UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 12th Apr 2018, 3:50 pm   #21
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by Argus25 View Post
I think the best way to handle this is to set up a situation where you know exactly what the peak current will be because you have placed a low value current limiting resistor in series,
Hey Hugo,

Excellent suggestion indeed.

I was at first was thinking of using resistors of a value such that by a combination of calculation and instinct, they are on the borderline of being destroyed instantaneously because of I^2R heating in the initial surge ( I would of course have suitable fire precautions etc.). The first few would then actually (intentionally) be destroyed instantaneously.

I would then incrementally reach a value such that the resistor is not destroyed instantaneously, and survives to a steady state when the caps are fully charged. I could probably get this right in less than five trials.

There are also, I noticed a while back, some film types that are designed for pulse currents, on a one-off basis, hundreds of times their RMS rating. Frustratingly, I wasn't as methodical as usual about this and haven't made a note of the types.

I'll look out for your paper on the other thread now...
__________________
Al
Al (astral highway) is offline  
Old 12th Apr 2018, 3:58 pm   #22
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by broadgage View Post

A value as low as 0.01 ohm implies a short circuit current of 24,000 amps, just about possible if you are next door to a very large transformer but almost unknown domestically.
I am about 5 metres above the main distribution board for the properties here - I've seen the giant busses when the door has been left open!!! I am 1KM from the main substation.


0.1 ohms implies a short circuit of about 2,400 amps, a very typical figure.
1 ohm implies a short circuit current of only 240 amps, a very low figure, which also implies significant voltage drop under load ...
A 1 ohm mains supply impedance is however much more likely at the point of use. A shed at the far end of a large garden, with an undersized cable from the house, could give such a figure.[/QUOTE]

That's a thoughtful way of putting these figures into context.

All in all, bearing in mind that I'm not in a shed, and I'm going to intentionally use a small, 2m long extension lead, I'd imagine I'm looking at rather less than 1R and rather more than 0.1R

0.5R seems too low, so let's say more than 0.25R but less than 0.5R

When I find a way to get more precise, I'll post the result.
__________________
Al
Al (astral highway) is offline  
Old 12th Apr 2018, 4:13 pm   #23
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: Source impedance of mains, a workaround?

Thinking laterally, the problem of high peak currents in rectifiers straight off the mains is already acknowledged in terms of its distortion of the mains waveform and poor power factor. I believe that the current rules regarding switch-mode power supplies require any supply larger than 70VA (from memory) to have power factor correction (or to achieve a certain power factor, which amounts to much the same thing).

If you're charging big capacitors, I'd look seriously at power factor correction circuits. They consist of a boost converter with a multiplier in its feedback loop so that the input current is always proportional to the input voltage, and thus causes no distortion and has a high power factor, but the output voltage is regulated to (typically) 400V. Many large switch-mode supplies already contain such a circuit which could be taken as-is, but you could build your own around a standard controller IC. I think the UC1854/2854 is the canonical example, but others are available. Charging big capacitors with low ESR straight off the mains is exactly what they're intended for.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 12th Apr 2018, 4:38 pm   #24
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by cmjones01 View Post
Thinking laterally, the problem of high peak currents in rectifiers straight off the mains is already acknowledged in terms of its distortion of the mains waveform and poor power factor.

If you're charging big capacitors, I'd look seriously at power factor correction circuits.

Hey Chris, that's a very helpful idea, thank you. I'll certainly investigate that and learn about it.

I should add, the obvious way for me to avoid any problems is simply to use the really established idea of a series R, timer and relay, to deal with inrush current. The problem is of course far from unique to my application and but also to high-powered amateur transmitters with beefy power supplies/ audio equipment with huge stacks of electrolytics.
__________________
Al
Al (astral highway) is offline  
Old 12th Apr 2018, 4:52 pm   #25
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by GrimJosef View Post
On a point of detail I'm not sure how relevant a datasheet ESR value might be when it comes to operating an electrolytic capacitor in a high-current regime.
Hey GJ,

An afterthought:

You raised an important point about ripple current and the capacitor's behaviour. I had a bit of a play with some duty cycles and other parameters.


Ripple current on the DC bus (let's call it Iripple) is proportional to the duty cycle times voltage, all divided by operating frequency times inductance (of the main inductor, not stray inductance, negligible here)

Where D=relative duty cycle,

Vbus=volts on the rails

f=200KHz (I don't know exactly yet, but this will be about right)

L=9uH

Vbus= 2 *1.414 *mains =740V

Duty cycle is deliberately going to be low, precisely because of overcurrent issues. Let's pick duty cycle =5%

So: Iripple=D *(1-D)*Vbus/f*L

Then Iripple =0.05 *(1-0.05)*740/200E3 *9E-6

Iripple=19.5A

And RMS ripple current=19.5/2*0.707=6.9A

So these figures seem fine. And of course we have our bi-directional TVS's....
__________________
Al
Al (astral highway) is offline  
Old 12th Apr 2018, 5:02 pm   #26
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by astral highway View Post
There are also, I noticed a while back, some film types that are designed for pulse currents, on a one-off basis, hundreds of times their RMS rating.
Once the pulsers start to get to a decent size e.g. this one (double-exposure person for scale towards the bottom of the picture)

https://cryptome.org/eyeball/aurora/aurora-13.jpg

copper sulphate solution resistors become popular. They will absorb huge amounts of energy in quite short pulses without coming to any harm. They do take a bit of maintenance and it's hard both to achieve and to sustain a very constant resistance. But in this type of kit 20-50% here or there rarely matters much.

It's very important, however, to contain the liquid in a flexible pipe - the ones you can see in the right foreground above will have been made in food-grade PVC tubing. This allows the solution to expand when it heats up, which it will. Large pulsers are almost always filled with transformer oil which, over a year or three, leaches the plasticiser out of the tubing, causing it to stiffen. I once made a largeish (about 10 litres ?) CuSO4 resistor in 8" diameter uPVC pipe without considering what the u stood for. This turned out to be a Very Bad Mistake.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline  
Old 12th Apr 2018, 5:24 pm   #27
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by GrimJosef View Post
Once the pulsers start to get to a decent size e.g. this one (double-exposure person for scale towards the bottom of the picture)

https://cryptome.org/eyeball/aurora/aurora-13.jpg
8

That's incredible! What's the application here? This is a power distribution network of some kind, right?
__________________
Al
Al (astral highway) is offline  
Old 12th Apr 2018, 5:27 pm   #28
McMurdo
Dekatron
 
McMurdo's Avatar
 
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,263
Default Re: Source impedance of mains, a workaround?

In industrial automation we use line reactors in series with the mains to mitigate peak currents on rectifier-fed loads. I've just specc'ed a mains mcb for an inverter drive and the rating table gives 20A, no reactors, or 15A, with reactors. Large installations in the 100kW+ range often use phase shift chokes and 6-phase rectifiers to minimise harmonic distortion / supply notching
__________________
Kevin
McMurdo is offline  
Old 12th Apr 2018, 5:34 pm   #29
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by Herald1360 View Post
If you know where this kit is going to be plugged in, why not just plug in a 3kW kettle there and measure the voltage
Hey Chris,

Smart idea, thank you.

And the results are very telling.

Test 1: Location, 13A outlet right next to cooker spur (not sure if connected or legal to be connected to eachother)

V1: 243V;
V kettle on: 241V
R=0.15R

Test2: 13A socket, adjacent room
V1:243V;
V Kettle on: 238V
R=0.38R

Test 3: 13A socket, adjacent room, plus 5m extension lead.
V1:243V,
V Kettle on 235V
R=0.6R

Accordingly, I'll def be using the 'extension lead current limiter' for the surge-testing phase, as it's an appreciable help. Pretty amazing for just 5 metres of twin and earth!
__________________
Al
Al (astral highway) is offline  
Old 12th Apr 2018, 5:38 pm   #30
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by astral highway View Post
That's incredible! What's the application here? This is a power distribution network of some kind, right?
No, it was a single pulse machine - an early nuclear weapons simulator https://cryptome.org/eyeball/aurora/aurora-eyeball.htm.

They're bigger now

https://share-ng.sandia.gov/news/res.../#.Ws-LFpdG19M
https://www.youtube.com/watch?v=eaopaLJk3-Y

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline  
Old 12th Apr 2018, 6:08 pm   #31
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by GrimJosef View Post
No, it was a single pulse machine - an early nuclear weapons simulator https://cryptome.org/eyeball/aurora/aurora-eyeball.htm.
Wow. That's some serious investment and an incredible research project.

I'm attaching a copy of an original photo I own of I think two young German research scientists in early atom-splitting experiments.
Attached Thumbnails
Click image for larger version

Name:	785F8168-245B-4959-9F2A-670666D7BD67.jpg
Views:	66
Size:	49.4 KB
ID:	161018  
__________________
Al
Al (astral highway) is offline  
Old 13th Apr 2018, 1:33 am   #32
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Source impedance of mains, a workaround?

Al
Trying to type on a tiny phone screen so sorry if messy..
There are two aspects to your problem , the surge when you initially switch on and the filter caps start from zero charge and the surge current on near the top of the sine wave when the rectifiers conduct to top up the charge. This latter one is the effect that power companies dislike as switch mode supplies dominate now so if you look at mains waveforms these days the tops of the sine waves are squashed down.

One thing that would help the turn on surge in this case is to synchronise the timing with mains Zero crossing switch , many pre made ones are available which are controlled by a standard current, voltage or logic level to switch them on . This way the rate of rise of the surge current is limited by the rise of the mains waveform to a lower peak value than it would be if the load of the rectifiers and discharged capacitors were connected at the moment the sine wave peaked. if a load is inductive it's not a good idea to switch on zero crossing, there was another thread about that.

The current peaks when the diodes conduct to top up the capacitors bewteen peaks can be ameliorated with a series inductor or reactor .

since over the time that the voltage is falling on the capacitors between peaks is only a small % of the total discharge time, you can model it as a linear decay and from Q= cv, Then dQ/dt = current = cdv\ dt, then the ripple voltage drop between the charging peak time = (load current divided by the capacitance) xtime. I always try to remember that for a 10,000uF cap on full wave 50 Hz or 10mSec charge timing and a current of one amp, the ripple voltage is one volt

One way to help contain it on turn on if you had a zero crossing switch and delay circuit you could ramp the voltage up in a series of steps with increasing amounts of each quarter cycle so the peak voltage gets a bit higher each time over many cycles like slowly turning up a light dimmer

Last edited by Argus25; 13th Apr 2018 at 1:59 am.
Argus25 is offline  
Old 13th Apr 2018, 3:49 pm   #33
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Hi Hugo,

I’m also now typing on a tiny screen. Thank you for your input on this double challenge. I’ve looked online at zero crossing solid state relays with DC 3-30 v on one side and the load on the other. One is made by a company you have cited a few times - Teledyne.

Is it as simple as merely closing the DC contacts at any arbitrary point , wiring the load o the other side , and the Zero crossing detector takes over?

I will consider this although I am already in ‘scope creep’ territory as the budget for the build grows again. I have to bear in mind that I will be operating at around just 5 per cent duty cycle, not CW, ever! This will be achieved by switching the whole circuit in bursts at a low frequency and low pulse width regardless of the RF frequency.

I can deal with the simpler problem of surge current easily as I have a suitable relay and resistor.

The load will be a low inductance (10uH) hard- switched in trialling and then self-resonant in the final design, if all the trials and data-gathering at that stage go completely to plan.

Afterthought, on the hoof ...

*would it help if I integrate the capacitor charging differential equation, solving for c?
*did you see my calcs for ripple, including the inductance and frequency?
__________________
Al

Last edited by Al (astral highway); 13th Apr 2018 at 3:54 pm.
Al (astral highway) is offline  
Old 13th Apr 2018, 8:45 pm   #34
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Source impedance of mains, a workaround?

Hi Al,

I saw your calcs for ripple current but I wasn't sure what your load circuit looked like, so my suggestion for the ripple voltage on your filter capacitor was just based on an average current drain over some time. As long as you know the approximate current and the time it is there for, it is easy to calculate how much the voltage on the filter cap will fall. The amount it will fall is simply inversely proportional to the value of the filter cap, proportional to the current and proportional to the time the current is taken . Of course the product of current and time is the charge taken from the capacitor and as mentioned this linear approximation is fine for a ripple voltage that is say 10% of the total capacitor voltage or less.

What is the average power that your DC supply will have to deliver ?
Argus25 is offline  
Old 13th Apr 2018, 9:53 pm   #35
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

Quote:
What is the average power that your DC supply will have to deliver ?
Hi Hugo,

It will go through iterations in development. I will trial it at low volts (80v) and with a transformer to limit the first test to 150W average.

The first mains increment (no isolation) will be at 300W.

If everything is completely glitch free at 500W, it will eventually run short bursts at probably double that. The components are capable, but there are other design considerations.

All of these stages will be carefully monitored, there's no rush.
__________________
Al

Last edited by Al (astral highway); 13th Apr 2018 at 10:18 pm.
Al (astral highway) is offline  
Old 13th Apr 2018, 10:46 pm   #36
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Source impedance of mains, a workaround?

From time to time I come across people who say that loads and source impedances should always be matched (usually in terms of RF transmitters)

So I use the mains source impedance as an example of how all hell breaks out if you do!

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 14th Apr 2018, 12:29 am   #37
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

A demonstration of that would be either extremely ‘lively’ or rather quiet , depending on which way the mismatch was configured!
__________________
Al
Al (astral highway) is offline  
Old 14th Apr 2018, 12:58 am   #38
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: Source impedance of mains, a workaround?

Al, can you post a schematic sketch of the parts you have been referencing in your posts? I'm unsure of where you have the TVS? And is the main filter cap after the mains bridge going to be the series connected caps of the half-bridge? I'm guessing the 1.5uF DC supply cap is physically positioned for the half-bridge transient currents, rather than the mains diode bridge transient currents?

With those D22-20 mains diodes, you will experience substantial reverse recovery. Have you simulated the charge up pulsing to see the pulse peak magnitudes over the first few mains cycles, and rationalised whether Tj will be ok given the 275A 10ms half sine single pulse limit rating? Did you look around for diodes with better Ifsm, or fast recovery specs?

Ciao, Tim
trobbins is offline  
Old 14th Apr 2018, 5:00 am   #39
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Source impedance of mains, a workaround?

Quote:
Originally Posted by astral highway View Post
A demonstration of that would be either extremely ‘lively’ or rather quiet , depending on which way the mismatch was configured!
No, there is no choice. I was talking about a load matched to the mains source impedance so there is only one value of load impedance. Let's say the mains at some particular socket looks like 0.25 +j0.25 Ohms. So we need a load of 0.25 -j0.25 Ohms for a proper conjugate match.

The reactive terms cancel, leaving 0.25 Ohms resistive Thevenin source driving 0.25 Ohms resistive load. The EMF is 240v, so the current taken is 240/(0.25+0.25) = 480A

The power in the load is 480^2*0.25 = 57.6kW

The power dissipated in the wiring is also 57.6kW

So the efficiency of getting power to your load is going to be 50% and then those transmission and generation inefficiencies which aren't modelled by the source impedance get a look in.

Quite lively until a fuse or breaker manages to open.

I use this as a way of illustrating that if you buy a transmitter it often has "50 Ohms" written by the output connector. This means "please connect me to 50 Ohms" and not "I am 50 Ohms". So an antenna tuning unit is used to load-up the transmitter with the impedance of its dreams. It might be matched, but then it often isn't. People often chuck the word "matched" in without thinking. Similarly, to many people, all screens are Faraday screens and they aren't aware that they've invoked a specialist item that doesn't do what they were thinking of.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 14th Apr 2018, 8:11 am   #40
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Source impedance of mains, a workaround?

David-that’s a neat demonstration of an underlying concept about source impedance and load matching: boldly memorable and illustrative. And then your point about what it means if a component says ‘50R’ ‘...doesn’t mean I am 50R, it means connect me to 50R’. Thank you!

That’s probably a ‘source’ of confusion to many!

Tim, thank you, but at this stage I don’t want to be posting circuit diagrams. My original question is pretty much done and I’d rather wait until I’ve been though a process of building successively more powerful versions under test conditions, attempting to solve at least the inrush current problem, if not also the phasing problem. The ‘as built’ versions of course can behave very differently to the design versions.
__________________
Al
Al (astral highway) is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:50 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.