H2O Posted July 5, 2020 Share Posted July 5, 2020 7 hours ago, NinOz said: I use a small adaptor based on a circuit published around 1995 on NCE yahoo groups, to measure the voltage using a DC meter, analogue or digital, or a DMM. I use a 20V DC analogue meter. Works a treat but a digital readout would be better for visibility at a distance. Basically stores averaged near peak voltage in a capacitor. Consists of two transistors, three capacitors and three resistors. Claimed to be accurate to within 100mV, probably due to the necessary charge and bleed resistors plus transistor biases. Should cost about 5 to 10 quid to make. Pity the topic wandered in the realms of expensive equipment and daft views of DCC/AC wave forms rather than just address the original simple question and application. Excellent suggestion. For an even simpler option that should give a consistant error of 0.7V too low on the dc voltage setting is a simple peak detector. This consists of a diode in series with a non-polarised capacitor - measure the voltage across the capacitor. Cost from about 10p + vero board and some leads. Although not super accurate it should be consistent, and much better than the ac setting of a typical home digital multi meter (dmm). An analogue multi-meter will give a lower voltage reading due to its lower input resistance. Link to post Share on other sites More sharing options...
RMweb Premium melmerby Posted July 5, 2020 RMweb Premium Share Posted July 5, 2020 14 minutes ago, H2O said: Excellent suggestion. For an even simpler option that should give a consistant error of 0.7V too low on the dc voltage setting is a simple peak detector. This consists of a diode in series with a non-polarised capacitor - measure the voltage across the capacitor. Cost from about 10p + vero board and some leads. Although not super accurate it should be consistent, and much better than the ac setting of a typical home digital multi meter (dmm). An analogue multi-meter will give a lower voltage reading due to its lower input resistance. Even better just a bridge rectifier without a capacitor feeding a DC voltmeter. It will read about 90% of the true RMS value. 4x 1N4148 diodes to make the bridge. Link to post Share on other sites More sharing options...
Nigelcliffe Posted July 5, 2020 Share Posted July 5, 2020 (edited) 23 minutes ago, melmerby said: Even better just a bridge rectifier without a capacitor feeding a DC voltmeter. It will read about 90% of the true RMS value. 4x 1N4148 diodes to make the bridge. Fit the rectifier diodes on a bit of PCB offcut, add a cheap three digit DC voltmeter display (numerous on sale for under £2), a couple of bent brass tabs (etched kit offcuts) at the end to push onto rails, and there's an under £5 track voltage tester. Addition of a couple of LEDs (with resistors) before the rectifier will mean it can distinguish DC from DCC. On DC, one LED will light, indicating direction of voltage (which rail positive). On DCC, both light. Nigel Edited July 5, 2020 by Nigelcliffe 2 1 Link to post Share on other sites More sharing options...
RMweb Premium kevinlms Posted July 5, 2020 RMweb Premium Share Posted July 5, 2020 15 hours ago, Dagworth said: No, that's the DC vs. DCC argument. Not quite what I had in mind, but still good. Link to post Share on other sites More sharing options...
Crosland Posted July 5, 2020 Share Posted July 5, 2020 9 hours ago, NinOz said: stores averaged near peak voltage in a capacitor. ??? DCC is a square wave so no need to average or detect peaks or do any RMS calculation. As already said, a simple bridge rectifier is more than adequate. 1 Link to post Share on other sites More sharing options...
H2O Posted July 5, 2020 Share Posted July 5, 2020 4 hours ago, melmerby said: Even better just a bridge rectifier without a capacitor feeding a DC voltmeter. It will read about 90% of the true RMS value. 4x 1N4148 diodes to make the bridge. A full bridge rectifier will have the disadvantage of an error of approximately 1.4V. This is 90% at a track voltage of 14V but may be easier to think of as a drop of 1.4V. If a full bridge rectifier is used a capacitor will still be useful as without it the voltage measured after rectification will fluctuate a little, especially if the system is under load (which is when some faults show up). The ideal dcc voltage will switch cleanly, the actual signal will not due to the layout being a complex load. A full bridge rectifier will have the advantage of it not being important which way round the circuit is connected to the track. This can be solved if using a single diode by simply switchIng the dmm leads around. For accuracy the transistor circuit above may be best but most expensive, followed by the single diode and capacitor solution which is cheap and then the full bridge circuit which is more expensive and less accurate (due to the 1.4V drop). As with everything, each solution has its own advantages. Take your pick... Link to post Share on other sites More sharing options...
NinOz Posted July 6, 2020 Share Posted July 6, 2020 Again we stand to go off into the realms of nit-picking. What are we trying to achieve with measuring the voltage, what was the OP trying to do? Just trouble shooting wiring problems or trying to fix booster/supply circuit problems, DCC signal quality? I assume the former for the OP. My NCE adaptor just monitors voltage at the booster outputs, really not doing much except noting shorts and indicating all is fine at the booster. For checking wiring quality I use my little DMM on AC setting usually. I don't care what the voltage or what the measurement bias really is as I am looking for any significant resistances in the supply to that point. Measure the voltage unloaded, drop a load on that point and measure the voltage, look at the voltage sag, if significant trace any anomalous results, fix. Doesn't matter if the bias is 0.1V, 0.7V, 1.4V or x.xxV as long as it is consistent. "??? DCC is a square wave so no need to average or detect peaks or do any RMS calculation. As already said, a simple bridge rectifier is more than adequate." The circuit I use averages due to balance between built in rate of charge and discharge of the capacitor on each +/- pulse. Nothing in my post demands that averaging, RMS or any such is required. It is just how the circuit works. Some people could start an argument about anything even if not said or implied. 2 Link to post Share on other sites More sharing options...
RMweb Gold Dagworth Posted July 7, 2020 RMweb Gold Share Posted July 7, 2020 On 06/07/2020 at 02:01, NinOz said: Some people could start an argument about anything even if not said or implied I hope the poster sees the irony of his somewhat argumentative post? Andi 2 1 Link to post Share on other sites More sharing options...
Junctionmad Posted July 8, 2020 Share Posted July 8, 2020 On 27/06/2020 at 23:50, melmerby said: My scope does this: Nice. Built in DCC decoder , what scope is it Link to post Share on other sites More sharing options...
Junctionmad Posted July 8, 2020 Share Posted July 8, 2020 (edited) By the way mention was made of using scopes, unless you have , an isolated scope, battery powered scope , differential probes, or setup to use two inputs in differential mode , don’t connect a standard scope probe across a DCC track , as the scope probe gnd is at mains ground and will short out the DCC track. Edited July 8, 2020 by Junctionmad 1 Link to post Share on other sites More sharing options...
RMweb Gold SHMD Posted July 8, 2020 RMweb Gold Share Posted July 8, 2020 47 minutes ago, Junctionmad said: Nice. Built in DCC decoder , what scope is it I asked that and Melmerby answered it in the second last post at the bottom of page 4. It's a PicoScope: https://www.picotech.com/products/oscilloscope It's worth going and having a look at that post. Kev. Link to post Share on other sites More sharing options...
RMweb Premium melmerby Posted July 8, 2020 RMweb Premium Share Posted July 8, 2020 3 hours ago, Junctionmad said: By the way mention was made of using scopes, unless you have , an isolated scope, battery powered scope , differential probes, or setup to use two inputs in differential mode , don’t connect a standard scope probe across a DCC track , as the scope probe gnd is at mains ground and will short out the DCC track. I use my Picoscope with the laptop. It's USB powered and current consumption is quite low. 1 Link to post Share on other sites More sharing options...
RAF96 Posted July 9, 2020 Share Posted July 9, 2020 On 08/07/2020 at 13:52, Junctionmad said: By the way mention was made of using scopes, unless you have , an isolated scope, battery powered scope , differential probes, or setup to use two inputs in differential mode , don’t connect a standard scope probe across a DCC track , as the scope probe gnd is at mains ground and will short out the DCC track. My standard method is probe to one rail and ground clip to the other. I have had all four probes on my Picoscope connected to the Elite at the same time... Main Track probe to left rail and ground to right rail Prog to Prog track ditto Boost output ditto but see note below. Xpressnet via a breakout cable. ...what you have to watch with a USB scope is connecting multiple devices via USB as the cable screen-ground may not be at equal potential on all devices leading to a fault. E.g. controller and scope and other external device connected to the same PC by USB. Note - I did find a crossover fault on the Elite motherboard which a user would never notice in that the Boost output logic was contra the main track, but as the Hornby Booster module is also a reversing loop module in practice this opposite polarity would be switched at the track as soon as detected. Be aware if trying this at home as it killed the Elite Boost and Prog circuit. Safer to probe at the track of each power district rather than the controller outputs. Link to post Share on other sites More sharing options...
WIMorrison Posted July 9, 2020 Share Posted July 9, 2020 Could you use the usb to usb opto-isolators that you can get for a few quid? the certainly allow me to connect multiple USB devices at the same time and when they are not used I find the devices tripping out because of ground conflicts Link to post Share on other sites More sharing options...
RMweb Premium melmerby Posted July 9, 2020 RMweb Premium Share Posted July 9, 2020 I've just checked the Z21* and the Lenz 100 set and neither has any connection between mains ground and track output. There also no connection between the case ground and track output on the Lenz. *In fact the Z21 floats about 40v AC above ground, the Lenz is only 2.5v above ground. Link to post Share on other sites More sharing options...
RMweb Gold Dagworth Posted July 9, 2020 RMweb Gold Share Posted July 9, 2020 (edited) 2 hours ago, melmerby said: I've just checked the Z21* and the Lenz 100 set and neither has any connection between mains ground and track output. There also no connection between the case ground and track output on the Lenz. *In fact the Z21 floats about 40v AC above ground, the Lenz is only 2.5v above ground. Odd, I’m pretty sure I’ve experienced my soldering iron tip short out my Lenz set to ground if I try to solder on live track. That said, there cannot be any connection to mains ground on my Lenz set, the transformer is double insulated. Now wondering if there’s a short somewhere between a running rail and one of the DC lines of the power bus around the layout somewhere? Andi Edited July 9, 2020 by Dagworth Link to post Share on other sites More sharing options...
RMweb Premium melmerby Posted July 9, 2020 RMweb Premium Share Posted July 9, 2020 (edited) Same here, the transformer I use is a quality 80VA toroid. The only earth on my layout is the screen of the cable I have used for the RS bus but only connected to mains earth, not the Lenz case EDIT The Z21 uses a laptop type SMPS. Edited July 9, 2020 by melmerby Link to post Share on other sites More sharing options...
RMweb Gold SHMD Posted July 9, 2020 RMweb Gold Share Posted July 9, 2020 The answer is in the insulation. (I note the use of Doubly Insulted "Square within a Square" symbols.) I possess 3 insulation meters (0 to 500v, 1000v, and a 2500v range models), "Megger's" if you will, and am also responsible for the department's shared two other "Megger's" of 0 to 5000 ( programmable) and the 30kV (selectable range). Most things have a resistance - and therefore a leakage current. The PSU outputs will find (float) to their equilibrium voltage - depending upon the leakage current through the insulation, usually mains/2. There is usually enough stored energy (joules) lying around to spark/trip when a differing (floating) potential is initially joined to that. Big old TV chassis used to be connected to the center-tap of '0-120 to 0-120' transformers. You didn't want to ground your scope probe on those! In my day job, I'm not happy unless i'm reading in the Giga-Ohm range across isolated floating potentials - at the required working voltages! Kev. Link to post Share on other sites More sharing options...
Suzie Posted July 9, 2020 Share Posted July 9, 2020 Was the USB interface connected by any chance? This I suspect will connect both the 'T' and 'M' terminals of the Lenz (they are electrically the same) to the 0V of the USB cable, and thence via the computer to mains earth. Link to post Share on other sites More sharing options...
RMweb Gold Dagworth Posted July 9, 2020 RMweb Gold Share Posted July 9, 2020 1 hour ago, Suzie said: Was the USB interface connected by any chance? This I suspect will connect both the 'T' and 'M' terminals of the Lenz (they are electrically the same) to the 0V of the USB cable, and thence via the computer to mains earth. No, but that has just made me realise that a lot of my Xpressnet connections use the layout 12vDC power bus rather than running four wires all the way around. Only the A and B wires of the Xpressnet are run around the layout independently. That may well be the source. Maybe time to redo bits of the Xpressnet, how many handsets can the LZV100 power at any given time? Andi Link to post Share on other sites More sharing options...
Suzie Posted July 9, 2020 Share Posted July 9, 2020 I think that the Lenz can provide 300mA on the 'L' wire. Overload it and the Lenz will crash (this is what happens when you plug a Multimaus in sometimes due to the inrush). Probably best to run the whole layout Expressnet from a separate 1A 12V supply - use just the 'M', 'A' and 'B' wires from the Lenz and feed the 'L' and 'M' on the bus with 12V from an independent 12V supply. You will still have the 5-pin DIN socket on the front of the Lenz in case of emergency, or you can use that for the interface like you have done. Not sure how many handsets can be powered from 300mA. 1 Link to post Share on other sites More sharing options...
Nigelcliffe Posted July 10, 2020 Share Posted July 10, 2020 Not sure how many Lenz handsets, but at least five (mix of LH100 and LH101) from my experience. Link to post Share on other sites More sharing options...
RMweb Premium melmerby Posted July 10, 2020 RMweb Premium Share Posted July 10, 2020 10 hours ago, SHMD said: Big old TV chassis used to be connected to the center-tap of '0-120 to 0-120' transformers. You didn't want to ground your scope probe on those! Kev. Big old TV chassis used to have the chassis conected to mains neutral (if you had 3 pin mains plugs). With 2 pin plugs it was a 50/50 chance of live or neutral! Even some radio sets adopted that approach. My parents had a small valve table radio and at first glance it looked like it was isolated as it had a transformer but that only fed the 6.3v valve heaters, the 250v DC supply came straight from the mains via a valve rectifier. 1 Link to post Share on other sites More sharing options...
RAF96 Posted July 10, 2020 Share Posted July 10, 2020 (edited) Dagworth said Quote Odd, I’m pretty sure I’ve experienced my soldering iron tip short out my Lenz set to ground if I try to solder on live track. It is always worth checking your soldering iron tip for continuity. Some cheaper ones are lethal to any electronics. Edited July 10, 2020 by RAF96 1 Link to post Share on other sites More sharing options...
RMweb Gold Donw Posted July 12, 2020 RMweb Gold Share Posted July 12, 2020 My Hakko soldering station quotes tip to ground as <2 ohms and <2mV. Mind you with HV across the end of the garden to a transformer and the elctrical earth comes from there I would imagine there could be more than 2mV between the ground there and at the house. Interestingly the spec states the plastics are not insulators but conductors all in the interest of avoiding static problems. Don Link to post Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now