Math Formula to calculate DC power supply capacitor value?

[Nigel Goodwin]

I worked in a TV repair shop part time 1970 new TVs came out with new tubes they were called, compactrons.

Can't say I've ever heard that term.

Crazy idea put 2 complete tubes inside 1 glass body I assume to save space tube sockets were mounted on printed circuit boards.

Nothing crazy about it, it makes great sense, and it was done LONG before the 70's - the American 'tube' numbers don't make much sense, but the European 'valve' numbers certainly do.

Such as ECC83 - E means 6.3V heaters, C means triode (so CC = two triodes) and 8 means B9A base (3 means it's different to a 1 or a 2 ) - it's an audio twin triode.

Likewise PCL85 - P means 300mA heaters, C is triode again, and P is pentode - it was the standard frame valve in B&W TV's for many years.

You even got valves with more than two in the same envelope.

Essentially it was a precursor for integrated circuits, creating composite devices for specific purposes - and back in the day, that purpose was radio.

That only lasted 3 years then circuit boards were plug in with transistors. To repair a TV unplug the bad circuit board and replace it. Tubes were gone from TVs by the end of 1973 except for the 1B3GT high voltage tube. OH.....almost forgot picture tube is a tube also.

Valves ceased to be used in the UK well before the US, and we never had transistor sets with valve EHT rectifiers. As far as repairs went, you simply repaired the boards rather than replacing them - it made in home service a lot easier. The outside engineers would fit an exchange board, and bring the old one back for me to repair - to that end I build two test jigs, one for Thorn 3000 panels, the other for RBM A823 series - the two most popular chassis's we sold.

 

[Nigel Goodwin]

Hi again,

Well ok, what would you suggest then for a 'universal' power supply with a four diode bridge and puts out 50 amps?
Line voltage 120vac let's say 60Hz.

Assuming low voltage unregulated? - transformer, bridge rectifier, reservoir capacitors - as everything you see uses.

Although these days, it's much likely to be switch-mode anyway.

 

[MrAl]

I worked in a TV repair shop part time 1970 new TVs came out with new tubes they were called, compactrons. Crazy idea put 2 complete tubes inside 1 glass body I assume to save space tube sockets were mounted on printed circuit boards. That only lasted 3 years then circuit boards were plug in with transistors. To repair a TV unplug the bad circuit board and replace it. Tubes were gone from TVs by the end of 1973 except for the 1B3GT high voltage tube. OH.....almost forgot picture tube is a tube also.

Hi,

Yeah i could not remember when tubes went out altogether in my area. I used to collect tubes when i was a kid.
One day i went to Radio Shack and they had removed the tube tester. I thought that was strange at the time.

 

[MrAl]

Assuming low voltage unregulated? - transformer, bridge rectifier, reservoir capacitors - as everything you see uses.

Although these days, it's much likely to be switch-mode anyway.

Hi,

Yes unregulated, to start anyway.


Yes i agree probably switch mode, but i felt that was too complex for this thread. I guess that depends too though on what the user wants to do.

But here's one of my points, which i might remind is still a side point by my view, and that is that we SEE a transformer, we see a bridge rectifier, we see reservoir capacitors (we also call filter caps i am sure you know), but what we DONT see is the built in leakage inductance in the transformer.
Even the small 3v, 6v, 9v, 12v unregulated wall warts have leakage inductance built into their power transformers. That is very significant because for one if we dont have some inductance there and we short out the secondary, we either burn up the transformer or blow some fuse somewhere. Secondly, the diode peak current is immensely reduced and the power line harmonics are also reduced (see first link again in my post with the two links).
Could you imagine what the diode peak current would be with a 50 amp output load?
With a tightly couple transformer (power transformers are tightly coupled unless specifically designed otherwise) we might even see a surge so high that it blows the mains breaker every time the power supply is turned on after a time being off.

So to me these issues are very significant even for a one off power supply.

Taking all things into consideration it's probably not a beginner's project though even with an inductor

 

[gary350]

Hi,

Yes unregulated, to start anyway.


Yes i agree probably switch mode, but i felt that was too complex for this thread. I guess that depends too though on what the user wants to do.

But here's one of my points, which i might remind is still a side point by my view, and that is that we SEE a transformer, we see a bridge rectifier, we see reservoir capacitors (we also call filter caps i am sure you know), but what we DONT see is the built in leakage inductance in the transformer.
Even the small 3v, 6v, 9v, 12v unregulated wall warts have leakage inductance built into their power transformers. That is very significant because for one if we dont have some inductance there and we short out the secondary, we either burn up the transformer or blow some fuse somewhere. Secondly, the diode peak current is immensely reduced and the power line harmonics are also reduced (see first link again in my post with the two links).
Could you imagine what the diode peak current would be with a 50 amp output load?
With a tightly couple transformer (power transformers are tightly coupled unless specifically designed otherwise) we might even see a surge so high that it blows the mains breaker every time the power supply is turned on after a time being off.

So to me these issues are very significant even for a one off power supply.

Taking all things into consideration it's probably not a beginner's project though even with an inductor

Is this the power supply your talking about?

What is, built in leakage inductance in the transformer?

This PS is built with a microwave oven transformer. It was originally wired to produce 15vdc on the PS. After getting the circuit running I changed it to 18vdc. After getting it running better I changed it to 22vdc that is the most wire that will fit in the available space of the transformer. I would have done 24vdc if I could got 2 more turns on the secondary coil. No room for transformers shunts I left them out.

There are 5 bridge rectifiers in parallel they are suppose to be rated 1000v 50a but I don't think so. In the beginning 3 or 4 bridge rectifiers use to go up in smoke in 30 seconds that was before I added 8 more 3300uf filter capacitors the .01uf ceramic caps in parallel and had a good 6uh choke coil. I once had several TV choke coils in series to get 6uh but 20 guage is too small it over heats. Now I have an LCR meter I know for sure my choke value is 6uh. My circuit drawing originally called for 240 ohm mosfet gate resistors I found another identical circuit that uses 330 ohm resistors. After changing resistors to 330 ohms idle current is 1a. Before idle current was 15a. I think the 220 ohm resistors were not letting mosfet turn completely off. It would be helpful to own a scope to learn what is going on in the circuit. The good thing is I am learning new stuff, learning how, mosfets, chokes, filter caps, PS, and all the parts work together. Better than college lab class 50 years ago. Microwave transformer primary coil is not continuous duty this circuit can never run more than 10 or 15 minutes. I have several 100a diodes, several 250a and 600a diodes I can build a better PS if I need to but this is mostly a learning project.

**broken link removed**

**broken link removed**

 

[Nigel Goodwin]

You need to be aware that you can't simply place rectifiers in parallel, minor differences in their EXACT specifications mean that one or more will take all the load, overheat, and blow. If this doesn't happen fairly promptly, it WILL happen after a year or two of use. It's a common fault in various TV's where they (stupidly!) do the same thing.

 

[gary350]

You need to be aware that you can't simply place rectifiers in parallel, minor differences in their EXACT specifications mean that one or more will take all the load, overheat, and blow. If this doesn't happen fairly promptly, it WILL happen after a year or two of use. It's a common fault in various TV's where they (stupidly!) do the same thing.

YES I learned that. The bridge rectifiers were all from China. After smoking a dozen rectifiers i found a very old good quality rectifier in my parts box see the black color rectifier on the left. After adding the good rectifier the circuit has been working fine up to 50 amps. I know I should change it but until it goes up in smoke again I will keep using it. I have several NPN & PNP stud mount 100 amp rectifiers i only need 2 heat sinks for 4 diodes. At the moment heat sinks are the number 1 problem. I don't have what I need and I am not finding what I need either. I need to go ahead and build something better before it go up in smoke again. Heat sinks are a learning project too if they are too small they need to be made larger or add cooling fans. I would rather not use fans on this sooner or later this project will junked and parts used for something else. Places that sell heat sinks i would like to have want $20 each plus postage. I have a lot of small heat sinks from junk TVs too small. I have several aluminum plates I could maybe cut and bend to make a suitable heat sink. I could replace the 13 capacitors 3300uf each with 1 larger 74,000uf capacitor I would hate to smoke it, I don't mind smoking the 3300uf caps they cost less they are doing good at the moment. On another thread someone mentioned several small caps in parallel work better than 1 large cap.

 

[MrAl]

Is this the power supply your talking about?

What is, built in leakage inductance in the transformer?

This PS is built with a microwave oven transformer. It was originally wired to produce 15vdc on the PS. After getting the circuit running I changed it to 18vdc. After getting it running better I changed it to 22vdc that is the most wire that will fit in the available space of the transformer. I would have done 24vdc if I could got 2 more turns on the secondary coil. No room for transformers shunts I left them out.

There are 5 bridge rectifiers in parallel they are suppose to be rated 1000v 50a but I don't think so. In the beginning 3 or 4 bridge rectifiers use to go up in smoke in 30 seconds that was before I added 8 more 3300uf filter capacitors the .01uf ceramic caps in parallel and had a good 6uh choke coil. I once had several TV choke coils in series to get 6uh but 20 guage is too small it over heats. Now I have an LCR meter I know for sure my choke value is 6uh. My circuit drawing originally called for 240 ohm mosfet gate resistors I found another identical circuit that uses 330 ohm resistors. After changing resistors to 330 ohms idle current is 1a. Before idle current was 15a. I think the 220 ohm resistors were not letting mosfet turn completely off they were probably oscillating between 10a & 15a. It would be helpful to own a scope to learn what is going on in the circuit. The good thing is I am learning new stuff, learning how, mosfets, chokes, filter caps, PS, and all the parts work together. Better than college lab class 50 years ago. Microwave transformer primary coil is not continuous duty this circuit can never run more than 10 or 15 minutes. I have several 100a diodes, several 250a and 600a diodes I can build a better PS if I need to but this is mostly a learning project.

**broken link removed**

**broken link removed**

Hi there,

Thanks for your question.

Well i was talking about a general 50 amp power supply, either with inductance or without inductance. This probably includes yours but i'd have to see a schematic of yours to be sure.
The one i was talking about was a very basic one, with a bridge rectifier and filter caps, and a transformer and either with or without an inductor or built in leakage inductance.

Leakage inductance (or built in leakage inductance) is the equivalent inductance in series with either the primary or secondary of the transformer. It's the inductance you get from loosely coupled windings. It's the same as what you would have if you used a tightly coupled transformer and separate inductor where the value of the separate inductor would be equal to the value of the leakage inductance.
It is purposely built into the transformer for various reasons especially in DC power supplies.
It comes from the way the laminations are stacked as well as how the two coils (primary and secondary) are placed on the core.
For an example, your microwave oven transformer shown in your picture has significant leakage inductance because for one it is butt stacked and also because the windings are placed one next to the other rather than one on top of the other. The other benefit there though is that the secondary has better electrical isolation from the primary so is safer than one winding on top of the other.
If that is your actual transformer then you already have inductance in the circuit and so it is quite possible you dont need any extra inductance. You could look up ways to measure it if you want to find out some approximate value.

Yeah a scope is used to test new power supply designs. There are some cheaper ones being sold today that would be just fine for 50 or 60Hz power supplies. Look on Amazon or something like that.

As Nigel was saying, rectifiers in parallel dont work out too well. You have to be a little lucky i guess to get them to work right, but what helps is if you use separate wires to each rectifier which are all the same length so that they have a little current balancing. You might be able to tell how well they are sharing current by seeing how much each one heats up.

By removing the shunts from the microwave oven transformer it is possible that you lowered the overall permeability and thus the transformer saturates and gets hot. If you could add turns to the primary that would help. If you could add a variac to the front end you could lower the voltage and stop the transformer from overheating, assuming everything else was right too.

 

[schmitt trigger]

Re: The compactron, they were multiple vacuum tube functions incorporated into a single glass envelope.
According to Wiki, these are some of the common devices:

• 6AG11 double diode similar to 6AL5, double triode high-mu similar to 12AT7. Designed for FM stereo multiplex service.
• 6BK11 triple triode. Two of the triodes are simmilar to 12AX7 and one of them is simmilar to 5751.
• 6C10 high-mu triple triode for color matrix amplifiers in television by Sylvania, etc.... not related to the Edison Swan (later Mazda) 6C10 triode-hexode
• 6M11 twin triode - pentode. Designed for sync separators and AGC amplifier circuits.
• 6K11 triple triode. Designed for sync separators and AGC amplifier circuits.
• 6LF6 beam power pentode with anode cap. Designed for horizontal output service.
• 8B10 twin triode - twin diode. Designed for horizontal phase detector service, and horizontal oscillator service.
• 12AE10 twin pentode. Designed for FM discriminator/detector, and audio output.
• 38HK7 pentode diode. Designed for horizontal output service and as a damper diode
• 1AD2 diode high voltage, used in flyback transformer rectification

But similar to the sub-miniature Nuvistors, these were an established technology's last stand against the transistorized steamroller which was rapidly replacing them across the board.

 

[unclejed613]

something about compactrons that often confused people were the odd heater voltages, 33, 38, 50, 21, etc... the reason for this becomes clear when you see that these tubes were used in "AC/DC" (also known as "hot chassis") TV sets. the heaters were wired in series, and in the USA, had to add up to 110 - 120 volts. "hot chassis" sets were made into the early 1970s because there were still a few rural areas with DC service. i just read that the last DC power system in the US was shut down in 2007 (?!?!?!?!?!) https://en.wikipedia.org/wiki/War_of_the_currents#Remnant_and_existent_DC_systems

as a side note, equipment using a typical SMPS will run from DC wall outlets, but anything using power transformers will "let the magic smoke out" of the transformers. there are some types of motors that can run on AC or DC (motors with brushes), but other types (induction motors) will burn out.

 

[unclejed613]

it's still not clear what the OP is trying to accomplish. the parallel string of bridge rectifiers and caps is a reliability nightmare (Nigel mentioned that the forward voltage on the diodes is going to be all over the map, so one of them is going to have most of the current burden). in additin to the diodes not sharing the current very well, is the fact that when there are a lot of electrolytics in parallel, they tend to fail catastrophically. it only takes one cap to have slightly higher ESR than the rest, and that one cap will soon self-destruct. there may even be a couple of threads here on ETO about the phenomenon.

 

[gary350]

Hi there,

Leakage inductance (or built in leakage inductance) is the equivalent inductance in series with either the primary or secondary of the transformer. It's the inductance you get from loosely coupled windings. It's the same as what you would have if you used a tightly coupled transformer and separate inductor where the value of the separate inductor would be equal to the value of the leakage inductance.
It is purposely built into the transformer for various reasons especially in DC power supplies.
It comes from the way the laminations are stacked as well as how the two coils (primary and secondary) are placed on the core.
For an example, your microwave oven transformer shown in your picture has significant leakage inductance because for one it is butt stacked and also because the windings are placed one next to the other rather than one on top of the other. The other benefit there though is that the secondary has better electrical isolation from the primary so is safer than one winding on top of the other.
If that is your actual transformer then you already have inductance in the circuit and so it is quite possible you dont need any extra inductance. You could look up ways to measure it if you want to find out some approximate value..

I have a factory made circuit with four 2N3055 transistors in parallel on each heat sink. Factory used nichrome wire to connect the transistor pins in parallel. I was told the nichrome wire provides the resistance needed to make the 4 transistors turn on and work together. There are 2 heat sinks both with 4 transistors, a transformer, 4 resistors, 2 capacitors, 2 diodes, when connected to a 12v car battery the transformer puts out 120 vac. This is an old 1970s type DC to AC inverter. It appears to be the same type circuit used in an induction heater circuit. I wonder if bridge rectifiers or mosfets can be connected in parallel with nichrome wire.

If I intend to use a microwave EI core for a continuous duty transformer I wind new primary coils the correct way on my lathe math came out to 200 turns on the primary of the last transformer I did. Microwave EI core spaces are too small for the extra 100 turn primary that leaves very littler room for a secondary winding. Need to ignore the core is 1500 watts wind it like it is a 600 watts then leave out 1/2 of the EI laminates us them for another transformer project.

I have a box of micro wave transformers primary coils put 2 primaries on the same EI core us 1 as the secondary to make a 1 to 1 ratio transformer. The secondary should be 120v 12.5a to proove the EI core is can handle 1500 watts. I have several heating elements from hot water heaters they are all 2 ohms each keep them in water for cooling. Amp clamp on primary and secondary wires with a load when primary reads 120v 12.5a secondary should read about the same minus maybe 2% for transformer losses. I have a 3KW transformer 240v primary 120v secondary I can test EI cores to 15a, 20s, 25a at 120v for a few seconds. Core saturation can be adjusted by adding or taking away EI laminates.


**broken link removed**

 

[Nigel Goodwin]

something about compactrons that often confused people were the odd heater voltages, 33, 38, 50, 21, etc... the reason for this becomes clear when you see that these tubes were used in "AC/DC" (also known as "hot chassis") TV sets.

Again, it's your silly American numbering system that causes 'confusion', the European system clearly labelled (most) TV valves as P, signifying 300mA series heaters (not all, as some were E - but just happend to be 6.3V 300mA). There was also another designation U, which signified 100mA series heaters, U series valves were commonly used in AC/DC radios.

So no confusion over here, as everything was clearly labelled.

As for mains supplies, likewise over here we had a mix of AC or DC, and various different voltages, and even different AC frequencies.

Only about 6-7 miles from where I live there was a small town that had it's electricity supply from the large industrial complex that was the towns reason for been, and that used a slightly unusual voltage (not quitesure what?), but more importantly it was only 25Hz instead of 50Hz. This was still in use for part of my working life, although it was outside of our service area, but talking to engineers from round there is was common practice to add extra electrolytics across the reservoir capacitor, to reduce the hum bars on the picture.

Obviously the introduction of the National Grid eventually ended all such unusual schemes, and everything became 240V 50Hz.

 

[MrAl]

I have a factory made circuit with four 2N3055 transistors in parallel on each heat sink. Factory used nichrome wire to connect the transistor pins in parallel. I was told the nichrome wire provides the resistance needed to make the 4 transistors turn on and work together. There are 2 heat sinks both with 4 transistors, a transformer, 4 resistors, 2 capacitors, 2 diodes, when connected to a 12v car battery the transformer puts out 120 vac. This is an old 1970s type DC to AC inverter. It appears to be the same type circuit used in an induction heater circuit. I wonder if bridge rectifiers or mosfets can be connected in parallel with nichrome wire.

If I intend to use a microwave EI core for a continuous duty transformer I wind new primary coils the correct way on my lathe math came out to 200 turns on the primary of the last transformer I did. Microwave EI core spaces are too small for the extra 100 turn primary that leaves very littler room for a secondary winding. Need to ignore the core is 1500 watts wind it like it is a 600 watts then leave out 1/2 of the EI laminates us them for another transformer project.

I have a box of micro wave transformers primary coils put 2 primaries on the same EI core us 1 as the secondary to make a 1 to 1 ratio transformer. The secondary should be 120v 12.5a to proove the EI core is can handle 1500 watts. I have several heating elements from hot water heaters they are all 2 ohms each keep them in water for cooling. Amp clamp on primary and secondary wires with a load when primary reads 120v 12.5a secondary should read about the same minus maybe 2% for transformer losses. I have a 3KW transformer 240v primary 120v secondary I can test EI cores to 15a, 20s, 25a at 120v for a few seconds. Core saturation can be adjusted by adding or taking away EI laminates.


**broken link removed**

Hello again,

Ok, well then i dont see how the first post relates to this schematic. The first post in this thread seems to ask about a DC power supply filtering capacitor or something. Maybe you could explain.

 

[gary350]

Hello again,

Ok, well then i dont see how the first post relates to this schematic. The first post in this thread seems to ask about a DC power supply filtering capacitor or something. Maybe you could explain.

You are right, It has nothing to do with it. We solved the PS question now we are just talking.

After you explained how valve number system on foreign valves work it does make good sense. I guess it is like everything else someone makes up a system and some how it becomes the system everyone uses. Electricity was all types of strange voltages 100 years ago about 1970 it was standardized to 120 volts in USA but most people still talk about 110 volts. A person still needs the data sheet for each valve when design circuits or substituting valves.

 

[schmitt trigger]

This is what is called thread drift.

Nigel: how did the town inhabitants cope with their electric lights? 25 Hz would have produced a headache-inducing flicker.

Of course, here the Niagara Falls power station was originally set for 25 Hz, too.

 

[Nigel Goodwin]

Nigel: how did the town inhabitants cope with their electric lights? 25 Hz would have produced a headache-inducing flicker.

I've no idea, I didn't live there

 

[schmitt trigger]

I read that the Swiss Railways utilizes 16 2/3 Hz as their catenary power.
How did they do with the carriage lights, previous to high frequency converters? Perhaps a rectifier and batteries?

 

[shortbus=]

If I intend to use a microwave EI core for a continuous duty transformer I wind new primary coils the correct way on my lathe math came out to 200 turns on the primary of the last transformer I did.

You seem to like to do things unconventionally. Most people leave the primary alone and wind a new secondary. Since the primary is made for the local mains voltage and usually people want much less than the high voltage that the MW secondary has.

After reading most of your postings over the years I'm surprised you haven't won a Darwin Award, or at least been a runner up.

 

[MrAl]

You seem to like to do things unconventionally. Most people leave the primary alone and wind a new secondary. Since the primary is made for the local mains voltage and usually people want much less than the high voltage that the MW secondary has.

After reading most of your postings over the years I'm surprised you haven't won a Darwin Award, or at least been a runner up.

Hi,

Well once the shunts are removed it changes all that i think. I havent done any measurements yet though but the current should eb higher without the shunts.