More On The B+ Mod

[John L Stewart]

Following for information only. Your results may differ.

The full-wave center tap rectifier (sometimes called full-wave diametric) on paper appears completely symmetric. But as usual, the devil is in the details.

The PS HV winding is usually wound serially, that is the 2nd half is simply wound over the first after a CT connexion has been brought out. As a result, the 2nd half has a higher resistance than the first half. For the simple low cost PT I’ve used for these tests the halves of the HV secondary measure 189R & 206R.

If you look carefully inside the glass of a directly heated rectifier such as the 5U4, 5Y4, Etc you will see the heaters inside each plate are connected in series to their outside 5 volt supply. So each heater is 2.5 volts at rated current. Better a short high current heater than two long thin 5 volt heaters at ½ the current, both for mechanical strength & the peak current emission required for capacitor input filters.

Altho most windings on PS PTs are color coded to ID their intended use, there is usually no polarity information given. As a result it is possible the heater winding will be connected either series aiding or series opposing to the HV winding, sofar as the rectifier is concerned. It would seem that a 5 volt supply would have little effect while in series with the applied HV. But the p-p voltage difference is 14 volts. The rectifier diodes are conducting only on the peaks of the HV supply into the following cap. The EI characteristic of the diodes is not linear but rather follows the 3/2 power law. That further accentuates the extra 5 volts.

Go to this link & see page 4 for the EI curve og the 5U4 rectifier-

http://frank.pocnet.net/sheets/093/5/5U4GB.pdf

This can result in a large difference in the resulting 60 Hz (50 Hz) component at the output of the rectifier, depending upon which connexion polarity relation occurs. The following filter section, whether LC or RC is not as effective at removing this ripple as for the 120 Hz (100 Hz) it is meant for.

A CT on the 5 volt heater supply goes only part way in respect to the problem described. The test results showed 6 to 8 db improvement over the CT remedy depending on how the circuit was connected.

Another partial fix would be to add an equalizing resistor to the winding half of lower resistance.

But the simple fix requires no other parts at all save the current sampling resistor in the HV CT lead to the first filter cap. Then try reversing to Rectifier heater supply leads for the best result.

All this doesn’t matter a lot if the amplifier is PP, since common mode voltages are for the most part rejected. But the PP is never perfect. With the SA I've seen IM sidebands at 880, 940, 1060 & 1120 Hz when a 1000 Hz test signal was applied to an otherwise good PP Amp.

This simple rectifier fix can help a lot in a SET amp where 2/3 to ¾ of the 60 Hz (50 Hz) can show up in the loudspeaker, possibly at a resonance. Getting the last few mv out near a speaker resonance could make a significant improvement.

Hope this will be useful information for someone.

Cheers, John

[Big Bad Bob]

On 04/10/11 17:16, John L Stewart so wittily quipped:
> Following for information only. Your results may differ.
>
> The full-wave center tap rectifier (sometimes called full-wave
> diametric) on paper appears completely symmetric. But as usual, the
> devil is in the details.
>
> The PS HV winding is usually wound serially, that is the 2nd half is
> simply wound over the first after a CT connexion has been brought out.

yeah, that's quite possibly true for CHEAP transformers. But it's just
as easy to use 2 parallel windings then put 'em in series for
(effectively) a center tap. Even cheapo radio shack transformers (the
ones I've seen) appear do that. No reason why NOT to do that if the
intent is a CT for a full-wave rectifier tube filament (or a low voltage
full wave rectifier even).

> As a result, the 2nd half has a higher resistance than the first half.
> For the simple low cost PT I’ve used for these tests the halves of the
> HV secondary measure 189R& 206R.

etc. - yeah, that kind of imbalance would induce SOME ripple/hum into
the circuit, and require larger capacitors + series inductors to get rid of.

[Phil Allison[_3_]]

"Big Bad Boob"

>
>> The PS HV winding is usually wound serially, that is the 2nd half is
>> simply wound over the first after a CT connexion has been brought out.
>
> yeah, that's quite possibly true for CHEAP transformers. But it's just as
> easy to use 2 parallel windings then put 'em in series for (effectively) a
> center tap.


** So the Boob knows nothing about transformers either.


> No reason why NOT to do that if the intent is a CT for a full-wave
> rectifier tube filament (or a low voltage full wave rectifier even).

** So the Boob did not even read the post he is now crapping on about.

It says " PS HV winding .... "



..... Phil

[John L Stewart]

On 04/10/11 17:16, John L Stewart so wittily quipped:
> Following for information only. Your results may differ.
>
> The full-wave center tap rectifier (sometimes called full-wave
> diametric) on paper appears completely symmetric. But as usual, the
> devil is in the details.
>
> The PS HV winding is usually wound serially, that is the 2nd half is
> simply wound over the first after a CT connexion has been brought out.

yeah, that's quite possibly true for CHEAP transformers. But it's just
as easy to use 2 parallel windings then put 'em in series for
(effectively) a center tap. Even cheapo radio shack transformers (the
ones I've seen) appear do that. No reason why NOT to do that if the
intent is a CT for a full-wave rectifier tube filament (or a low voltage
full wave rectifier even).

> As a result, the 2nd half has a higher resistance than the first half.
> For the simple low cost PT I’ve used for these tests the halves of the
> HV secondary measure 189R& 206R.

etc. - yeah, that kind of imbalance would induce SOME ripple/hum into
the circuit, and require larger capacitors + series inductors to get rid of.

----------------------------------------------------------

Opinions are for lawyers, solicitors & politicians. Big Fat Bob is none of these. So lets leave it to him to present proof that the common PT used in tubed amps really does have equal resistance HV halves.

Bob, pls be sure to state the PT manufactuerer & p/n along with the resistance data so we can cross check.

Cheers to all, John

[Big Bad Bob]

On 04/11/11 16:22, Phil Allison so wittily quipped:
> "Big Bad Boob"
>
>>
>>> The PS HV winding is usually wound serially, that is the 2nd half is
>>> simply wound over the first after a CT connexion has been brought out.
>>
>> yeah, that's quite possibly true for CHEAP transformers. But it's just as
>> easy to use 2 parallel windings then put 'em in series for (effectively) a
>> center tap.
>
>
> ** So the Boob knows nothing about transformers either.

you are still an idiot. you shouldn't be up this late. your mommie
told you to go to bed early, as 6-year-olds need their sleep.

troll harder, your 'fail' level is showing.

[Phil Allison[_3_]]

"Big Bad Boob"
>
>> The PS HV winding is usually wound serially, that is the 2nd half is
>> simply wound over the first after a CT connexion has been brought out.
>
> yeah, that's quite possibly true for CHEAP transformers. But it's just as
> easy to use 2 parallel windings then put 'em in series for (effectively) a
> center tap.


** So the BB Boob knows NOTHING about transformers either.


> No reason why NOT to do that if the intent is a CT for a full-wave
> rectifier tube filament (or a low voltage full wave rectifier even).

** So the BB Boob did not even read the post he is now crapping on about.

It says " PS HV winding .... "



..... Phil

[Phil Allison[_3_]]

"John L Stewart"
>
> Opinions are for lawyers, solicitors & politicians. Big Fat Bob is none
> of these. So lets leave it to him to present proof that the common PT
> used in tubed amps really does have equal resistance HV halves.
>
> Bob, pls be sure to state the PT manufactuerer & p/n along with the
> resistance data so we can cross check.


** Don't hold your breath waiting for that ....



.... Phil

[Big Bad Bob]

On 04/12/11 17:51, Phil Allison so wittily quipped:
> "Big Bad Boob"
>>
>>> The PS HV winding is usually wound serially, that is the 2nd half is
>>> simply wound over the first after a CT connexion has been brought out.
>>
>> yeah, that's quite possibly true for CHEAP transformers. But it's just as
>> easy to use 2 parallel windings then put 'em in series for (effectively) a
>> center tap.
>
>
> ** So the BB Boob knows NOTHING about transformers either.

you are SO full of crap. Troll harder next time. You've invented new
levels of 'fail'.

>
>> No reason why NOT to do that if the intent is a CT for a full-wave
>> rectifier tube filament (or a low voltage full wave rectifier even).
>
> ** So the BB Boob did not even read the post he is now crapping on about.
>
> It says " PS HV winding .... "
>

and I still say you're an idiot. quit wasting bandwidth with your
ad-hominem attacks. And while you're at it grow a brain. There's no
reason why you can't use parallel windings on a power transformer. In
fact, I say it SHOULD be done that way. Apparently, so do THESE guys:

http://www.hammondmfg.com/300Pseries.htm

"concentric wound for for low stray field and low noise"

[Doug Bannard]

"Big Bad Bob" > wrote in message
...
> On 04/12/11 17:51, Phil Allison so wittily quipped:
>> "Big Bad Boob"
>>>
>>>> The PS HV winding is usually wound serially, that is the 2nd half is
>>>> simply wound over the first after a CT connexion has been brought out.
>>>
>>> yeah, that's quite possibly true for CHEAP transformers. But it's just
>>> as
>>> easy to use 2 parallel windings then put 'em in series for (effectively)
>>> a
>>> center tap.
>>
>>
>> ** So the BB Boob knows NOTHING about transformers either.
>
> you are SO full of crap. Troll harder next time. You've invented new
> levels of 'fail'.
>
>>
>>> No reason why NOT to do that if the intent is a CT for a full-wave
>>> rectifier tube filament (or a low voltage full wave rectifier even).
>>
>> ** So the BB Boob did not even read the post he is now crapping on about.
>>
>> It says " PS HV winding .... "
>>
>
> and I still say you're an idiot. quit wasting bandwidth with your
> ad-hominem attacks. And while you're at it grow a brain. There's no
> reason why you can't use parallel windings on a power transformer. In
> fact, I say it SHOULD be done that way. Apparently, so do THESE guys:
>
> http://www.hammondmfg.com/300Pseries.htm
>
> "concentric wound for for low stray field and low noise"
>
>
Well Bob...you're wrong again! Concentric winding and parallel winding are
NOT the same.

In a concentric winding, one winding is wound on top of the other with
insulation separating them. Could be a secondary wound on top of a primary,
or a secondary wound on top of another secondary. This is the usual and
rather conventional construction for an inexpensive power transformer.

A parallel winding occurs when two (or more) strands of magnet wire are
wound together. They may be then connected to form a centre tapped winding
with equal half winding resistances if you like, and this is often done on
low voltage transformers. It is NEVER done on a high voltage secondary and
I can assure you that Hammond would never construct a high voltage
centre-tapped winding in this way.

The words "high voltage" should be your tip-off as to why concentric rather
than parallel construction must be followed.

Balancing the resistance of HV windings is a piece of cake even with
concentric construction, without even changing the wire gauge...just costs a
bit more.

Doug Bannard, VE3SPF

[Big Bad Bob]

>> http://www.hammondmfg.com/300Pseries.htm
>>
>> "concentric wound for for low stray field and low noise"
>>
>>
> Well Bob...you're wrong again! Concentric winding and parallel winding are
> NOT the same.

leave out concentric, concentrate on 'low stray field and low noise'. I
was quoting their page.

[you ESTJ types need to stop circling around minor details]

[Phil Allison[_3_]]

"Doug Bannard"
>
> Well Bob...you're wrong again! Concentric winding and parallel winding
> are NOT the same.

** Kinda obvious really .....


> In a concentric winding, one winding is wound on top of the other with
> insulation separating them. Could be a secondary wound on top of a
> primary, or a secondary wound on top of another secondary. This is the
> usual and rather conventional construction for an inexpensive power
> transformer.
>
> A parallel winding occurs when two (or more) strands of magnet wire are
> wound together.


** A technique also known as " bi-filar " winding.

Much loved by the RF brigade.



> They may be then connected to form a centre tapped winding with equal half
> winding resistances if you like, and this is often done on low voltage
> transformers. It is NEVER done on a high voltage secondary and I can
> assure you that Hammond would never construct a high voltage centre-tapped
> winding in this way.


** For the simple reason that the enamel insulation between adjacent turns
would not stand up to the voltage stress ( same as the voltage of whole
winding ) for very long.


> Balancing the resistance of HV windings is a piece of cake even with
> concentric construction, without even changing the wire gauge...just costs
> a bit more.

** Resistors are cheap.



..... Phil

[Big Bad Bob]

On 04/13/11 07:00, Phil Allison so wittily quipped:
>> Balancing the resistance of HV windings is a piece of cake even with
>> concentric construction, without even changing the wire gauge...just costs
>> a bit more.
>
> ** Resistors are cheap.
>
>

better to wind in parallel, and if high voltage arcing is a concern, put
the two halves on the same later IN PARALLEL such that # of turns AND
inductance AND wiring resistance match. Per layer.

Stop being an ass