Go Back   Home > Forums > >

Transistor Based, Switched Capacitor Power Supply
Transistor Based, Switched Capacitor Power Supply
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Closed Thread
 
Thread Tools
Old 1st December 2020, 12:32 AM   #1
StevenStanleyBayes is offline StevenStanleyBayes  Canada
diyAudio Member
 
Join Date: Jan 2012
Location: Ottawa
Default Transistor Based, Switched Capacitor Power Supply

This project has NOT been practically made and may not be soon, because of difficulties to home make this project and lack of equipment, such as an oscilloscope, etcetera.

This is, also, why I want to see the comments and the opinions of others.

A possible idea is to make an integrated system with a power supply, an amplifier, two speakers, etcetera, which, does not have any mechanical inputs and outputs, such as jacks, sockets, speaker connectors, potentiometers, etcetera, instead, everything is integrated and controlled by a remote control or through Bluetooth.

The same applies to the input signals, which are not mechanically connected to the system with jacks and sockets, but, instead, through Bluetooth only.

Because the system does not have any mechanical inputs, outputs and control, the case can be grounded to Ground and Neutral can be used as an internal, system ground. Thus, Ground will NOT be connected to Neutral, which means, such a system is as safe as a stove or power tools, even, safer.

The document was supposed to be very tiny, but, became still tiny, yet, not as tiny as intended. The schematics are in the addendum, so, please, scroll all the way with Ctrl Home.

Thanks for your help and advise!

Here is the document :

Transistor Based, Switched Capacitor Power Supply - Google Drive
 
Old 1st December 2020, 12:34 AM   #2
StevenStanleyBayes is offline StevenStanleyBayes  Canada
diyAudio Member
 
Join Date: Jan 2012
Location: Ottawa
One of the possible, interesting questions is whether 2SC5200 and 2SA1943 would heat up significantly just by switching them without any load?
 
Old 1st December 2020, 04:19 AM   #3
IanHegglun is offline IanHegglun  Australia
diyAudio Member
 
Join Date: Mar 2017
Hi Steven,

Bear in mind diyAudio topics according to the rules must be safe. I would think that excludes direct-off-the-mains supplies with amplifiers and speaker outputs which float with mains voltages to ground (relative to metal chassis etc).

This would apply to full-wave bridge rectified mains where the capacitor terminals has alternate mains voltage to ground.

It may not apply to the half-wave rectifier or the voltage doubler - effectively two half-wave rectifiers (as you discuss in your document) where one side of the capacitor IS connected to the mains neutral which is also 'bonded' to earth in most countries.

I say "may not apply" because I am not an authority on the legality of direct-off-the-mains half-wave rectifier or doubler arrangements even in my country. It may depend on whether it is for DIY and a never for sale item and a never for export. It depends on each countries regulations and standards.

Anyone, please correct me if I am wrong or missed something important or if I confused some.

In the 1960's through 1970's most tube based TV's in New Zealand and Australia had direct-off-the-mains supplies and you could not touch their chassis.and anything to earth at the same time. The UHF tuner had a ferrite core to isolate the mains from the TV aerial - but sometimes the insulation failed and sent mains down to another TV sharing the same aerial giving a shock to anyone touching the other TV's aerial connector (as was my experience). Also IIRC Bob Carver's first "cube amp" before he sold any was a transformerless direct-off-the-mains amp.

BTW I suggest as a precaution you should add a note at the beginning to the effect that the ideas involve direct-off-the-mains power and anyone attempting to make these should take extra precautions with earthing as well as the higher voltages can be lethal. Using an isolation transformer during development would be a wise idea. Also an earth leakage relay in the mains would be a good idea.as well. Once developed and in a suitable enclosure (with any metal that can be touched grounded) then an isolation transformer is not needed (it may depend on each countries regs), but the earth leakage relay would still be a good idea.

Quote:
One of the ways to make a transformerless power supply from the standard, mains, hot and neutral voltage is to convert this AC voltage to a DC with a bridge, scale down the DC and use a standard, switched flying capacitor, charge pump inverter, such as LTC7820 and the magnificent TDK Lambda converters
I'm not sure exactly what circuit you have in mind here. Could you add a basic circuit to clarify please.

BTW An article on direct-off-the-mains supplies was published in Electronics World Sept 1997 and can be seen on MyDrive/My Articles "Switched-capacitor-power-supplies" and the patent referred to is also on MyDrive. One idea for isolation without a transformer to 3V for PC's was to use relays and flying batteries (below) but I have never seen this used commercially. For audio amps you don't need the second down converter on the isolated side.
Transistor Based, Switched Capacitor Power Supply-scps-battery-relay-mains-kv-isolation-fig13-png
Attached Images
File Type: png SCPS-Battery+Relay-mains-kV-isolation-Fig13.png (13.0 KB, 166 views)

Last edited by IanHegglun; 1st December 2020 at 04:27 AM. Reason: Gdrive isn't allowing Get link just now - try later
 
Old 1st December 2020, 04:45 AM   #4
IanHegglun is offline IanHegglun  Australia
diyAudio Member
 
Join Date: Mar 2017
Oops, did anyone spot the problem in the relays in the above post? Well done.
Corrected below
Transistor Based, Switched Capacitor Power Supply-scps-battery-relay-mains-kv-isolation-fig13-png
Attached Images
File Type: png SCPS-Battery+Relay-mains-kV-isolation-Fig13.png (12.6 KB, 160 views)
 
Old 1st December 2020, 06:39 PM   #5
Holmer is offline Holmer  Germany
diyAudio Member
 
Join Date: Apr 2019
If one of the relay contacts gets stuck and the other switches over someone can get killed.
Even if safety relays are used which prevent this by locking the contact pairs together, the contact clearance is below legal requirements.

Big contactors may provide the clearance, but they are LOUD.
 
Old 2nd December 2020, 12:09 AM   #6
StevenStanleyBayes is offline StevenStanleyBayes  Canada
diyAudio Member
 
Join Date: Jan 2012
Location: Ottawa
No, this is not a commercial power supply and is not to be made for to be sold.

Yes, the best way would be to use a transformer. However, the transformer must not be very powerful and not of high voltage, or, the device would be even more unsafe than the straight off the mains ones. I have found a switching transformer, which, may be suitable for audio : FA2443-AL.

I think the most important safety requirement is to have the chassis grounded and not connected to neutral. Thus, whatever happens inside, the case will always be connected to ground and whether a faulty hot or neutral touches the case is irrelevant as they would go to ground ( and the fuses would blow in case hot touches case ). This, however, does not allow to connect equipment, because, the outputs and inputs provide case connections, connected to their input and output connections and this would connect neutral to ground. This is why I have mentioned such a power supply be used only without inputs and outputs and without any potentiometers and any galvanic control. Just an infrared or radio ( Bluetooth ) controls and a radio ( Bluetooth ) inputs. No speaker out[puts too. Speakers must be integrated. Even better, the enclosure must be fiberglass or plastic.

Stoves, Air Conditioning, etcetera, take 240V and do not have transformers. Even in case anyone is to put a transformer in a stove, this transformer would be very powerful and high voltage and would not provide any protection despite the isolation. Thus, the transformers used must be low voltage and, even better, low power.

Most standards specify only a single fault protection. This means any metallic case must be grounded, so, in case of a fault when hot touches case, hot goes to ground and nothing happens. However, in case of this and a faulty ground connection, the case may be disconnected from ground. However, this is a double fault scenario. 1. Hot touches case and 2. Ground connection to case is broken. Because this is a double fault and not a single fault, such scenario is considered safe.

Yes, another condition is any inputs and outputs not to be connected to mains in any way. I would add, they must not be, indirectly or directly, connected to a high voltage and power transformer. Thus, any tube amplifiers with high voltage and high power transformers and high voltage converters may also be unsafe.

As far as I remember, LTC7820 is a voltage converters, which, with one of the arrangements ( the schematics is in the datasheet ), LTC7820 can be configured to be a 10A, 24V, voltage inverter and make a good negative voltage from a positive one.

The " magnificent TDK Lambda converters " are AC to DC or DC to DC integrated converters. I am not sure whether they have a switching transformer inside. An example for AC to DC integrated converter is KMS40-15. There are other AC to DC converters in the KM brand. They ca be connected directly to mains. An example of a DC to DC integrated converter may be PH150A280-15. There are other DC to DC converters in the PH-A280 brand.

All TDK Lambda converters are expensive.
 
Old 2nd December 2020, 12:40 AM   #7
StevenStanleyBayes is offline StevenStanleyBayes  Canada
diyAudio Member
 
Join Date: Jan 2012
Location: Ottawa
The document has been updated with Figure 14c. and explanations thereof.
 
Old 2nd December 2020, 01:14 AM   #8
IanHegglun is offline IanHegglun  Australia
diyAudio Member
 
Join Date: Mar 2017
Quote:
Originally Posted by Holmer View Post
If one of the relay contacts gets stuck and the other switches over someone can get killed.
Even if safety relays are used which prevent this by locking the contact pairs together, the contact clearance is below legal requirements.
Big contactors may provide the clearance, but they are LOUD.
Hi Holmer,

Isn't sticking of relays due to welding? But in my arrangement welding is prevented by using 3 relays, so when one is 'flying' it doesn't break voltage, so in principle no arcing or welding.

Clearance is then determined by air breakdown, which I think must be several kV for relays like the relays we use for speaker protection.

Since this is a non-standard application existing standards are probably not relevant. But as a lone designer wouldn't want to get special approvals at today's rates.

It was just a crazy idea that I thought could be used for audio amps to reduce weight. Don't all new ideas appear crazy at first?
 
Old 2nd December 2020, 01:29 AM   #9
StevenStanleyBayes is offline StevenStanleyBayes  Canada
diyAudio Member
 
Join Date: Jan 2012
Location: Ottawa
Another clarification for the general document, I would like to add, is, the document shows there is a transformer, but, this transformer is provided either by the power distribution industry or by the building construction company. This is a high power transformer. Thus, an additional, low power transformer is necessary for most any design.

However, the circuit and the circuit idea can be used after an additional transformer, although, perhaps, there are much better circuit solutions for such a case.

The circuit can be used for high voltage transformers for multi voltage applications, but, again, although higher voltage transformers are considered safe, they are not.

Any voltage higher than 60V is considered unsafe. This is why people can lick 9V batteries and can touch the standard, 12V, high power, low voltage car battery, people should not do this with higher than 60V batteries and supplies.

Thus, the maximum, safe, dual power supply, whether from batteries or transformers is lower than +- 30VDC, because, to make +- 30VDC one needs a converter, which, usually needs some overhead voltage.

Thus, a power supply of +- 25V can be considered the highest safe power supply. This leaves 5V overhead, which is OK with most 78xx and 79xx converters with bipolar buffers.

However, when the converter is buffered by a MOSFET, the overhead voltage is Ugs, which, typically, is 10V and, therefore, the highest safe voltage is +- 20V.

In other words, the rule is, the highest voltage difference at and after the secondary of a transformer must be lower than 60V.

60VDC is considered maximum. Better to be lower than this.

Then, on the top of this, there may be people with different sensitivity, different body and skin impedances, etcetera. Also, these 60V have been defined for the general case and, the voltage may need to be even lower when there is a possibility for the user to touch with two hands, in which case the path of the current goes through the heart.

There was a joke amongst digital and computer engineers : " Any voltage, higher than 5V, do not touch! "

I would say, any dual voltage higher than +- 18V, do not touch!
 
Old 2nd December 2020, 02:08 AM   #10
IanHegglun is offline IanHegglun  Australia
diyAudio Member
 
Join Date: Mar 2017
Ah, Steven, now I can see what you are trying to do. The "Vc" and "Ve" in Figs 0-2 connects to the series transistors in Fig.3 to drop the rectified mains voltage of up to +/-180V down to +/-18V for supplying amplifiers.

Even if you switch the series transistors fast you will always end up with the same power losses as a linear series-pass .regulator. This is a fact of life when charging a capacitor without using any inductor (like switching power supplies use). This issue was a question covered in my engineering course and it is covered in some physics texts.

In your circuit you drop 180-18V or about 160V so the efficiency (Pout/Pin) is only 11%. Again it is the same if you pulse the power through as when you use a series pass regulator.

BTW the EW 1997 article I mentioned covers variable ratio charge pumps with minimal power losses. The link is EW-WW_Hegglun_SCPS-Switched-capacitor-power-supplies_Sep-1997.pdf - Google Drive

Another way to make charge pumps variable ratio using PWM is to incorporate a small inductor in the diode path as shown in this patent Hegglun_SCPS-AU729687-patent-sealed_1998.pdf - Google Drive

Thanks for mentioning the LTC7820. I think I missed that one. It is similar to the one in the EW 97 article. I'm pleased someone finally made an IC for this job -- albeit only for a fixed ratio. Nice work ADI.. The patent can be seen here US9484799B2 - Switched capacitor DC-DC converter with reduced in-rush current and fault protection
- Google Patents
by Jindong Zhang and Jian Li.

Last edited by IanHegglun; 2nd December 2020 at 02:12 AM.
 

Closed Thread


Transistor Based, Switched Capacitor Power SupplyHide this!Advertise here!
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Converting un-switched to switched outlet (vintage receiver) maurycy Power Supplies 2 25th February 2016 08:56 PM
Switched mode power supply MMicke Car Audio 3 19th February 2005 11:42 AM
Switched power supply? Johan_Ch Tubes / Valves 38 31st January 2004 12:53 PM
Buildind switched power supply for amp soundNERD Chip Amps 21 30th November 2003 03:29 PM
Switched Mode vs Linear Power Supply royalmp2001 Solid State 4 12th July 2002 09:04 PM


New To Site? Need Help?

All times are GMT. The time now is 02:08 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki