Odd behavior or normal voltages in heater supply?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I think there is a fundamental flaw in you heater circuit. The dc heaters have one side connected to ground but the tubes fed with ac heaters are floating. However, you cannot float heaters, they have to have a path to 0V so the dc voltage between heater and cathode is controlled. Unfortunately you cannot just connect one side of the ac heaters to 0V as this will short out one of th diodes in the bridge. The only safe way to provide a single 0V connection to all the heaters is from a centre tap on the neater winding. Check out The Valve Wizard's page on this topic:

http://www.valvewizard.co.uk/heater.html

The section on dc heaters is towards the bottom.

Cheers

Ian
 
ruffrecords said:
I think there is a fundamental flaw in you heater circuit. The dc heaters have one side connected to ground but the tubes fed with ac heaters are floating.
They're not floating. They are referenced to half the rectified voltage.

However, you cannot float heaters, they have to have a path to 0V
They do have a path to 0V, via the rectifiers.

Indeed I agree with the fact that only one connection to 0V should be made, but it's not necessary a center-tap on the secondary.
It could be a virtual center-tap, or even one side of the secondary, or any point of the circuit. Some options are better than others, but they are all correct in providing galvanic path.
 
scott2000 said:
I see I may have mis-worded what I was saying.... I meant you have to subtract the voltage drop before multiplying...... not that the voltage is lower before the rectifier.....
..

Thank you sir

I will have a serious look at it today :)

Best regards

/John
 
abbey road d enfer said:
Voltage drop is due to 2 diode in series (1.4V), plus ohmic and inductive loss in the transformer. This must be put in regard with the peak voltage, that is supposed to be 6.3x1.414.
The resistive and inductive voltage loss happens at the secondary and clips the waveform; that reduces the peak voltage.
The diode loss is almost fixed at 1.4V.
Then the smoothing cap tries to fill the gaps between the charging peaks (they happen only when the input waveform is high enough to make the diodes conduct). You never mentioned the value of the smoothing cap, however that's an important factor in the design. Too small a cap results in low voltage and inadequate noise. Anything less than 10 000uF is a waste of money. Now there should be a small resistor in series to drop the excess voltage.
You cannot properly troubleshoot this issue with just a voltmeter, you need an o'scope.
You probably never noticed this 2V drop when you deal with 300V B+.

Thank you sir

Yes, there is a dropping resistor in series, but the problem is that 6.0V before the rectifying bridge is 6.15V after the bridge. The capacitor is 6800uF and it works in several other stuff I have built so I am a bit confused why the voltage is so low and why the rectifying bridge only produce an increase of 0.15V.

Even if the unloaded secondary is 6.0V and with the diode loss of 1.4V you'll get 4.6V x 1,414 =6.5V. Then I would have a tiny voltage to drop with the series resistor.

I have used this layout before in some LA2A's to make them a bit less "humming" and also in three "mono versions" of the same design as this project. I have not seen the secondary low voltage 6.3VAC or 12VAC (in some designs) drop so much voltage that there is nothing left to filter for the DC.

I guess that it would be fairly simple to replace the first capacitors with a 10000uF and see if things changes?


Best regards

/John
 
ruffrecords said:
I think there is a fundamental flaw in you heater circuit. The dc heaters have one side connected to ground but the tubes fed with ac heaters are floating. However, you cannot float heaters, they have to have a path to 0V so the dc voltage between heater and cathode is controlled. Unfortunately you cannot just connect one side of the ac heaters to 0V as this will short out one of th diodes in the bridge. The only safe way to provide a single 0V connection to all the heaters is from a centre tap on the neater winding. Check out The Valve Wizard's page on this topic:

http://www.valvewizard.co.uk/heater.html

The section on dc heaters is towards the bottom.

Cheers

Ian


Thank you sir

I have read that article some time ago, but the design has worked just fine in many projects before. Perhaps a beginners lucky "mistake" as Abbey mention later in this thread? I guess that I can alter the reference to ground by creating a artificial centre tap before the bridge and see if there is a difference?

Best regards

/John
 
abbey road d enfer said:
They're not floating. They are referenced to half the rectified voltage.
They do have a path to 0V, via the rectifiers.

Indeed I agree with the fact that only one connection to 0V should be made, but it's not necessary a center-tap on the secondary.
It could be a virtual center-tap, or even one side of the secondary, or any point of the circuit. Some options are better than others, but they are all correct in providing galvanic path.

Thank you sir

Would it be a difference if using an artificial centre tap on the AC compared to this case where the reference to ground is at the DC?


Best regards

/John
 
... and I must say that the total capacitance is 23000uF spread out between 2 x 6800uF and 2 x 4700uF. Yes, I know it is a bit costly but it usually gives me a very clean DC for the heaters in small tube projects.


Best regards

/John
 
johnheath said:
Would it be a difference if using an artificial centre tap on the AC compared to this case where the reference to ground is at the DC?
I would think using an artificial center tap may make the DC heater line dirtier, which would not be a good thing.
 
johnheath said:
Yes, there is a dropping resistor in series, but the problem is that 6.0V before the rectifying bridge is 6.15V after the bridge. The capacitor is 6800uF and it works in several other stuff I have built so I am a bit confused why the voltage is so low and why the rectifying bridge only produce an increase of 0.15V.
It points out to the transformer; its resistance/inductance is probably too high.


I guess that it would be fairly simple to replace the first capacitors with a 10000uF and see if things changes?
I don't expect changing from 6.8m to 10m to make any serious difference.
 
abbey road d enfer said:
I would think using an artificial center tap may make the DC heater line dirtier, which would not be a good thing.

Thank you sir

No, I have had good result from the design shown in the schematic in the first post so I guess I am good at that.


Best regards

/John
 
abbey road d enfer said:
It points out to the transformer; its resistance/inductance is probably too high.

I don't expect changing from 6.8m to 10m to make any serious difference.

Thanks again sir

No, I don't see how changing the first capacitor would alter things very much. As I said the total capacitance is high as is.

So... you say the transformers resistance/ inductance is too high? The transformer is an American made and also brand new. The company has a fairly high reputation for making "good" transformers... perhaps not top-notch but good.

Is there a way for me to solve the problem with the same transformer (yes, I know I haven't mentioned exactly which one it is) in general? As mentioned the project is a stereo version of a mono version I have made a couple of times. In the mono version I use a toroid transformer with secondary winding of 250VAC and 6.3VAC and it works like a charm.

For this stereo project I had to get another power transformer that could handle the now increased need for current, but it fell flat on the ground.

Best regards

/John
 
scott2000 said:
Where? It's in there now? What value?

Oh..sorry, but I did not include the whole schematic... but there are in fact one inductor and two series resistors and totally four capacitors.

Best regards

/John
 
abbey road d enfer said:
That could have dispelled some doubts...

Well, the thing is that I simply followed a design that usually works for me, but as mentioned it did not in this case. I didn't really believe that leaving a few components out would make all that difference in the analysis about what could be happening with the voltages.

Problem mentioned as: Not getting the expected voltage... before and after the rectifying bridge.

Best regards

/John
 
abbey road d enfer said:
Everything matters in a circuit, even the pilot light.

Yes I know, but since I described the total current draw and the voltages I thought that you could discuss the problem anyway? I guess I was wrong?

Best regards

/John
 
Back
Top