TI track circuits

[PJW]

Hi Peter,

could you tell me  how TCs is tested? and what could be the load for TCs for testing?

Secondly, as TCs is directly connected to track, with what value of Inductor and Resistor to simulate the track for testing purpose.

---

I  am not sure quite how to interpret your questions; are these specifically asked for TI21 track circuits or more generally?

I am also not sure what level of existing knowledge you have, so I am afraid this post will be more asking you to clarify your question than give an answer! 


However I shall give some information that might help us refine your request.


TI21 track circuits are audio frequency track circuits, 

1. primarily designed to be used in jointless configuration using a pair of tuning units separated by a length of rail to form a tuned zone to separate into different sections, therefore not needing Insulated Block Joints.  Peter has described this in more detail above,
   
2. they can however be used in jointed configuration with IBJs more in the way of a dc track.
   

Testing is obviously a little different in each case.

I'll assume for now you are familiar with testing a dc track.  

Case 2 is quite similar to dc in some ways except that

• there is more equipment to be checked to be of the correct type, installed in accordance with specification etc,.
  
• voltages have to be taken at the transmitter end PSU, the Tx unit itself, the connections to the ETU (End Tuning Unit) as well as on the rails.  The voltages taken on the ETU and rails etc. must be taken with a frequency selective meter (aka blue  handbag) as need to differentiate "this" track circuit voltage from that of the "adjacent track" (though of course if the IBJ and other insulations are ok then there shouldn't actually be any other voltage)
  
• voltages also have to be taken at similar placed at the Receiver end
  
• unlike the dc track where any adjustment is made at the feed end, for the TI21 the track is adjusted by varying the gain setting at the Rx end. [Actually where the track is short there is a configuration change to Low Power by changing the connections on the ETU, but I'll ignore this].  For the traditional TI21 the gain is changed by altering strappings for 3 different coils of a transformer, altering the turns ratio by selecting a gain of 1, 2,3,.....13.  The modern form of the track Ebitrack200, is conceptually similar but the Rx is digital and it sets its own gain internally by being given two scenarios: track clear and a test shunt applied (1 ohm for Normal Power, 1.5 ohm for Low Power)
  
• the track is current operated and so the tester needs to measure the voltage across a 1 ohm resistor in the input circuit, so as to get the numeric value of the current without interrupting the circuit; this has to be compared to the length of the track and gain setting to confirm that track operating in accordance with expectations.   For the Ebitrack, the current a can simply be read off the Rx unit display
  
• broadly the rest of the process is analogous to the dc case, except that there is practically no hysteresis between the drop shunt and pick shunt and the TI21 is inherently slow to pick (2 seconds) so the shunt box must be operated far more slowly or a false result will be obtained
  
  

Case 1 needs more than case 2, because the rejection ration of the tuned zone must be checked.  

• Adjacent track circuits use different frequencies in a frequency pair A/B, C/D, E/F or G/H; let us assume a boundary between a "A" track and a "B" track.  
  
• The voltages on the rails at the "A" TU are taken both for frequency "A" and also for frequency "B"; similarly the voltages on the rails at the "B" TU are taken both for frequency "A" and also for frequency "B",
  
• The rejection ratio for each frequency must be calculated; this involves dividing the voltage present at its own TU by the same frequency at the other TU,  This proves how good the zone is at preventing one track's voltage leaking through into the next track of the opposite frequency; the achieved figures must be checked to meet the relevant standard.
  
• When testing the limits of the track, this is more involved in a tuned zone as the drop shunt in this area will be less than within the track itself, and it is important to check that there is an are in the middle where both tracks drop but the further track is not dropped by a dead short at its neighbouring TU.
  

The above is deliberately a bit simplified (and from memory- I don't think I have tested one myself for some 25 years!) but gives the most salient differences.

I think that covers the first part of your first question; the second part may also have been covered to some extent, depending on precisely what you meant.

• Generally one would set up a TI21 track to drop at about 1 ohm; it is then likely to pick at 1.1ohm.  
  
• However when testing to limits, the test is a 0.5ohm shunt.  In truth in a tunes zone between two Tx (not really a preferred configuration as would generally use an ETU as a centre-fed configuration of two track sections of the same frequency), then one is lucky to achieve 0.3 ohm throughout and in extremis if we could get the tracks to drop with a galvanised bond wire in the centre of such a TU we accepted it (though I suspect nowadays that the justification of this non-compliance would be very difficult- doubt whether the physics has changed, so I would bet that a "blind eye" is still turned at local level to avoid escalating as a problem!)
  

The second question is the one I don't really understand at all.
The rails are continuous through the tuned zone so yes electrically the two track circuits are separated by an equivalent circuit which has both resistance an inductance.  The dc resistance of 20m of rail is so low as to be negligible (however if it is jointed track with a fishplate joint bonded over by a couple of galvanised bonds then it isn't completely negligible); however the impedance presented at audio frequency is significant because of the inductance. I do not know numerically what it is; I suppose I could reverse engineer it from typical voltage readings, but really that is the concern of the product designer 30+  years ago; it is what it is and it is that value which dictates the TUs should be 20m apart on the typical 113 rail section.
There is no simulation involved in real applications; I guess that some training schools which don't have the luxury of the space etc to have a 30m length of railway line may well have got some simulated circuit of resistor and inductor to get sensible values for trainee's experience, but I do not know what those values would be.

Have I understood what you were actually asking?