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 > Specific Vintage Equipment > Vintage Radio (domestic)

Notices

Vintage Radio (domestic) Domestic vintage radio (wireless) receivers only.

Closed Thread
 
Thread Tools
Old 22nd Oct 2020, 9:11 am   #21
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,287
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

I have just returned from the workshop having replicated Trevors experiment.
The first thing I have to observe is that 22uf is too small for the reservoir capacitor.

I used a transformer that had 23V on the secondary and with a 1.5K load I measured some 13V of ripple across the 22uF capacitor. This will confuse any voltage measurements across the diode. I increased the capacitor to 220uF.
I connected the ground of the capacitor to the positive of the capacitor / cathode of the diode as this is the most stable part of the circuit. The probe was connected to the anode.
I measured some 64V p-p ac across the diode which agrees with the expectation of 2X peak ac.

Peter
Electronpusher0 is offline  
Old 22nd Oct 2020, 9:24 am   #22
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Diodes fail in two ways.
Excess forward current causing the junction to burn out. The other way is when the PIV is exceeded, the junction actually flashes over, in a way like capacitor can break down.
A capacitor is not made like a Diode..... Obviously, on my drawing and the practical circuit I don't see where this 700v is coming from. The proof of the pudding has been why not only myself but many others have used allegedly under rated caps in this circuit with no issues for years.

I'll do another test today with a 400v dc rated cap and the same circuit, incorporating a 50ma IC fuse and bring the voltage up gradually to 275v and see if the cap fails, if not I'll try a cap rated at 250 or 300v dc if I can find one.

Looking forward to others trying the same or similar test.
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 9:29 am   #23
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Quote:
Originally Posted by Electronpusher0 View Post
I have just returned from the workshop having replicated Trevors experiment.
The first thing I have to observe is that 22uf is too small for the reservoir capacitor.

I used a transformer that had 23V on the secondary and with a 1.5K load I measured some 13V of ripple across the 22uF capacitor. This will confuse any voltage measurements across the diode. I increased the capacitor to 220uF.
I connected the ground of the capacitor to the positive of the capacitor / cathode of the diode as this is the most stable part of the circuit. The probe was connected to the anode.
I measured some 64V p-p ac across the diode which agrees with the expectation of 2X peak ac.

Peter
Em.... Interesting Peter. I'll redo this later with a bigger cap. I'd agree the ripple was quite high.
What was the P-P off the mains tx on your scope?
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 9:56 am   #24
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,287
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Hi Trevor, I did not measure the p-p of the secondary on the scope but it measured 23V rms on a digital meter.
The ripple current agrees with theory, to a first approximation the peak to peak ripple on a capacitor is:
8 times the current in the load in Amps divided by the capacitor value in thousands of uF for full wave and
16 times the current in the load in Amps divided by the capacitor value in thousands of uF for half wave.
The load current is about 15mA using a 1.5K resistor so the calculation is
16 X 0.015 / 0.022 (current in Amps, C in thousands of uF)
= 10.9 V p-p

As I say the formula above is an approximation but comes quite close in practice and shows that a measured ripple of 13V p-p is about what is to be expected.
Do try again with a bigger cap, the ripple voltage will just confuse the diode voltage measurements. The smaller the capacitor the larger the ripple and the lower the voltage on the diode eventually when the cap value becomes zero the diode voltage becomes the peak of the ac instead of 2 X peak

Peter
Electronpusher0 is offline  
Old 22nd Oct 2020, 10:29 am   #25
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Hi.
I just about remember this being discussed in my final year at college about capacitors in power supplies. This was 1974 and by that time silicon rectifiers were the norm. The mains filter was usually a 0.1uf 250 or 300v ac cap. Failures were common as we know, even RS and Duramold types and the replacement would be a 0.1uf 1000v dc cap. This was brought up in college and we were told that a 1000v dc cap is derated to 250v ac so is a suitable replacement in that circuit. The lecturer said that the relationship is 4x of 250v ac cap at DC and vice versa, whether that was correct I don't know but that stuck over the years.
Caps across the silicon rectifier were also discussed as switching noise was noted even then and also in thyristor power supplies a filter cap was fitted in some sets across the rectifier and some thyristor power supplies too. I even remember the caps rated at 250 or 300v ac. He did say that the applied voltages on a capacitor across the rectifier shouldn't be confused with PIV and the normal theory didn't apply, he didn't elaborate though. So in fact according to theory I could be wrong but this doesn't answer the question why I've never had a failure. Perhaps the rule of 4 is valid today and the AC cap is still rated at 1kv?
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member

Last edited by murphyv310; 22nd Oct 2020 at 10:35 am.
murphyv310 is offline  
Old 22nd Oct 2020, 10:37 am   #26
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Trevor, you say in your video that the peak to peak voltage shown on the 'scope when multiplied by a factor of 10 is 300 volts (you counted three vertical divisions) that means the the peak to peak voltage of the waveform at the input to the diode is 30 volts, that means the RMS voltage is 10.65 volts, you also said that the AC voltage into the rectifier was either 24 or 25 volts, unless I'm missing something nothing about the video adds up.

Lawrence.
ms660 is offline  
Old 22nd Oct 2020, 10:46 am   #27
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

OK Lawrence.
Most of what you say has been negative about my video.
If that's what think then fair enough. I'll remove the video if you think it's wrong and ask the moderators to remove the thread.
I do not wish to be goaded into a disagreement so I'd rather have this closed. Perhaps some useful help and encouragement would have been a better way to go.
Thank you.
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 11:05 am   #28
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

I've now deleted the video.
I may consider redoing this later at 230v with a 16uf reservoir cap, 1N4007, 22ohm series limiter and a 4. 7k load. I'll double check the scope calibration and leads.
If I'm wrong I'll put my hands up and admit my error. The question then is why no failures
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 11:39 am   #29
smiler411
Tetrode
 
smiler411's Avatar
 
Join Date: Apr 2008
Location: Rhondda Cynon Taff, Wales, UK.
Posts: 87
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

This site may explain it better than me

https://www.quora.com/Why-is-peak-in...pacitor-filter

If you look at the graphs for the half wave with capacitor you can see that the cathode of the diode is always at a positive voltage that is at or slightly below (depends on load and capacitor size) the peak with respect to 0V. The anode side varies between +peak voltage and -peak voltage with respect to 0V.

So the maximum voltage across the diode is almost 2 X Vp-p as it will be the sum of the peak voltage plus the voltage on the reservoir capacitor. Normal 250V ac capacitors are designed to be fitted across the mains or similar voltages where the voltage across the capacitor is never more than Vpeak.. That is the voltage between the live lead and neutral lead is never more than around 350V. This is similar to case 1 in the link where there is no reservoir capacitor. Once you add the reservoir capacitor you should no longer use these 250V AC capacitors across the diode. They may survive as they may well be designed with a margin anyway, but they are being used out of their specification, you should not rely on this margin which may vary between manufacturers and capacitor type and should use a suitable 1000V DC capacitor, this will ensure long term reliability and safety

Any measurements with multimeter are not valid here because of the nature of the waveforms and the meter reading will not give any indication of the peak voltage.
smiler411 is offline  
Old 22nd Oct 2020, 11:59 am   #30
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Quote:
Originally Posted by smiler411 View Post
This site may explain it better than me

https://www.quora.com/Why-is-peak-in...pacitor-filter

If you look at the graphs for the half wave with capacitor you can see that the cathode of the diode is always at a positive voltage that is at or slightly below (depends on load and capacitor size) the peak with respect to 0V. The anode side varies between +peak voltage and -peak voltage with respect to 0V.

So the maximum voltage across the diode is almost 2 X Vp-p as it will be the sum of the peak voltage plus the voltage on the reservoir capacitor. Normal 250V ac capacitors are designed to be fitted across the mains or similar voltages where the voltage across the capacitor is never more than Vpeak.. That is the voltage between the live lead and neutral lead is never more than around 350V. This is similar to case 1 in the link where there is no reservoir capacitor. Once you add the reservoir capacitor you should no longer use these 250V AC capacitors across the diode. They may survive as they may well be designed with a margin anyway, but they are being used out of their specification, you should not rely on this margin which may vary between manufacturers and capacitor type and should use a suitable 1000V DC capacitor, this will ensure long term reliability and safety

Any measurements with multimeter are not valid here because of the nature of the waveforms and the meter reading will not give any indication of the peak voltage.
Hi.
I agree with you here and yes I fully understand how a rectifier with a cap will have a higher voltage across the diode. I removed the video just in case it was misleading. It is possible on the video I was getting confused due to the low value of the smoothing cap at 25v.
I hope to check this out today later on at 230v. I realise reading with a multimeter is rather pointless.
When you get told things at college when a teenager you take them as gospel especially when the lecturer seems to make a very valid discussion. To be fair it never really dawned on me that I and others were running the caps on the cusp.
Thanks again.
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 12:09 pm   #31
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Quote:
Originally Posted by Electronpusher0 View Post
Having read this thread and watched the video I am confused!

Only one thing for it. I'm off to the workshop to replicate Trevors experiment.
Hello Peter, Lawrence and Trevor.

1) I am not confused but I do think this thread is an example of when two experts are looking at the same problem with a different 'lens.' The result can be confusing.

This is a shame as some newcomers could be really at sea.

2) I think Lawrence's contributions repeatedly stress a fact that is relevant and should be at the foreground. Take a look at post #7 when Lawrence politely says...

''...the PIV across the diode [is] VAC peak and if a reservoir capacitor is fitted then the PIV becomes 2*VAC peak under no load conditions?''



3) Trevor, I think your experiment is well-intentioned but a little confusing as it is apparently at 1/10th scale.

Yet you also round down some values instead of to the nearest significant figure. This is important.

The clipping of the sinusoidal waveform is convenient from the viewpoint of your argument as it knocks off the peaks. As it stands, the experiment is inconclusive and confusing.

The experiment should definitely be off-load as off-load is a real-life condition. Again, Lawrence has already said this.

My take on this...

Trevor, you say that 250VAC rated capacitors haven't failed in the past in your experience (which, also by your description includes a long time ago in the repair workshop when these components were far more expensive). This is interesting to note. However, this shouldn't trump current best practice.

The level-shifted voltage stresses amply noted in this thread by Lawrence should be taken into account.

Components are relatively cheap. Seems like a no-brainer to select components that have a margin of tolerance. That would mean 1,000VDC rated capacitors... (Oh and I think we haven't even mentioned fast transients...)


I think novices reading this now or in the future need really clear rules. We don't all need to be heading to the bench to replicate a situation that has seemingly been analysed in production and test labs many times already! I'm not trying to have the last word here but I think it would be helpful if there was one, really clear conclusion of this for others.

Thanks to everyone for staying chilled-out about this...
__________________
Al

Last edited by Al (astral highway); 22nd Oct 2020 at 12:32 pm.
Al (astral highway) is offline  
Old 22nd Oct 2020, 1:00 pm   #32
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Hi chaps.
Thanks to all for the comments.
I think I have to put up my hands and say I made a blunder on the video and am glad I pulled it. I checked again and yes the result was misleading. I removed the load and it didn't add up. I rechecked my probe and scope calibration and it was out quite a fair bit. Then it dawned on me. I checked the calibration yesterday prior to doing the video. The probe got caught though and I nearly pulled the scope off the bench. I obviously damaged something. On rechecking just now with another probe it all falls into place. There is no need to do another video at 230v as not only did I confuse myself but others too which I apologise for.

The advice of using up to date parameters is wise and duly noted.
Thanks everyone, a learning curve and reminder of theory that you forget over time.
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 1:26 pm   #33
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,287
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

My contribution to this discussion was not to doubt best practice which is well established.
I "headed for the bench" to try and replicate and understand the anomolous readings Trevor was getting.

I am happy that the very high ripple voltage was confusing the diode measurements and that when the value of the capacitor is increased, such that the ripple voltage becomes negligable, then the measured voltage across the diode (measured with the scope) is indeed the expected 2 X Vpeak of the transformer secondary (2 X 1.414 X Vrms)

Peter
Electronpusher0 is offline  
Old 22nd Oct 2020, 1:42 pm   #34
murphyv310
Dekatron
 
murphyv310's Avatar
 
Join Date: Sep 2006
Location: Kilmarnock, Ayrshire, UK.
Posts: 5,420
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Two new scope probes on order.
Wow some of them would make you broke at the cost.
__________________
Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member
murphyv310 is offline  
Old 22nd Oct 2020, 2:13 pm   #35
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

Quote:
Originally Posted by murphyv310 View Post

The advice of using up to date parameters is wise and duly noted.
Thanks everyone, a learning curve and reminder of theory that you forget over time.
Thanks, Trevor. That's gracious of you and helpful to the progression of this discussion. I personally appreciate you doing this.

I wonder if mods might link this thread with the one called:-

'Diode switching noise on mains power supplies'?

For example, trobbins points out in post #5 that there are more appropriate ways to handle hash/RFI - e.g by using a snubber across the transformer output.

Also moderator David (RadioWrangler) addresses RFI in his detailed remarks on that thread: - e.g
"There are two issues here. Adding capacitors to prevent diodes blowing up, and adding capacitors to prevent snappy turnoff diodes generating lots of RFI. The same capacitors can do some of both jobs.

Modern high voltage rectifiers are much faster and turn off more cleanly. There is still some RFI but the level is greatly reduced. You probably still want some filtering in the mains path, but RIFAs aside, it's easier.
Seems like a great opportunity to merge the two threads?
__________________
Al

Last edited by Al (astral highway); 22nd Oct 2020 at 2:19 pm.
Al (astral highway) is offline  
Old 22nd Oct 2020, 3:28 pm   #36
Cobaltblue
Moderator
 
Cobaltblue's Avatar
 
Join Date: Dec 2010
Location: Exeter, Devon and Poole, Dorset UK.
Posts: 6,823
Default Re: Discussion on rectifier and capacitor replacing metal rectifier

There has been some very interesting contributions to this thread but as the ground is now covered I will close this one.
__________________
Invisible airwaves crackle with life or at least they used to
Mike T BVWS member.
www.cossor.co.uk
Cobaltblue is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:22 pm.


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.