near the rear exit door, and with their passenger cars fully lighted (non led) they pushed the find engine button on the remote,  within 10 seconds the engine was found and loaded into the remote.  They pushed start and it started. 

The second MTH engine (ps3) was not placed on until later in the day on the lower level green line, with the same results.   Only this time a legacy engine was in operation, a conventional engine was operation and the first MTH with all those lighted cars were in operation.  We have never been able to add an MTH engine when other engines were operating.

Just for drill while the both MTH engines above were operating and the legacy engine was in operation, I got out my rail-king evo ps3 placed it on the blue track near mell's diner and got "out of range".  But the remote loaded the other two engines.  I then moved the evo to behind the roundhouse on the blue line and the evo loaded in seconds, and ran all over the layout.  I was not willing to unplug the legacy system while a legacy engine was in operation.

 

during the day we have removed what we believed are 100 percent of the bulbs and the bulb holders. 

The only other thing we need to do is remove the loose ground wires.

I believe we are 100 percent better with our MTH set up.

Of course we must get the members who have had issues to come and run their MTH engines and report any issues they have.

Members who have issues must report what issues they have and not just state "my engine does not work"...

We need members to report what issues they have like "out of range", "no engine on track", "engine not found" and the location that they have these issues plus any other thing the remote tells you so we can address these issues.  

We are working on a check list to report issues and hope to have this check list available soon.

 

This is good news!

Your club already had a nice antenna in the ceiling but the connector is different (BNC) than the one on the TIU I gave you (SMA) so we couldn't hook it up yet and you're stuck with the short whip antenna I made out of a coax cable. The RF out of range will probably go away if you go back to the good antenna. I will order the SMA-to-BNC adapter and send it to you guys this week. Once you have good DCS in most places, we can use the telemetry train to look at the remaining bad spots and try to figure out why they are bad.

Adrian, SD3R

This is a great discussion, which I am hopping will address some of the DCS issues we are having at the NJ-Hi Railers club.  I will be at the club on Wednesday and will take track measurements of the TMCC & DCS signals and will disconnect the TMCC signal and see if it is easier to add/control a DCS engine. 

But in the mean time, I have some questions:

 Adrian –

-Can you provide some details on the DCS module which is mentioned above (is it a device that disconnects the TMCC “U” terminal from the track?). 

-You mentioned a DCS telemetry train, please tell me more about this device.        

 SD3R – Can you tell me more about your external DCS antenna system.

 Thanks,

Bob D

rad400 posted:

Adrian! posted:

bigdodgetrain posted:

Showed up Tuesday morning (5-29) for my regular run day.  The new board was installed per above.  We cleaned track like normal and began setting up for our running.  Two members brought MTH engines.  One MTH engine has not been in command mode for over 5 years.  We set this engine up (ps2) on the top black line near the rear exit door, and with their passenger cars fully lighted (non led) they pushed the find engine button on the remote,  within 10 seconds the engine was found and loaded into the remote.  They pushed start and it started. 

The second MTH engine (ps3) was not placed on until later in the day on the lower level green line, with the same results.   Only this time a legacy engine was in operation, a conventional engine was operation and the first MTH with all those lighted cars were in operation.  We have never been able to add an MTH engine when other engines were operating.

Just for drill while the both MTH engines above were operating and the legacy engine was in operation, I got out my rail-king evo ps3 placed it on the blue track near mell's diner and got "out of range".  But the remote loaded the other two engines.  I then moved the evo to behind the roundhouse on the blue line and the evo loaded in seconds, and ran all over the layout.  I was not willing to unplug the legacy system while a legacy engine was in operation.

 

during the day we have removed what we believed are 100 percent of the bulbs and the bulb holders. 

The only other thing we need to do is remove the loose ground wires.

I believe we are 100 percent better with our MTH set up.

Of course we must get the members who have had issues to come and run their MTH engines and report any issues they have.

Members who have issues must report what issues they have and not just state "my engine does not work"...

We need members to report what issues they have like "out of range", "no engine on track", "engine not found" and the location that they have these issues plus any other thing the remote tells you so we can address these issues.  

We are working on a check list to report issues and hope to have this check list available soon.

 

This is good news!

Your club already had a nice antenna in the ceiling but the connector is different (BNC) than the one on the TIU I gave you (SMA) so we couldn't hook it up yet and you're stuck with the short whip antenna I made out of a coax cable. The RF out of range will probably go away if you go back to the good antenna. I will order the SMA-to-BNC adapter and send it to you guys this week. Once you have good DCS in most places, we can use the telemetry train to look at the remaining bad spots and try to figure out why they are bad.

Adrian, SD3R

This is a great discussion, which I am hopping will address some of the DCS issues we are having at the NJ-Hi Railers club.  I will be at the club on Wednesday and will take track measurements of the TMCC & DCS signals and will disconnect the TMCC signal and see if it is easier to add/control a DCS engine. 

But in the mean time, I have some questions:

 Adrian –

-Can you provide some details on the DCS module which is mentioned above (is it a device that disconnects the TMCC “U” terminal from the track?). 

-You mentioned a DCS telemetry train, please tell me more about this device.        

    

 SD3R – Can you tell me more about your external DCS antenna system.

 Thanks,

Bob D

 

Hey there,

This lionel/DCS module is still being worked out so I haven't decided exactly how it should work yet. Part of me wants to go low tech and simple and just chop windows into the carrier for the DCS to pass through. That seems like it could come with new problems, so having it triggered is probably better. I'm thinking to use a pass transistor so it would adjust the signal voltage of the legacy signal. We could have an "1" voltage and "0" voltage for the two switch states. I think turning it all the way off (disconnecting it) is probably not necessary. I'm prototyping on my ARTY7 FPGA but still waiting on some test PCB layouts I drew up to arrive.

The DCS telemetry train is one of my older projects though it's constantly evolving. In the 5th generation we have now, I have a raspberry pi3 host bolted onto a flat car and one of those 4th order RC filters going to the pickups. I have a home-made scope (fast 12b A/D chip and a FIFO) triggering off the DCS waveforms, and then reporting the DCS excursion voltage to an LCD screen glued on top. Basically like an oscilloscope built into a train with a DCS matched filter already inside.

The SD3R basically did what our AGHR club did too. You lift the sad antenna (stick of wire) off the RF board, solder a coax to the same terminal (and solder the shield to the ground on the RF board). Then you goto a bulkhead (they use BNC we use SMA) and just buy a 900 MHz antenna on Amazon. I also did the same thing on my remote. It makes a big difference in terms of reception. Some discussion of the RF engineering basis is here. 

What kind of DCS issues are you having?

Adrian

The Lionel/DCS module you mentioned above, is that what you left behind with the SD3R club last week or was it something else?

Would you mind sharing the drawings on the DCS telemetry train.  I would like to build one for the NJ-Hi Railers.  It should make it easier to diagnose DCS signal issues on our 30'X200' layout.  Currently we are starting  to do DCS signal testing with a digital scope the club just picked up, but to have a scope on train wheels, that would make life a lot easier. 

DCS  issues:  At times, hard to load an engine into the hand held/tablet or to start up a DCS engine. We get one of those engine not found messages.  I will do a test on Wednesday to see if engines load easier when TMCC is turned off.  Last I checked the TMCC signal it was in the 5vpp range.  We also lose control of engines at various parts of the layout.

Thanks for your help!

Bob D

this is the module Adrian made for us.  The Angles gate Hi-railers use several. it is only a dcs system the legacy is connected elsewhere.

rtr12 posted:

GRJ, it sure does! Do you think any of Dale M's findings would help with this? Or is Adrian already aware of that info? I know you and PLCProf are and I know Dale was boosting the Legacy signal which seems to be somewhat opposite of this?  Just a thought here. It's all over my head...but I try to follow along anyway. 

Adrian,

just a lay person following along, but I noticed a difference in your tone generator frequency and what Dale Manquen found to be good Legacy/TMCC signal. That was 455Khz with a 5 volt excursion.

rad400 posted:

Adrian

The Lionel/DCS module you mentioned above, is that what you left behind with the SD3R club last week or was it something else?

Would you mind sharing the drawings on the DCS telemetry train.  I would like to build one for the NJ-Hi Railers.  It should make it easier to diagnose DCS signal issues on our 30'X200' layout.  Currently we are starting  to do DCS signal testing with a digital scope the club just picked up, but to have a scope on train wheels, that would make life a lot easier. 

DCS  issues:  At times, hard to load an engine into the hand held/tablet or to start up a DCS engine. We get one of those engine not found messages.  I will do a test on Wednesday to see if engines load easier when TMCC is turned off.  Last I checked the TMCC signal it was in the 5vpp range.  We also lose control of engines at various parts of the layout.

Thanks for your help!

Bob D

 

What I left in their club is a panel with the PSX breakers, chokes, and a rev "M" TIU with my diode boards inside.  The lionel/DCS switching thing isn't even finished being designed yet! (I'm thinking maybe getting into it this weekend).

I can share the telemetry train details (maybe on a separate post), but the thing is there's a lot of C++ code that drives it. I'll try to put up a thread about it this week.

Your DCS issues are typical of poor correlation (low DCS excursion or low effective SNR) similar to everyone else. A 5V carrier underneath sounds like too much based on the measurements above. I'm thinking of doing a video tutorial on some of these effects at some point...

Moonman posted:

just a lay person following along, but I noticed a difference in your tone generator frequency and what Dale Manquen found to be good Legacy/TMCC signal. That was 455Khz with a 5 volt excursion.

That would be before the addition of the TMCC Buffer.