 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : SMALL UPS...
  Page 3 of 5   |
|||||
| Author | Message | ||||
phil99 Guru  Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Little of that surge is due to the SMPSU starting. Most of it is just charging the main filter cap. The higher supply impedance just increases the charge time, which is a good thing. Less stress on the rectifier and filter cap. After the start voltage is reached the SMPSU runs in current limiting mode till the output caps are charged and voltage regulation takes over. |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6783 |
Thinking about it, many SMPS circuits have a PTC thermistor (sometimes with a bypass device) in series with the mains to limit the starting surge anyway. Whether the series resistance is in that or is in the wire feeding it probably doesn't matter that much. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
Grogster Admin Group  Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
I wonder then, why mains inverters of old, would flatly refuse to start anything with a SMPSU in it? Perhaps it was the square-wave output? Or modified square-wave output? Pure sine-wave inverters are quite affordable now, and they don't seem to have any issues. Perhaps it is a combination of inverter development, and SMPSU design as we march merrily forward with technology!  Smoke makes things work. When the smoke gets out, it stops! |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
The main filter cap is the problem for a square-wave inverter (often called modified sinewave if it has PWM voltage regulation). A vertical rise in voltage at every half-cycle produces a massive current spike as it tries to instantly charge the cap to 300Vpeak. It can also happen with a sinewave inverter if the SMPS is switched on close to the AC peak. It happens less now as newer devices often have some form of surge limiting as noted in Mick's post above. Edited 2024-06-09 09:54 by phil99 |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
I've been playing with this a bit today, and we might be dead in the water.  As soon as I simulate the phone-line resistance IN ANY WAY, the RX end of the link refuses to then start the VCR. I'm down to LESS then 100-ohms of inserted series resistance at this point(82-ohms, 1W carbon), and the VCR never starts. This resistor is ONLY to see if I get start-up at the other end. I'm not leaving it there, even if the VCR DID start, as I would cook the 1W resistor pretty fast I think!  You can HEAR the VCR TRYING to start, but it never starts, so this might have killed this idea on the bench before I even try to test it on-site. Does anyone have any ideas of other things I could perhaps try on this idea, before I declare it a no-go? Basically, the phone-pair DC resistance is going to be WAY more then 100-ohms, even on the short runs. Phones don't care about that, but this concept looks like it will. I wonder if some caps placed on either side of the RX transformer could help to just get things started? Seemed like such a nice idea, and perfect use for the old mostly-redundant grease-filled telephone cables! Oh well....  I might still be able to use these cables and the LV AC concept, but I would probably have to hack the amplifiers so I can tap into the LV side of the SMPSU inside them - perhaps also just outright disconnecting the internal SMPSU and having a local LV regulator fed from the LC AC on the phone cables. That'll be my next experiment. Edited 2024-06-16 10:14 by Grogster Smoke makes things work. When the smoke gets out, it stops! |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
How long is the line? 80Ω = 800m for 2 pair in||. Most VCRs power the motors briefly at startup so you are comparing apples with oranges. A better test may be a 12V 1A plug-pack with a 15Ω load. Better still, take your test rig to the site and put it between the power point and launch amp. Start with 10Ω in the 30VAC line and work your way up till it won't start. Edit. Next step is use 48V transformers. Both the current will be lower (30/48 times) and the tolerance for voltage drop will be higher (each volt drop is a smaller % of 48 than of 30). The losses are reduced to 30/48 squared = 0.390625 times those at 30V. Edited 2024-06-16 11:16 by phil99 |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
Sounds like good advise, I will proceed along the lines you suggest. I would love to use 48v transformers, but they are unavailable over here in Kiwiland. I'd have to get some custom made especially, and that is WAY expensive. If you have any suppliers I should be looking at, please do link to them here. Edited 2024-06-16 15:35 by Grogster Smoke makes things work. When the smoke gets out, it stops! |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
I've done more bench-testing, and the VCR can ALMOST start with 22-ohms series resistance. The LCD display lights up, but then the SMPSU fails, and it switches off. It then cycles, but never is able to start. Using 15-ohms, it DOES start, but the 1W carbon resistor just about goes into nuclear meltdown!  But that is only cos a 1W carbon resistor, is simply not grunty enough for what I am subjecting it to!  On 18-ohms, the VCR cycles and refuses to start. Smoke makes things work. When the smoke gets out, it stops! |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
I see the problem. Looked up Jaycar NZ and their range has shrunk. Altronics has some toroids with 2 x 25V secondaries that would do. The other option would be 2 x 24V at each end with the secondaries in series. A bit messy and more expensive. Worth persisting with the 30V ones until you are certain the launch amps won't work with the measured line resistance. Perhaps check how many pairs are are available to parallel up. I have simply been assuming 2 but there may be more. Also if the distances are long enough they may have used Exchange cables which are 0.4 sqmm, so half the resistance per pair. |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6783 |
For 48V transformers use two 24V ones. Put the 24V windings in series and the mains windings in parallel. It's preferable if they are the same type and rating, but no problems if they aren't as the rating is the same as the smallest one. Yet another possibility are 100V line transformers, intended for PA speakers. These have multiple low voltage taps to play with. Edited 2024-06-16 18:06 by Mixtel90 Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
The cables are fully-underground burial rated cables with the very hard black-plastic sheath, and the pairs inside are all packed with grease to repel moisture in service. The trunk cables that run around the outside of the village are a combination of either 100-pair or 50-pair, with 25-pair and 7-pair grease-filled cables of the same type, routing lines off to the smaller areas. But the point I am trying to make here, is that there will be HEAPS of spare pairs - paralleling them up to reduce the line resistance, won't be an issue, as around 90% of these cables are now redundant, as all the residents have been ported over to either fiber or wireless connections, as Chorus(the company that maintains the communications network across NZ) want to decommission the entire copper network, and I think the deadline for that is next month, actually. In other words, those still on copper who have NOT made the move by next month, will loose their phone connection when they finally switch off the entire copper network. But that leaves the village with a lovely network of then-redundant copper cables that I am planning and hoping to use - one way or the other - for the power to the amps.  Smoke makes things work. When the smoke gets out, it stops! |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4223 |
Grogster, Using the phone lines for power of TV distribution means maintaining the ancient copper network (or have someone maintain it). It is not only the installation cost, also a maintenance fee. Are you planning to install power supplies in the old telephone exchange building ? Is the Telephone company willing to leave the copper cabling to you, or do they plan to sell it (copper is expensive now). I hear technial challenges, but also political and strategic. Volhout PicomiteVGA PETSCII ROBOTS |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6783 |
Of course, they won't allow any connections to those wires while they are still connected to any external network (which doesn't have to be telephones. In the UK pairs are sometimes rented out to private companies for telemetry stuff, I believe). Once they have no use for them whatsoever they will probably disconnect them with a hacksaw as the connectors might be rescued - they won't be bothered to insulate the ends. It's even possible, given the current cost of copper, that they may take up any major lengths for recycling. Of course, if the cables were that horrible copper plated aluminium they wouldn't bother. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
No, no, no..... Volhout, you are mis-interpreting. The village is classified as a "Private Network" here in NZ. Basically, the Chorus network STOPS at the village demarcation point - the point where Chorus(the network company) drops off the phone connections. Anything PAST that point, is private network, and 100% owned by the village. We used the same cables that Chorus would have used, for the village infrastructure, but the village is a TOTALLY unique and fully isolated part of the old copper network. We can do ANYTHING we like with those old cables, as the village OWNS them, and those cable runs have NOTHING to do with the official network. Once isolated at the village demarc, those cables are totally and utterly under the sole control of the village, and have ZERO to do with ANY telecommunications company. No need to access any exchange, those underground cables belong to the village, and therefore, the village can do whatever it likes with them - once they no longer have any telephone services running over them.  Smoke makes things work. When the smoke gets out, it stops! |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6783 |
Oh, that's an interesting system! I've not come across anything like that before. Useful! :) To get more W over the network you could run LV 3-phase synthesized from a Pico... :) Do you think 100V would be pushing that cable too far? The traditional 48V/50V telephone system uses a 90V AC ring voltage, albeit at low current. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
It might be a NZ only thing - private networks?  But yeah - the village owns the cables, so ZERO issue with whatever the hell we do with them, ONCE Chorus cut the link to the old copper exchange. 100v - Perhaps. Never thought above 50vAC for these cables..... Smoke makes things work. When the smoke gets out, it stops! |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6783 |
If they are in good condition I bet they'll withstand 250VDC on a pair (that's not a working voltage, it's a test one). A normal electrical tester would soon tell you (they only put about 1mA on). Careful, it does charge the cable capacitance so discharge it after testing. :) Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| phil99 Guru Joined: 11/02/2018 Location: AustraliaPosts: 2135 |
Once you go over 50VAC the LV regulations apply. The original problem is that the launch amps are fed from a LV supply separate from the building supply, which is no longer allowed. The reason for this regulation is to ensure that when the main switch at the switchboard is turned off all LV is off. ELV has less stringent rules. |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6783 |
AFAIK We could do this, but the supply from an external source has to have local isolation before you can access the terminals without a tool. If a tool is necessary (e.g. a screwdriver) then only an isolator and warning notice are required, without an interlock. There also has to be a label by the mains switch informing that there is an external supply coming into the premises that isn't controlled by that switch. This is because firemen isolate the building supply at that point. Of course, the incoming external supply can only be accessible in a sensible zone. Not a bathroom or somewhere like that. Mains supplies are actually a major problem in the UK. The incoming mains supplies traditionally have only the supply fuse as a means of isolation. However, it is illegal for anyone but the supply authority to pull that fuse. Hence, whenever an electrician needs to work on a consumer unit his only means of isolation is the built-in switch. Replacing a consumer unit means either getting the supply authority to pull the fuse (which they charge for) or break the law and pull it anyway. Luckily most of the meter companies will now install an isolator after the meter, but some still want to charge a lot of money to do so. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9306 |
I have an insulation-breakdown tester. I could certainly test a few pairs NOT connected to any house at the other end! My tester can do 250v, 500v and 1000v breakdown tests. I've actually used that thing before in the past, when EVERYONE was on copper, to confirm a pair that has had an insulation breakdown fault, which usually causes noise on the line such as popping or scratching, humming or buzzing kinds of complaints. Happily, I never have to worry about chasing those phone faults now, as most people are either on wireless or fiber, either of which, does not have to worry about any of those old copper line faults. Assuming I can get something that does eventually work, I would prefer to keep the supply AC voltage to 50v or less, then you don't have to worry about any regulations per-se', and it is still a pretty safe LV to have floating around. Fused, naturally, but as soon as I go above 50vAC, electrical regulations and compliance come into effect. And fair enough too. It's winter here at the moment, so I don't really get any fine days lasting long enough without rain, to do the on-site cable testing, but once we get out of the winter, I will be doing some tests on cable pairs not in use. EDIT: I still remember in the days when copper was king, fault-finding on crook pairs, and you'd be clipping your test phone onto the line where it connects to the house(after isolating the house), to test and see if the noise was cos of wiring in the house, or a bad pair feeding the line to that house......and someone would try to ring into that house, while you had you hands on the wires. The ringing-voltage certainly let you know it was there, especially if you were kneeling on damp ground. After that happened once or twice, I started using rubber mats(and shoes) on the ground to kneel on, to isolate me and prevent that! It also prevented my knees getting damp! Memories.....Edited 2024-06-18 10:17 by Grogster Smoke makes things work. When the smoke gets out, it stops! |
||||
| Page 3 of 5 |
|||||
 Print this page |
