 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 : Solar : High power DC relay Setup.
Page 1 of 2   |
|||||
| Author | Message | ||||
| George65 Guru  Joined: 18/09/2017 Location: AustraliaPosts: 308 |
I want to switch some high power ( 300V, 10A) DC current, preferably with a SSR. This way I can use an ordinary switch with a small 12V relay and use that to drive the SSR to switch the major load. I bought a couple of DC Solid state relays but I have read they don't stand up to DC switching as advertised. I'd like to learn a simple way of switching high power DC in a simple and straightforward way as I have seen illustrated but now can't find. I am sure I saw a diagram somewhere not too long ago that from memory was 2X DPDT relays kind of cross wired. I have plenty of Mechanical as well as SSR's both in AC and DC form so any suggestions would be appreciated. |
||||
| Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138 |
Here is a simple circuit I just drew up that could be used as the basis for a high voltage isolated switch; uses two mosfets back to back so to be bidirectional as to in and out. Requires an isolated low power DC rail to supply the FOD3180 opto driver chip. Note I haven't drawn in any circuit protective devices or bypass caps. Shouldn't be too difficult to locate suitable 500-600Vdc mosfets with a low on resistance.  The above circuit can switch at high speed 50Khz or so, you haven't mentioned intended usage, personally I would use an evacuated relay if just slow on/of switching was required see these US$60 each. High Volage Relay Cheers Mike |
||||
| George65 Guru Joined: 18/09/2017 Location: AustraliaPosts: 308 |
Thanks for taking the time and effort with that Mike. I have some mosfets but will have to check what they are. I have been looking for the relay setup I -think- I saw but can't find anything. Maybe I was dreaming or am thinking of something else. I always save or bookmark these things but I can't find anything so maybe I was dreaming. Seems so difficult to perform such a straightforward task. |
||||
| Solar Mike Guru Joined: 08/02/2015 Location: New ZealandPosts: 1138 |
>> 2X DPDT relays kind of cross wired. You could wire the relay contacts in series, so they all open at the same time, thus having multiple air gaps to help prevent any arc, however with 300 vdc, if any air gaped contact opens carrying current the air will ionize and the arc current jump across the open contacts burning them out in the process. You could try a 4 pole relay with all in series, but I feel the total contact gap would still be too small to prevent an arc forming at 300v, perhaps 2 x 4 pole relays giving 8 contacts in series may work; that's why they go to vac relays - no air to ionize. The mosfet switch is efficient but susceptible to voltage spikes etc, shorting out the mosfet, especially if any inductive loads are in the circuit. Mike |
||||
| hotwater Senior Member  Joined: 29/08/2017 Location: United StatesPosts: 120 |
There is another way if the relay has NO/NC contacts. Place a capacitor from those two contacts. From the NC contact place a high value resistor from the NC to common supply. This gives you nearly the full AC rating of the relay on DC. This assumes you are not switching really fast. |
||||
| George65 Guru Joined: 18/09/2017 Location: AustraliaPosts: 308 |
Is this for an AC relay or either? I have some double pole relays I think have both NO and NC contacts I bought recently. Does the cap and resistor quench the arc or something? The switching will be every 5mi9n to 6+ hours so no worries there! I have been thinking about just doing what I do with Breakers for my Arrays, Wire a DC relay on each pole so they break together. Haven't had any problems with the Breakers and as these DC relays are supposedly for that purpose, Doubling them up should add a measure of overkill anyway. Hmm, Just saw the DC ssr's I was going to use are only rated to 22oV on the output. Wanted to use them around 350. The AC SSR's I have are rated at 380. Will voltage actually make a difference to these things? The control side is 3- 32V which is fine. |
||||
solarwind Regular Member  Joined: 03/02/2008 Location: South AfricaPosts: 51 |
The combination of a low value resistor in series with a capacitor and then placed across the contact, is called a "snubber" or "snubber" network. I once used it across an ac circuit breaker which I had and used it to interrupt the dc coming from the PV into the DC panel. I used a filament lamp in series with a high voltage electrolytic capacitor. When the circuit breaker is opened, the lamp glows for a while diminishing soon to nothing while the capacitor charged up. The DC from the PV was effectively isolated. When the circuit breaker is closed again, the lamp glows as the charge on the capacitor dissipates through it. Once again, no arcing of contacts. Using an electrolytic capacitor of the correct voltage and connected with the correct polarity is of course very important. The only weak link, so to speak, is a scenario where the lamp fails............! Any way, in an application such as this, the lamp is not stressed and should last for ages. To be more secure, two lamps can be wired in parallel. If one fails, the time constant will change, but the circuit will still perform its function.The same or similar circuit could be used across relay contacts that operate at low frequency, such as the thermostat of a geyser when connected directly to dc. You don't have success until you've tried it! |
||||
| hotwater Senior Member Joined: 29/08/2017 Location: United StatesPosts: 120 |
RC networks tend to be ineffective in many DC applications. The capacitor has to be so large and resistance so small that it tends to burn contacts on closure. Electrolytic capacitors are also prone to failure. In the method I talked about, the capacitor charges while the load is on. Hence, the need for only a high resistance to charge it. When the relay contacts switch the capacitor places a reverse current on on the arcing contact and quenches it. A FET seems like an easy cheap solution. |
||||
| George65 Guru Joined: 18/09/2017 Location: AustraliaPosts: 308 |
I'm going to try a couple of other things which probably amount to one in the same. I'm going to try having a small load like you mentioned, a light bulb ( in this case in series to handle voltage) connected across the switched sides of the relay all the time. The small amount of current this carrys will be of no consequence to what I want to do when the relay is off. The second thing is rather than switching the load on and off, transfer it to a dummy load. I'm thinking that when the contacts in the relay switch, they will be effectively grounded by the other terminals on the DPDT relay and the arc will quench that way. The relay switches in Milliseconds so while the contacts may still have to endure some arcing, even if I get a year out of $3 relay, I won't be worried about a shortened life. I am going to wire the relay across both sides of the output from the panels. I have a feeling this will make a difference as well rather than just switch one side as normally done. If this works, I'll go one step further and wire both contacts of the double pole relays to each side of the outputs thus increasing amperage capacity and providing some more isolation. Looks like it might be a bit sunny today so I'll have some decent power to test with hopefully. Now just have to go and dig out some lights and things to load up with for the tests. :0) |
||||
| George65 Guru Joined: 18/09/2017 Location: AustraliaPosts: 308 |
Just tried the DC switching with the Relay. Switching Both poles from the panels was a Fail. Arc maintained and would have till something melted or caught fire. Switching with the DPDT relay to a Dummy load however seemed highly successful. Certainly a good spark but nothing I would not expect to see with a high wattage AC load. I tested at 330V @around 500W and have no reservations the AC mechanical relay I used would not last a decent service time. May not go the 10M cycles the data sheet says but If it got 1m or even 100K cycles, I'd be plenty happy with that as it would still represent years of service for what I want to do. The Dummy load I used was 1100W so more than the power being produced. I'll now try the same test with a couple of 50W 240V bulbs in series to allow for the over voltage from the panels and see how that goes. If that works to keep the arcing from teh main load manageable, then maybe I can do a secondary switching with the globes as the load and keep the arcing manageable that way? |
||||
yahoo2 Guru Joined: 05/04/2011 Location: AustraliaPosts: 1166 |
it depends on what the risk is George. or how valuable the component is that needs protection, relays that cycle everyday and can be seen to work are not a problem. The one that is a problem is where it is supposed to disconnect when other components develop a fault. for me an undetected welded contact is a bigger risk than a relay failure open circuit. even with flames  as long as the arc stops eventually it is all good.usually there are two service cycle numbers one is mechanical cycling (no load) the other is electrical and thermal load most fancy contactors and relays have a graph on their data sheet of current vs cycles. What you do is match the cycles you need to the load rather than taking the advertised nameplate amps at face value. example a 50 amp contactor will do the same job as a 400 amp version of the same type if we switch 5000 times instead of half a million. I'm confused, no wait... maybe I'm not... |
||||
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
I have MOSFETS switching tens of thousands of times a day to regulate my solar array connected to my GTI which then charges my batteries via the off-grid inverter. It would not take much to setup something similar for what you are doing, Mine is operating 2 arrays at close to 500VDC and 10A. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| George65 Guru Joined: 18/09/2017 Location: AustraliaPosts: 308 |
Not taking much to you is a lot different to what it would take for me but none the less, I'd be very interested and appreciative if you would advise on how you suggest to do it and hopefully I can follow and put it into practice. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
You need these MOSFETs one of these Isolated powers supplies. 2 20K 0.5W resistors silicone insulator pads for the MOSFETS and a heat sink. It is very important that the MOSFETs are electrically isolated from the heatsink and keep in mind you are dealing with high voltage DC. You can get the data sheet for the MOSFETs here so you will know which pin goes where.  Turn on the AC power and it will conduct, turn it off the PV power stops. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
That is interesting Gary, thanks for posting. In other words, an alternative 'bang, bang' switching method, this time switching the DC side of the inverter. While my AC side switching (fooling the GTI there is a mains outage) works fine for me I am wondering about long term effects on the GTI, having the mains drop out every day. Switching the panels on or off simulates the day/night cycle to the GTI. I shall try that DC switching when my work bench looks for another project  .I would only require 1 array switching (10A max) and, from that Mosfet spec sheet, a relatively small heat sink, capable of dissipating 10W or so. Klaus |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Klaus you could take it one step further and use it as a PWM regulator based on your battery voltage. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Here is a PDF of the circuit that will be easier to read. 2018-04-19_213607_schemeit-project.pdf There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| George65 Guru Joined: 18/09/2017 Location: AustraliaPosts: 308 |
Thank You very much. I did some reading up on Mosfets and this looks like a great solution. Amazing such small devices can handle so much power! I have ordered the parts, now just need to find the silicone sheets. This is a nice simple solution to what otherwise seems like a very difficult problem. Thank you again. |
||||
| yahoo2 Guru Joined: 05/04/2011 Location: AustraliaPosts: 1166 |
whew, got there in the end! I have been waiting for you to jump in Madness, explained it a lot better than I could. I agree, things have changed a lot in the last decade. I think you will be happy you had a go at this, gives your build so much flexibility. I'm confused, no wait... maybe I'm not... |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Insulators George who knows where it will lead to with a little dip of your toe into the world of electronics. Once you gain some knowledge it is easy to expand on it. Henry Ford said, "It does matter if you think you can or you think you can't you are probably right". There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Page 1 of 2 |
|||||
 Print this page |
