Suzie

It can be a serious problem - I saw one very large chap lean all his weight up to his elbow on to a large 0-gauge layout supported on slightly fewer trestles than it should have had which dipped about six inches under his weight. I think it survived but there could have been a very nasty mess.

Tune in app works, but stops working when you are out of cellular telephone range.

It would be nice if that radio app was there...

That is a controller setting and not a decoder setting.

I just got a secondhand Tesla X and it seems like quite a nice car, but it is looking like a radio is an optional extra on modern cars as I cannot find the radio. Satnav appears to use Google which I guess is O.K. although I would rather have Waze. It has taken a fortnight to get access to the glovebox which was protected by a PIN number - which could only be cleared by sending a photo of the V5 to Tesla, and DVLA do not send the V5 out as quickly as they might. It has been quite a learning experience. I am especially not getting used the doors and everything being motorized - even the charging socket cover has a motor to open it. I was quite impressed that the granny charger it comes with is 3-phase - so it will charge at full 11KW AC rate if I find a suitable red commando socket to plug in to! Looks like lots of room to cart bits of layout around which is the important thing of course.

You need to look at the original post - it asks a general question. To which you can only give a general answer. I gave up with sub £25 decoders for general use some time ago because they all had shortcomings that made the cost saving not worthwhile most of the time (along with a lot of the more expensive decoders). If £25 is too expensive, then people will need to say what compromises they are prepared to make (poor motor control, unreliable packet decoding, noisy motor drive, complicated motor setup, insufficient function outputs, wrong size, inadequate power output, etc.) I am sure that there are some cheap decoders that work well in some locos, but there is no general cheap decoder that is universally good. I have tried a lot of decoders, and there is only one brand that makes decoders that can just be dropped in to any loco which will work without any configuration, and they do that for around £25.

The cheapest is most unlikely to be good enough. Plan to spend £24 for a direct plug-in decoder, or £25 for one with wires and get a Zimo MX600 (wires/wired plug), MX637(PluX22) MX638(21MTC) or MX618 (Next-18).

Also how fat is the wire in the 25-way cable? You can solder 16/02 wire quite easily in to a 25-way 'D' connector and a couple of these should suffice to carry the DCC a short distance, but 25-way cable is only likely to have 7/02 wires in it at best (and most are thinner than this) so you will really need four per bus wire if the 25-way cables are any length.

The DS64 will operate four signals but you will have to set it up to operate 'slow motion machines'. Manual for the DS64 is here:- https://www.digitrax.com/media/apps/products/stationary-decoders/ds64/documents/DS64_flattened.pdf

Yup! That is the problem. Electrofrogs are so much easier with DCC, it is no surprise that the only trouble you are having is with the Insulfrogs.

I think you will find the problem is with the Insulfrog point and not the Electrofrog one. Try putting insulated joiners on the Insulfrog point frog rails and see if the problem goes away. Insulfrog points specifically are not suitable for use with DCC because the frog rails are too close together and can be bridged by the tyre of the wheel - other dead frog points don't have this problem.

What is the resistance of the coils on the failed relay? they should be between 864 Ohms and 1056 Ohms and if they are you probably have not damaged the coils. If the coil is damaged it is probably down to not using an inverse parallel diode to divert the back EMF from the SEEP motor (1N4148 diodes are ideal for this and very cheap).

I suspect that you may have a faulty relay here. Those you have are good for up to 27V so you cannot cook them from the 24V output of your CDU. The most likely thing that has happened is too much current through the contacts welding them together due to a shorted load when testing. Sometimes a good beating with the wrong end of a big screwdriver can free the contacts but it might be permanently dead. Carefully check that you have wired the contacts correctly, they are not always connected intuitively. I would not give up yet - plenty of people have achieved excellent reliability so stick with it.

There is a lot of wire and choc-blocks in line with the DCC feed from the ECoS to the track. I think that simplifying it might help a bit - much as it goes against the grain for making fault-finding easy.

Just to add a few notes:- Hongfa relays are fine for this. I have some twin coil latching HFD3 relays (slightly smaller than the HFD2 types but adequate for the task). They will still switch 4A (see point 2) and carry quite a bit more current. Use both sets of contacts in parallel to increase the amount of current that the frog switching contacts can handle. This will improve reliability over the long term especially if people are prone to run the point the wrong way and short on the frog. Because of the extremely low resistance of the SEEP motors the voltage at the motor is often very low and may be insufficient to fire a 12V relay if mounted near the motor even when fed by a CDU. It is best to mount the relays near the control panel to ensure that sufficient voltage is available. I understand this can be inconvenient on a DC layout, so if you don't want the extra wiring you might need to select the voltage of the relay coils carefully to ensure reliable operation. I have seen installations where even when using a CDU giving out 24V (the typical output of a CDU fed with 16V AC) there was less than 6V available at the point motor due to resistance in the switches and wiring. Some CDUs will run at up to 35V - it can sometimes be helpful to use a regulated 35V DC supply to get the most out of a CDU if your wiring is not the fattest cable. Consider using a cheap adjustable DC-DC convertor running from a 12V supply to give you a cheap source of 35V to the CDU that can be adjusted lower if required. While the inverse parallel diodes across the coils are only essential when using semiconductors to switch the points - they will reduce arcing across the switches and extend their life. There will be far more back EMF (in terms of power) from the point motors than the relays so even if you don't have relays you should use inverse parallel diodes (no one will tell you this!). You say that you don't want to change your motors in which case you will have to live with the shortcomings of what you have - but in future consider using the solenoid motors with higher resistance coils (Peco PL10E or PL10EL for example) because the lower currents involved result in much higher voltage being available near the points for powering latching relays and you don't need to worry so much about super-fat wiring to the solenoid coils, CDU and switches. Using latching relays is super-reliable if you can implement them correctly. I got my good results by beefing up the wiring (especially from the CDU to the control panel), upping the CDU output voltage to 35V, and mounting 24V latching relays near to the control panel. The Hongfa relays will typically fire at about 75% of the rated voltage and will take 150% of the rated voltage so you do have a nice big window to play with (for example a 24V relay will fire at 18V and will be OK at 36V so will not be damaged if a solenoid goes open-circuit and will take a bit of loss in the wiring). I have probably raised more questions than I have answered but what you are trying to do is not as simple electronically as it first appears.