OK, so now that the application is known here is a thought, which is that such an application does not need a low ripple voltage unless you are actually doing electroplating of something like show car chrome.
My guess had been that it was for operating old automotive accessory items. Oh Well.
The most efficient rectification system will be a center-tapped full wave arrangement, because of only one diode drop in the loop. But that does require twice as many turns in the transformer secondary, which might be a challenge. Each diode will only be dissipating half as much power, so that is a benefit as well. And there is the easy option of cutting the power simply by switching off one side.

That’s exactly why I wasn’t mentioning the application. It does require a low ripple, as far as rectification I’m clueless as to what is the best/easiest/cheapest, I was just gonna use the alternator diodes for free and easy. If building a full bridge rectifier is cheap and easy, I’m all ears. Any filter can be used to smooth out the signal, I only know about pi filters.
The absolutes are 7 volts dc 24 amp, low ripple, and long term run times.

 

You don't know or understand how mrbill operates. It has no bearing on you, which as a new member have been very good. It has to do with him only reading parts of a thread(usually the parts with pictures) then making comments on things that have already been discussed and worked out.

I see, thanks for letting me know. Good to have when I want to mass produce.

 

Hi,

I've worked a lot to use mots, and I can say the following:

1. If your main is 110V AC, it's OK to use the primary connected to the main. But if it is 220V AC, it is not: the mot will heat more and more until it become damaged. You can try the following: connect the primary to the 220V AC main, and wait for some time (say 20min). You'll probably observe that you can fry a chicken on the mot. How this works in the microwave oven is probably a mixture of an exact magnetic design, and of a strong fan at the back of the oven.

2. So, if you work with 220V AC, the only way to use the mot is to connect the secondary (that you have removed) to the main, and to use the primary. Interestingly, this may be exactly what you need, for it will give you about 20V at the primary (that is, your secondary), and a lot of current. From 20V, you have to extract 7V.
Nevertheless, pay attention that the Voltage at the output of the mot is strongly dependent upon the current used: this is because mots are essentially very lossy and bad transformers. So, if you use 24A, the voltage will probably drop strongly, perhaps near 7V, and you are done for this part of the work. So, simply try this solution: connect the secondary of the mot to the 220V main, find a dummy load that can handle 24A AC, and watch up the voltage at the load, when the mot is feeding 24A. If the voltage is near 7VAC, you have chance.

3. You see, 24 amp is huge. It is difficult to filter such a big current. First, you'll need to build your own bridge rectifier, using several big diodes in parallel (that's the easy part), then you'll have to use a bank of 10mF capacitors for energy storage (thats the costy part). Then I would use a capacitance multiplier (with 2 or 3 transistors in parallel) to filter the ripple (that's the electronic part). Finally, you end up with an unregulated DC power supply. If you need regulation, you need to build the regulator with an oamp and 2-3 big transistors in parallel (electronic part again).

Good luck.

 

I suggested a full-wave center tapped arrangement because it only uses 2 diodes and so has half the diode drop, and also the internal loss is split between two sections of winding so half the heat on any part. AND, for starters, 0.7 volts less ripple before filtering.
Next, 7 volts and 24 amps is about 175 watts, which is medium power and less than the transformer is capable of. One thing to know about microwave oven transformers is that they are intermittent duty in normal operation. But that is at a higher power level. So you will need to watch for excessive heating for the first few hours.
And yes, occasionally I do skip sections of the thread when they seem to have wandered off the topic.

 

I suggested a full-wave center tapped arrangement because it only uses 2 diodes and so has half the diode drop, and also the internal loss is split between two sections of winding so half the heat on any part. AND, for starters, 0.7 volts less ripple before filtering.
Next, 7 volts and 24 amps is about 175 watts, which is medium power and less than the transformer is capable of. One thing to know about microwave oven transformers is that they are intermittent duty in normal operation. But that is at a higher power level. So you will need to watch for excessive heating for the first few hours.
And yes, occasionally I do skip sections of the thread when they seem to have wandered off the topic.

Unfortunately, the mot has no center tape, only 2 terminals at the primary and the secondary.

 

I suggested a full-wave center tapped arrangement because it only uses 2 diodes and so has half the diode drop, and also the internal loss is split between two sections of winding so half the heat on any part. AND, for starters, 0.7 volts less ripple before filtering.
Next, 7 volts and 24 amps is about 175 watts, which is medium power and less than the transformer is capable of. One thing to know about microwave oven transformers is that they are intermittent duty in normal operation. But that is at a higher power level. So you will need to watch for excessive heating for the first few hours.
And yes, occasionally I do skip sections of the thread when they seem to have wandered off the topic.

I should mention I was using 120 volt mains. I have seen two transformers provide 7 volts at 24 amps, both mots. It used two giants caps as a pi filter and 6 alternator diodes. This is where I got the idea. Now I believe he wound a secondary that provided 14 volts and ended up with 7 after all of that, but I’m not sure. All I know is that it can be done the way I’m trying to do it. I’m sorry I’m not understanding everything you guys are telling me. This is over my head, but here are the constants:
- 120 mains
- alternator diodes
- giants caps for a pi filter
- 7 volts dc hopefully 24 amps (if not possible then as many amps as possible)
Can this be done? If not please dumb down your other solutions using as many of those constants or comparable solutions that an amateur is capable of implementing. Thanks everyone!

 

OK, if a winding of 14 volts was made then it probably had two seven volt sections. This is what would normally be called "a full-wave center-tapped" arrangement. It will need to have more turns on the secondary, the part you wind, and it will need two diodes, not 4 or 6. It will not need any change in the primary connection, which you would not be changing anyway. There are a lot of references that will describe the connections of this arrangement, so it shouold not be hard to find a detailed drawing. The other benefit is that each of the two sections is only delivering power for half of a cycle, and so each side has half as much heating. The diodes also have half as much power, and so half as much heating. But still, if the voltage drop on each diode is 0.7 volts, and the current is 25 amps, the power is close to 20 watts. So you will need heat sinks for the diodes, and possibly a fan. This will be a fairly high powered supply, no question there.
The filter inductor will be a challenge because it will need to have anough iron to not reach magnetic saturation at the full current. So it will certainly be big and heavy.
OK, there is an explanation with only a small bit of basic multiplication, which goes volts times amps gives watts. And electrical watts of loss is about equal to heat watts that must be kept from raising the temperature. Magnetic saturation is the point where adding more magnetizing force does not increase the magnetization any more. When saturation is reached in a filter inductor bad stuff happens. Others can provide a more detailed description ifyou are interestd.

 

Maybe use one the 'rejected' MOT, remove the primary as well and wind the smoothing inductor on that?

 

Maybe use one the 'rejected' MOT, remove the primary as well and wind the smoothing inductor on that?

That may possibly have enough iron to do some filtering. The math to check for saturation is complicated, though. But a few turns of #10 should provide some filtering.

 

DON'T USE A SMOOTHING INDUCTOR. This will be huge and is an horrible price to pay. In addition this will (probably) not work or not work well. Use big Schottky diodes with heat sink, or several big diodes in parallel.
If you can use a centered taped configuration, do it, but that is not so essential.
You DO NEED big storage capacitors.
To smooth the ripple, use a capacitance multiplier with 2-3 big transistors in parallel (or a single one that can handle 24 A). This is a much smaller price to pay than using smothing inductors as a passive filter. A capacitance multiplier is simply a transistor whose base is connected to a RC filter. This works very well, but for the huge current you need, you'll have either to choose a transistor that can handle this current, or to put several transistors in parallel (also need an heat sink for the transistor).
Finally, you'll have to decide if you need voltage regulation (this is probably the case). If so, you'll have to build the regulator (not so complicated, even for an amator).

 

I feel the need to add other things.
You are an amateur (like me 4 years ago) and you are completely unaware of many problems inherent to power supplies. This is not an easy task, even for 3 A current, and requires a good technical level. That is not an accident if "the art of electronics" has dedicated a big chapter just for that. For example, when you'll turn on the power, there is a huge so called "inrush" current (several hundreds of amp) charging the capacitors. Your main is probably not able to handle that, and you'll end up with global main shut down. So, a current limiter is needed to charge slowly the capacitors when turning on.
Another thing you have to think about is overcurrent protection. Without it, it is very easy to turn your supply into an incendiary device (I'm serious, without required protections there will be a fire). Finally, if you want to be (not so) pedant, you'll have to think about filtering the spikes into the main. You can use the filter from your (canibalized) microwave oven near the connection to the main. There are other things and the best I can do is to recommend you "the art of electronics", chap. 9.

 

I feel the need to add other things.
You are an amateur (like me 4 years ago) and you are completely unaware of many problems inherent to power supplies. This is not an easy task, even for 3 A current, and requires a good technical level. That is not an accident if "the art of electronics" has dedicated a big chapter just for that. For example, when you'll turn on the power, there is a huge so called "inrush" current (several hundreds of amp) charging the capacitors. Your main is probably not able to handle that, and you'll end up with global main shut down. So, a current limiter is needed to charge slowly the capacitors when turning on.
Another thing you have to think about is overcurrent protection. Without it, it is very easy to turn your supply into an incendiary device (I'm serious, without required protections there will be a fire). Finally, if you want to be (not so) pedant, you'll have to think about filtering the spikes into the main. You can use the filter from your (canibalized) microwave oven near the connection to the main. There are other things and the best I can do is to recommend you "the art of electronics", chap. 9.

Why was an acquaintance of mine able to accomplish this with two mot’s two foot long caps with a six inch diameter and 6 car alternator diodes? He may have been running 220, but I don’t think so considering it was a standard 120 mains plug. This is all he used. He definitely used the “full wave center tap” method mr.bill is informing me on. He did one winding for 7 volts then another for 14.

 

Inread the remarks by coquelicot and find them a bit overexcited. Since you asked about winding an inductor it will not be a "huge price" although it will also not be small. And iif that person has learned over 4 years, good for them. I have 6 years of technical college folowed by almosy 50 years of engineering experience, much of it designing industrial equipment to be used on automotive production lines. That is a rather demanding application.
You do need to provide some overload protection but the reason is not so dramatic as presented.

 

You do need to provide some overload protection but the reason is not so dramatic as presented.

A single fuse should suffice or, to be really posh, a circuit breaker.

 

Inread the remarks by coquelicot and find them a bit overexcited. Since you asked about winding an inductor it will not be a "huge price" although it will also not be small. And iif that person has learned over 4 years, good for them. I have 6 years of technical college folowed by almosy 50 years of engineering experience, much of it designing industrial equipment to be used on automotive production lines. That is a rather demanding application.
You do need to provide some overload protection but the reason is not so dramatic as presented.

I agree with you Mr.Bill, even though my .02 cents weighs about as much as a penny. I want to reiterate that I have seen this done with my own eyes with the list of Constants I provided above. This setup ran for 150+ hours sometimes. Cooling a system and making it safe is the easiest part of any project similar to this. I appreciate your input Co and honestly I don’t know enough to say your right, or your wrong. I do know that I don’t want fire, but I also know that what I’m talking about doing and the way I’m going to do it will not spontaneously combust without warning. I will watch for the warnings and plan for that contingency.
Mr.Bill I like where you were going. Let me recap, I need two windings (secondary) that output 14 volts total? You believe that the alternator diodes are not a great route, can you expand on rectification and filtering, please. Everyone else is welcome as well, maybe the secondary smoothing is still an option, or is co correct about that being a bad idea?

I apologize for the mistakes in my writing I’m working at the same time.

 

I feel the need to add other things.
You are an amateur (like me 4 years ago) and you are completely unaware of many problems inherent to power supplies. This is not an easy task, even for 3 A current, and requires a good technical level. That is not an accident if "the art of electronics" has dedicated a big chapter just for that. For example, when you'll turn on the power, there is a huge so called "inrush" current (several hundreds of amp) charging the capacitors. Your main is probably not able to handle that, and you'll end up with global main shut down. So, a current limiter is needed to charge slowly the capacitors when turning on.
Another thing you have to think about is overcurrent protection. Without it, it is very easy to turn your supply into an incendiary device (I'm serious, without required protections there will be a fire). Finally, if you want to be (not so) pedant, you'll have to think about filtering the spikes into the main. You can use the filter from your (canibalized) microwave oven near the connection to the main. There are other things and the best I can do is to recommend you "the art of electronics", chap. 9.

I will definitely give this book a read, but no one else here has told me how I’m completely unaware of problems inherent to power supplies. It is not necessarily true that I am completely unaware. It is the fact that I’m not the kind of person to do things that are dangerous without having someone with the wisdom to help point me in the right direction. I know I am asking basics, but I assure you I’m not incompetent. Due to the nature of large amounts of current and uncovered transformers, I’m willing to leave my ego and knowledge at the door in order for someone experienced to get me where I need to be safely.

 

I didn’t want anyone else talking about electrolysis. If I didn’t mention what it was for I knew the creativity and conversation would flow much easier. If their is a specific objective in mind everyone will just give me the exact way it’s supposed to be done and I don’t get to learn from you guys about the hardcore electrical engineering side of it(I know we haven’t discussed anything hardcore yet)!

Fair enough! I know little about electrolysis,

Hi,

I've worked a lot to use mots, and I can say the following:

1. If your main is 110V AC, it's OK to use the primary connected to the main. But if it is 220V AC, it is not: the mot will heat more and more until it become damaged. You can try the following: connect the primary to the 220V AC main, and wait for some time (say 20min). You'll probably observe that you can fry a chicken on the mot. How this works in the microwave oven is probably a mixture of an exact magnetic design, and of a strong fan at the back of the oven.

2. So, if you work with 220V AC, the only way to use the mot is to connect the secondary (that you have removed) to the main, and to use the primary. Interestingly, this may be exactly what you need, for it will give you about 20V at the primary (that is, your secondary), and a lot of current. From 20V, you have to extract 7V.
Nevertheless, pay attention that the Voltage at the output of the mot is strongly dependent upon the current used: this is because mots are essentially very lossy and bad transformers. So, if you use 24A, the voltage will probably drop strongly, perhaps near 7V, and you are done for this part of the work. So, simply try this solution: connect the secondary of the mot to the 220V main, find a dummy load that can handle 24A AC, and watch up the voltage at the load, when the mot is feeding 24A. If the voltage is near 7VAC, you have chance.

3. You see, 24 amp is huge. It is difficult to filter such a big current. First, you'll need to build your own bridge rectifier, using several big diodes in parallel (that's the easy part), then you'll have to use a bank of 10mF capacitors for energy storage (thats the costy part). Then I would use a capacitance multiplier (with 2 or 3 transistors in parallel) to filter the ripple (that's the electronic part). Finally, you end up with an unregulated DC power supply. If you need regulation, you need to build the regulator with an oamp and 2-3 big transistors in parallel (electronic part again).

Good luck.

Well coquelicot said pretty much what I'd been thinking.... which was "If I were you, I wouldn't be starting from here".

To be honest, to get a ripple spec of 1%, you're looking at a 30mH choke with a 30A capability, and 4000uF input, 10000uF output capacitors.

Based on your 10 turns gave 11.8V, your secondary winding needs to be 9 - 10turns wound with 8awg wire (based on
700CMils/Amp for continuous operation), but I recommend you wind it as 4 x 14awg wires together, it'll be easier to wind and less lossy. You need to produce 10v - 11v on load.

You're losing nearly 100W in the diodes, you probably should put 2 in parallel for each and they'll need a big heatsink. A synchronous rectifier solution would be better.

The choke is tricky, a commercially available one is nearly £80. You could, as has been suggested, use one of the other cores. It needs about 50 turns of 4 x 14AWG to give 30mH. Thats a lot of wire and loses about 15W



A Capacitance multplier might be a better solution, I've used them before on 4 -5Amp systems, though my experience of them for high current applications isn't a positive one. I'll knock a solution up and try one out for you.

 

A single fuse should suffice or, to be really posh, a circuit breaker.

I don’t remember there being a circuit breaker for the setup I saw besides the one in the shop on the wall. That will be another easy task though, right?

 

There are a lot of suggestions flying around. I’m trying to respond to everyone and learn from it all. Everyone has a different opinion about how to do this. I got on YouTube even though I was warned against doing this and I found almost exactly what my friend had done. I’m thinking now that maybe he got the idea from YouTube.

 

Fair enough! I know little about electrolysis,



Well coquelicot said pretty much what I'd been thinking.... which was "If I were you, I wouldn't be starting from here".

To be honest, to get a ripple spec of 1%, you're looking at a 30mH choke with a 30A capability, and 4000uF input, 10000uF output capacitors.

Based on your 10 turns gave 11.8V, your secondary winding needs to be 9 - 10turns wound with 8awg wire (based on
700CMils/Amp for continuous operation), but I recommend you wind it as 4 x 14awg wires together, it'll be easier to wind and less lossy. You need to produce 10v - 11v on load.

You're losing nearly 100W in the diodes, you probably should put 2 in parallel for each and they'll need a big heatsink. A synchronous rectifier solution would be better.

The choke is tricky, a commercially available one is nearly £80. You could, as has been suggested, use one of the other cores. It needs about 50 turns of 4 x 14AWG to give 30mH. Thats a lot of wire and loses about 15W

View attachment 211719

A Capacitance multplier might be a better solution, I've used them before on 4 -5Amp systems, though my experience of them for high current applications isn't a positive one. I'll knock a solution up and try one out for you.

Wow.... you guys are next level. I’m still working on calculus to get my associates. What does a synchronous rectifier entail. Watch the video I posted and see what you think about it.