Re New quad II from scratch
Andre Jute
The sound of the Quad II depends to a far greater extent than most
people understand on the cathode feedback topology and the general
design which balances out quite extraordinary tube mismatches.
Exotic components simply do not come into it.
I rebuilt several sets of Quad II with exotic components and
eventually sold them all. My current set, saved for my retirement, is
bog-standard.
A replacement parts kit for a rebuild, sent by Acoustical (that's the
manufacturer of Quad) in Huntingdon, currently in my hand, contains
components that I would be careful not to display to audiophiles in
any other amp . . . In particular, polyprops alter the quality of the
sound. Stick to good quality low ESR electrolytics. You also want to
think long and hard before you change the value of any component, as
distinct from its rating. There are subtleties in this amp that are
not instantly obvious that can be ruined by a bigger-is-better
approach.
There is one component the rating of which you must change because it
is underspecified. (No, Walker was not incompetent. Parts availability
after the war was appalling.) R12 should be at least 6W, preferably
10W, rather than 3W as shown.
HTH
Andre Jute
<luc.de...@free.fr> wrote:
> Hello,
>
> I plan to make new quad II for my father (i'm an happy owner of a
> pair).
> I have collected documentation, read lot of discussions, lot of web
> pages,...
> Time has come now to conclude. And i need you comments and advices.
>
> Clearly make a new pair of quads will cost me around 1000•-1100•,
> incl. new tubes. No way to find a kit for those, so i have to do the
> sopping myself.
>
> I plan to buy the transformers from woodside (sowter are not so precise
> copies of the original) link :
> http://freespace.virgin.net/m.davis/homepage.htm
>
> I plan to buy components from Angela http://www.angela.com/ at least
> the capacitors.
>
> For that point i still have some doubts :
> I will buy paper in oil (copper angela branded) for C1 to C3 0.1mF
> What do i choose for C4 C6 16mF, an for C5 25mF ? (black gate?)
>
> For the resistors i doubt, i really don't have the money to buy
> tantalum, so anybody can suggest an alternative ? With the place to buy
> them that would be very appreciated.
>
> also i don't know how much W resistor i have to choose, anybody have
> this information ?
>
> I still have to buy aluminium chassis (any advice ?), sockets,
> connectors,... But i think this is more simple.
> your comments are welcome,
> Luc.
John Stewart
Andre Jute wrote:
> Luc
>
> The sound of the Quad II depends to a far greater extent than most
> people understand on the cathode feedback topology and the general
> design which balances out quite extraordinary tube mismatches.
>
> Exotic components simply do not come into it.
>
> I rebuilt several sets of Quad II with exotic components and
> eventually sold them all. My current set, saved for my retirement, is
> bog-standard.
>
> A replacement parts kit for a rebuild, sent by Acoustical (that's the
> manufacturer of Quad) in Huntingdon, currently in my hand, contains
> components that I would be careful not to display to audiophiles in
> any other amp . . . In particular, polyprops alter the quality of the
> sound. Stick to good quality low ESR electrolytics. You also want to
> think long and hard before you change the value of any component, as
> distinct from its rating. There are subtleties in this amp that are
> not instantly obvious that can be ruined by a bigger-is-better
> approach.
I would certainly agree with that. The Quad, like many other commercially
available amps is designed as a whole, not in pieces. Some folks want to
as an example increase the value of coupling caps. Those chosen by the
designer are chosen to result in stability. A different value often results
in unexpected results. In a multistage amp such as the Williamson we
are already bordering on instability. Larger coupling caps are often
enough to push things over the edge. Then the loudspeaker breaths
very heavily indeed.
Cheers, John Stewart
Patrick Turner
Andre Jute wrote:
> Luc
>
> The sound of the Quad II depends to a far greater extent than most
> people understand on the cathode feedback topology and the general
> design which balances out quite extraordinary tube mismatches.
This may appear to be the case after superficial examination.
I don't doubt the wisdom of the CFB in the output stage,
and it renders the KT66 with triode like gain and thd but with beam tetrode
power.
The balancing out of "quite extraordinary tube mismatches" is only so good.
When these amps are 1/2 way through their tube life,
the tubes all become mismatched, both KT66 have drifted apart,
and the EF86 have also drifted apart, and the result is that thd can be up to
20 dB above the
optimal thd when everything was sparkling new.
I rewired two Quad II earlier this year, and carefully measure the thd,
and I was able to lower it 15 dB in one amp and 10 dB in the other simply
by swapping outputs around, and the EF86, so that optimal 2H distortion
cancelling
occured between the drive and output stages.
Then and only then was I able to witness less than 0.05% thd at any normal
listening levels.
But audio systems are to me the sum of their parts, and so the low thd I was
able to get
in these repaired (and re-engineered ) amps must have helped the owner to
enjoy his music imho.
He sure noticed the difference.
I doubt the otherwise 0.15% thd would affect the sound much though.
There would be a shirtload more thd if there wasn't any NFB, since there
are no triodes used in Quad II amps.
The amp's tubes will slowly drift apart again, maybe I will still be around
for their next service...
A pair of other Quad II where I'd used KT88 in triode in class AB showed the
same
drift towards more 2H after several years of use, since the output tubes have
drifted in their gm,
and thus the 2H currents in each tube do not fully cancell in the OPT.
However, the odd order products with the KT88 and the
all triode LPT and input tube, 12AU7, seemed to be lower than with existing
Quad II,
so I didn't worry that the KT88 AB triode version had a smidgin more thd than
the originals.
There was less applied CFB and global.
>
> Exotic components simply do not come into it.
I have wooed some very critical listeners with bog standard Taiwanese parts.
I don't do it any more, I am reformed of my cheap set ways, and I now
place only Welwyn resistors and Wima MKP coupling caps into the amps I
re-engineer
or build new. I used to use Beyshlag R, but they got taken over by Vishay,
and the local distributor for Beyshlag got swallowed by some US company,
and the price went from 10c for each 1 watt beautifully made resistor to a
whopping
28c in 3 years.
And they won't do 100pcs of each value, it had to be 1,000 pcs minimum.
Hence my switch to Welwyn, and they are a very generous 3/4 watt metal film.
The 2w metal films I use are from WES components here, I don't know who makes
them but they
look very nice, act very blamelessly, so I trustv them with Motzart's music.
>
>
> I rebuilt several sets of Quad II with exotic components and
> eventually sold them all. My current set, saved for my retirement, is
> bog-standard.
>
> A replacement parts kit for a rebuild, sent by Acoustical (that's the
> manufacturer of Quad) in Huntingdon, currently in my hand, contains
> components that I would be careful not to display to audiophiles in
> any other amp . . . In particular, polyprops alter the quality of the
> sound.
Well, that's news to me. Audiophiles beg me to use polypropylene coupling
caps,
although none have begged me for anything else.
Maybe I just haven't lived long enough to know all the audiophiles worth
knowing....
> Stick to good quality low ESR electrolytics. You also want to
> think long and hard before you change the value of any component, as
> distinct from its rating. There are subtleties in this amp that are
> not instantly obvious that can be ruined by a bigger-is-better
> approach.
If one wants only to restore these amps, then staying to original R values is
essential,
since the balanced working of the EF86 depend on the values chosen.
The common cathode 180 ohm R is deliberately allowed to run hot.
I guess if a tube goes haywire, and becomes saturated with 300 mA
in a bias fault condition, one would want that cathode R to burn out,
rather than the fragile thin wire of the expensive OPT.
>
>
> There is one component the rating of which you must change because it
> is underspecified. (No, Walker was not incompetent. Parts availability
> after the war was appalling.) R12 should be at least 6W, preferably
> 10W, rather than 3W as shown.
Yes, its amazing that Walker did get such build quality into his amps.
Probably there were many rows with metalworkers and suppliers before
Walker would accept what they wanted to foist onto him....
But the only 3W R is the common cathode R.
with 140 mA, it runs at 3.5 watts, but its one of those fairly well made
"3W" R that was a little underated.
Nothing really bad happens if it burns out, and may save an OPT in doing so.
Maybe Walker knew this....
Let us give him the benefit of the doubt.
I would have to say that anyone now wanting to privately
copy Quad II amps has quite some measure of
devotional zeal indeed.
Patrick Turner.
luc.de...@free.fr
In fact i am in the "don't change nothing" posision for capacotirs and
resistors.
I have now another problem :
Woddside is not available to do the transformers at the moment (selling
business ?), and Sowter says that they are not doing perfect replicas
(instability ?).
Who knows where to buy good quad transformers at reasonable price ?
Thank's
Luc.
Patrick Turner
luc.de...@free.fr wrote:
I sure don't, but I would roll my own if I was suitably bribed
to perform the task.
I'd try to find a way of improving on the horrors of the Quad OPT
with its high winding losses when connected for 8 ohms.
Damn it, there is no 4 ohm connection.
And the method of changing Z involves different current densities
in windings, not good practice....
Using a non potted transformer which is the same overal size as Quad's
potted original would allow the core to be just a little larger, and the
wire dia,
and the turns count and interleaving to be more effective.
In fact if I was forced to make a Quad II amp as a jail sentance for
openly criticizing Quad II and Good British Products then I'd sure be able
to make a pair.
I'm presently on bail for charges and awaiting a trial by jury, but finding
12 good men
and women is proving difficult.
I'd beaver away for months, and presto, there they would be, glowing
gloriously
on the bench for all the jailors to enjoy, all painted up like nice grey
seductive audio tarts they'd be.
( The jails are still able to obtain a supply of grey WW2 navy surplus
paints from the Navy )
But one thing would be wrong ( tee hee!! ) and the overall dimensions
would be 1" larger all over, scaled up in size from the original toy
dimensions of Quad II.
Hauled before the judge yet again, I will claim that the scale ruler given
to me by
a visiting friend to Her Majesty's Inn was of East Polish manufacture
which still used an ancient
obscure measurement unit.
Well, it migh fail, that defence, even after the Polish helped the Poms in
WW2,
and the buggers will hang me anyways.
But such is life, as Ned Kelly said when they hanged him.
But two proper "Quad" amps will remain on display at the jail museum,
way down the back where folks seldom wander to visit.
Patrick Turner.
shiva
"Patrick Turner" <in...@turneraudio.com.au> wrote in message
news:41C0FA1B...@turneraudio.com.au...
Jeesh, I'm re -re doing a quad2 *preamp*, this time with a real phono stage
& *truly* justified toss of the "pushbutton radio" switches into the trash
(speakin' figuratively here - they're still with me - if anyone need them
(they work about as good as they always do, and most of the plastic buttons
are there), pay for the shipping... So many neat things 'bout those preamps
(tiny, nice knobs, clean layout) - but the RIAA sec & use of EF86's is
confusing. Can someone explain the RIAA section in those preamps -
I'mhaving a hard time understanding how it *could* sound like a RIAA curve
(It's one of the worst phono preamps I've heard. The tone section's just
killah, though, since I do play with tone controls... And, hey, should I
really keep that triple 16Mf cap? Is that a part of "the grand scheme", or
can I let the two channels be a bit more isolated? really wondering, 'coz
this is the second time I'm re-doing this thing... Thus far the ef86's are
triode-strapped, fils are DC regulated, B+ is 330Vwith the thing warmed up
(external power supply, no HT regulation, but there's room for that, if need
be), and the switch just carries a small DC signal, so no AC in the power
cable. Any advice /explanations of the (seemingly non) RIAA section?
Thanks...
-dim
Phil Allison
"Patrick Turner" = autodidactic QUAD hater and congenital LIAR !!!
>
>
> I'd try to find a way of improving on the horrors of the Quad OPT
> with its high winding losses when connected for 8 ohms.
** The Quad Mk2 winding resistance is 1.6 ohms when set for 16 ohms and is
0.85 ohms when set to 8 ohms.
So 10 % in both cases.
The Turneroid is a fuckwit.
> Damn it, there is no 4 ohm connection.
** Damn it - there is.
R goes to P and T goes to Q.
Resistance will then be about 0.3 ohms.
> And the method of changing Z involves different current densities
> in windings, not good practice....
** Pedantic Turneroid horse manure.
.............. Phil
Phil Allison
"Phil Allison"
Correction:
The Turneroid puked :
>
> Damn it, there is no 4 ohm connection.
>
** Damn it - there is.
The "Q" link on the OP tranny is a 4 ohm tap.
Gives 15 - 18 watts at 4 ohms, no problem.
........... Phil
Patrick Turner
Phil Allison wrote:
> "Patrick Turner" = source of much interesting info....
> >
> >
> > I'd try to find a way of improving on the horrors of the Quad OPT
> > with its high winding losses when connected for 8 ohms.
>
> ** The Quad Mk2 winding resistance is 1.6 ohms when set for 16 ohms and is
> 0.85 ohms when set to 8 ohms.
>
> So 10 % in both cases.
10% is consumer electronics grade winding losses for PP circuits.
However, I measured 17% winding losses when set for 8 ohms....
>
>
>
> > Damn it, there is no 4 ohm connection.
I have never seen any reference to the 4 ohm connection,
so assumed it was not recommended.
>
>
> ** Damn it - there is.
>
> R goes to P and T goes to Q.
>
> Resistance will then be about 0.3 ohms.
>
> > And the method of changing Z involves different current densities
> > in windings, not good practice....
>
>
>
> .............. Phil
I don't think Phil has the facts quite right about the secondary windings on the
Quad II
OPT.
He certainly fails to tell us of the facts, and just says I am wrong, which is
typical Phil.
There is a schematic of Quad II at
http://www.geocities.com/ResearchTriangle/Lab/6722/quadiicirb.html
I happen to have a Quad II amp on the bench as we speak,
and I just measured the unloaded voltages at the secondary windings when
I set up the signal to get 10.0 vrms output when set for 16 ohms, as the
schematic shows.
The voltage from P to Q was 5 vrms, and from Q to the link joining R&S the
voltage was 2.5 vrms.
From the link at R&S to T, the voltage was 2.5 vrms.
So the winding from T to R has 2.5 v, and the S to Q part of the winding SQP
also has 2.5 v.
So thus when the output is taken from point S and the winding TR is strapped
across
SQ, as it is for 8 ohms, there are two parallel windings of equal turns with 2.5
v in series with the
QP portion of SQP winding which has 5vrms.
The match is not really to 8 ohms when connected as such, but really to 9 ohms.
The connection of the TR winding across the QP section of SQP is ILLEGAL,
since you would have two windings where one has 1/2 the turns of the other, the
other,
and serious transformer shunt currents would flow if the OPT was connected
this way as Phil suggests it can be, and this would gravely overload the output
tubes..
It simply should never be done, and sorry Phil, I cannot agree that you
are correct about Quad this time.
Indeed the secondary winding resistance when connected for 16 ohms as the
schematic above shows
does in fact measure 1.6 ohms.
The Quad II OPT losses do not only include the secondary, but also the
reflected primary winding losses, themselves perhaps 7%, I am not sure at this
time.
But last time I measured the winding resistance total, it was 17% of the anode
to anode RL,
so for 15 watts of output there is 2.55 watts lost as heat in the OPT.
This is "toy" quality, or consumer grade quality. Good OPTs have losses
less than 6%.
The measured resistance for a '16 ohm' winding should be 0.48 ohms,
which is 3% of the Sec RL.
The primary DCR including any tertiary cathode windings should be less than
120 ohms where the a-a load is about 4k, as it is for Quad II.
I recall the DCR to be a lot more than 120 ohms, and its because of the
terribly thin wire used, which leads to early OPT failure if ever an output tube
becomes saturated for some considerable time.
Quad would have had to make they Quad II amp about 30% larger all over
dimensionally
to achieve lower OPT wind losses.
It works OK, the Quad II, but it does not represent the pinnacle of what can be
achieved with
tubes, wire, and iron.
Patrick Turner.
Patrick Turner
Phil Allison wrote:
The winding from T to R cannot be strapped across the QP portion of the
SQP winding, and I have outlined the reasons in my post of a few minutes
ago.
Indeed the Q point can be used for a 4 ohm connection.
But that means the S to Q part of SQP winding and the TR winding will
have to remain unused.
Therefore the total winding losses will indeed be awful with 4 ohms
connected as you suggest.
Sure there is some power to be had, but perhaps 25% of it is wasted as
heat in the OPT
P&S windings with a 4 ohm connection.
Nobody much used 4 ohms for speakers in 1955.
Walker didn't want to confuse ordinary folks
with a properly done lot of secondaries with many more necessary
connection points
which give the same winding losses and leakage inductance referred to
the primary
regardless of whether connected to
4, 9, or 16 ohms.
Walker simply tried to give ppl 8 or 16, rather than 4 and 16, which is
easier to do
with just two windings, and without current density and losses problems.
There are virtually no OPTs with exactly 4, 8 and 16 ohm load matches.
Patrick Turner.
Patrick Turner
shiva wrote:
The Quad 22 switch bock is a reliable switching module and I have never seen
one which was intermittent, so why not retain it?
> So many neat things 'bout those preamps
> (tiny, nice knobs, clean layout) - but the RIAA sec & use of EF86's is
> confusing. Can someone explain the RIAA section in those preamps -
> I'mhaving a hard time understanding how it *could* sound like a RIAA curve
> (It's one of the worst phono preamps I've heard.
The RIAA is partially achieved with passive eq and shunt NFB between
the g1 and anode of the EF86.
When you draw up the actaul circuit, all should become clear; its really very
basic.
Its also prone to suffering drastic loss of bass and treble when the R&C
components used drift
a lot over the 50 years since they were made.
Also the Quad 22 phono amp is really only fit for
high output MM; forget MC, unless you have a step up transformer, in which case,
its OK.
The line stage is a two stage amp using 2 halves of 12AX7,
and has a lot of NFB, and a complex way of incorporating filtering
and tone control, all of which can be easily improved upon.
> The tone section's just
> killah, though, since I do play with tone controls... And, hey, should I
> really keep that triple 16Mf cap?
Separate R feeds to say a pair of 100 uF caps won't hurt.
Maybe even a separate B+ PS.
The line stage is not all that wonderful imho, and almost anything is better.
Here is a pic of a Quad 22 control unit I have...
http://www.turneraudio.com.au/webpics02/quad22coverOff450w286h.jpg
> Is that a part of "the grand scheme", or
> can I let the two channels be a bit more isolated? really wondering, 'coz
> this is the second time I'm re-doing this thing... Thus far the ef86's are
> triode-strapped, fils are DC regulated, B+ is 330Vwith the thing warmed up
> (external power supply, no HT regulation, but there's room for that, if need
> be), and the switch just carries a small DC signal, so no AC in the power
> cable. Any advice /explanations of the (seemingly non) RIAA section?
> Thanks...
Triode strapping of the phono stage should prevent the tube
from providing anything like the wanted RIAA eq, since the EF86
needs to have a lot of gain for this circuit imho.
But you could replace it with a two stage RIAA circuit using a 12AX7, like
http://www.turneraudio.com.au/htmlwebpgs02/schemsimplephono.htm
Patrick Turner.
>
> -dim
Phil Allison
"Patrick Turner" = fucking LIAR !!!!!
>
> Phil Allison wrote:
>>
>> Correction:
>>
>> The Turneroid puked a MASSIVE lie !!!
>> >
>> > Damn it, there is no 4 ohm connection.
>> >
>>
>> ** Damn it - there is.
>>
>> The "Q" link on the OP tranny is a 4 ohm tap.
>>
>> Gives 15 - 18 watts at 4 ohms, no problem.
>
>
>
> But that means the S to Q part of SQP winding and the TR winding will
> have to remain unused.
>
> Therefore the total winding losses will indeed be awful with 4 ohms
> connected as you suggest.
** Rated power is available into a 4 ohm load.
The Turneroid autistic is psychotic about Quad.
And damn near every other thing too.
.............. Phil
Phil Allison
"Patrick Turner" = fucking LIAR !!
>
>> >
>> > with its high winding losses when connected for 8 ohms.
>>
>> ** The Quad Mk2 winding resistance is 1.6 ohms when set for 16 ohms and
>> is
>> 0.85 ohms when set to 8 ohms.
>>
>> So 10 % in both cases.
>
>
>
** The amp works exactly as advertised.
I have a Quad Mk 2 on my bench - my figures are correct.
The Turneroid just plucks them out his arse.
>> > Damn it, there is no 4 ohm connection.
>
> I have never seen any reference to the 4 ohm connection,
> so assumed it was not recommended.
** The Turneroid presumes too bloody much.
Judging things by irrelevant criteria is his forte.
The forte of a fuckwit.
.............. Phil
Patrick Turner
but no need to worry folks, its just Phil A.
up to his usual boring tricks.
But this reply from Phil completely snipped out all my content in a post I made
at 12.51 am Oz time
where I had a lot to say about the inadequacies of Quad II engineering
standards.
I also pointed out Phil's gross error in advising everyone that its OK to
parallel
the TR winding with the QP portion of the SQP winding; these windings
are clearly shown on Quad schematics, but I have never seen a reference to
connecting
Quad II secondaries the way Phil advised us to do.
He is very quick to accuse other where they snip out bits of his posts,
but here he does it himself, and the word hypocrite comes to my mind.
He completely refuses to address the issues I raised, but sinks into
a tirade of personal insults probably due to inadequate medication for his
bi-polar condition.
I don't claim I am right on this issue. I just related what I have measured,
and conveyed what I think is relevant in today's world where some of us
enjoy or practice standards better than those achieved by Quad II.
I would welcome any proof that my measurements are incorrect.
It is possible to use any speaker with an impedance of 4 ohms or above
when using the connection offered between the Q terminal and ground.
But I repeat that the winding losses are high simply because 1/2 or a large
portion of the
secondary winding wire is not used when using a 4 ohm load across QP.
Even if we note that Phil's figure for the 8 ohm setting
of Quad II gives Rw = 0.85 ohms, that means the sec losses are over 9.6%.
But then we have the dcr of the primary, and the winding losses there are
similar
to the sec, and it explains my calculation of total winding losses of 17%.
The winding losses are the equivalent of a series resistor between
the amplifier and the output, but it is enclosed by the global NFB loop
which tries to reduce the effect of this resistance along with reducing the
effective plate resistance of the tubes, so that Ro of the amp
is around 1 ohm, despite the high winding resistance, depending which impedance
match one selects.
Below Phil says he has a Quad II on the bench, but he fails to measure it
like I did last night like I did with the one I have on my bench.
Phil would do well to stick to the topic, leave personality judgements
out of arguments, and get on with trying to be be rational,
and if he wants to be believed, then he must present what seems to be a cohesive
and detailed argument rather than a childish tantrum.
Patrick Turner.
Phil Allison
"Patrick Turner" = something non human.
> I also pointed out Phil's gross error .....
** The minor oversight was corrected by me only 9 minutes after making it
!!!!
** Here is just one of the Turneroid's LIES
> Damn it, there is no 4 ohm connection.
** Damn it - there is.
The "Q" link on the OP tranny is a 4 ohm tap.
Gives 15 - 18 watts at 4 ohms, no problem.
> He completely refuses to address the issues I raised,
** MASSIVE LIE
I totally refuted the Tunneroid's psychotic anti Quad LIES !!
> I don't claim I am right on this issue.
** The Turneroid sub human does claim he is right:
" I'd try to find a way of improving on the horrors of the Quad OPT
with its high winding losses when connected for 8 ohms. "
" Damn it, there is no 4 ohm connection."
> I would welcome any proof that my measurements are incorrect.
" ** The Quad Mk2 winding resistance is 1.6 ohms when set for 16 ohms
and is
0.85 ohms when set to 8 ohms. So 10 % in both cases. "
> It is possible to use any speaker with an impedance of 4 ohms or above
> when using the connection offered between the Q terminal and ground.
** Rated power is available into a 4 ohm load.
The Turneroid autistic is totally psychotic about Quad.
And damn near every other thing too.
> Even if we note that Phil's figure for the 8 ohm setting
> of Quad II gives Rw = 0.85 ohms, that means the sec losses are over 9.6%.
> But then we have the dcr of the primary, and the winding losses there are
> similar to the sec, and it explains my calculation of total winding losses
> of 17%.
** The loss of power in the OT is trivial.
With audio program it would never exceed 300mW on average.
The Quad Mk2 amp meets all its specs.
End of story.
> Below Phil says he has a Quad II on the bench, but he fails to measure it
> like I did last night like I did with the one I have on my bench.
** The Turneroid presumes far too bloody much.
Judging things by irrelevant criteria ( his own ) is his forte.
The forte of a fuckwit.
The Turneroid fuckwit would do well to stick to the topic, leave personal
judgements out of arguments, and get on with trying to be rational, and if
he wants to be believed, then he must present what seems to be a cohesive
and detailed argument rather than a childish anti Quad tantrum.
But the Turneroid never will - because he is quite insane.
............. Phil
Patrick Turner
Phil Allison wrote:
> delete insult.
>
> > I also pointed out Phil's gross error .....
>
> ** The minor oversight was corrected by me only 9 minutes after making it
> !!!!
You didn't tell everyone that what you'd said 9 minutes earlier was plain wrong.
You just said use the Q terminal.
Here are our posts :-
> > Damn it, there is no 4 ohm connection.
> >
>
> ** Damn it - there is.
>
> The "Q" link on the OP tranny is a 4 ohm tap.
>
> Gives 15 - 18 watts at 4 ohms, no problem.
>
> ........... Phil
The winding from T to R cannot be strapped across the QP portion of the
SQP winding, and I have outlined the reasons in my post of a few minutes
ago.
What you should have said said was, " I made a mistake,
sorry folks, don't ever connect the TR winding across the PQ winding"
I hope nobody amoungst the hundreds of Quad users don't get their wires crossed
as result of what you said.
I said there wasn't a load match for 4 ohms, but only because it is to me a
lousy load match because it is so lossy. Sure the QP winding is a match to 4
ohms,
but the *total* winding losses are high
because not all the available turns in the transformer are optimally utilised.
Much higher than 10%.
I don't know anyone who uses Quads set to 4 ohms.
If an 8 ohm speaker is used on the 4 ohm setting, the losses decrease.
The load then seen by the KT66 tubes would be 8k instead of 4 k,
and the distortion will be a lot lower, because the output tube gain is higher
and
the FB has more effect. But power output is lower.
So if you want fidelity at the expense of lower
power, then that's a way to get it.
One could also set up Quad II to suit 8 ohms, but then use the SQ part of the
SQP winding
which will have the TS winding also in parallel and get a match to 0.88 ohms.
But losses will be even higher than with the QP 4 ohms match.
A far better way to set up the output transformer secondaries is at
http://www.turneraudio.com.au/webschema02/optschemano1550w390h.gif
The 6 windings in four layers give secondary windings where there are
4 paralel windings of 57 t = say 4 ohms,
3 parallel windings of 76 t = 7.1 ohms,
2 parallel windings of 114 t = 16 ohms.
The % losses are the same for each load impedance,
and HF response remains unchanged.
Walker must have thought it too costly to adhere to the better practices known
at the time for transformer construction, and described in RDH4. Better he
provide something ordinary
folks could carry around in a suitcase than try to succeed with a
perfectionist's amplifier.
Leak also dumbed down their transformers, and so did many others.
Competition wrecks many a fine principle.
Patrick Turner.
John Stewart
I think what is being discussed here comes under the heading of 'Utility
Factor'. How well used are the various windings of the transformer? Very
important when dealing with rectifiers, but here as well.
At first look the Quad OPT connexion options look very much the same as
what Hammond does today. Only a minor part of the secondary is left out of
the circuit while connected for 4 or 16 ohms. I have some of their 60 watt,
1600 Series here so when I get some time I will make 4-terminal
measurements on the various windings to get a handle on what the losses
will be.
See http://www.hammondmfg.com/1608.htm
Cheers, John Stewart
Phil Allison
"John Stewart"
> Phil Allison wrote:
>> ** Rated power is available into a 4 ohm load.
>>
>> The Turneroid autistic is psychotic about Quad.
>>
>> And damn near every other thing too.
>>
>> .............. Phil
>
> I think what is being discussed here comes under the heading of 'Utility
> Factor'.
** Wrong.
The subject is the Turneroid's insane dogma and irrational prejudices.
The result of an inflexible mind besotted with self inflicted rote learning
and possessing no technical insights.
............. Phil
Phil Allison
"Patrick Turner" = an evil, criminal LIAR.
>
>
> Phil Allison wrote:
>>
>> > I also pointed out Phil's gross error .....
>>
>> ** The minor oversight was corrected by me only 9 minutes after making
>> it
>> !!!!
>
> You didn't tell everyone that what you'd said 9 minutes earlier was plain
> wrong.
** I used the heading word "correction " then posted the facts.
> What you should have said said was,
** The Turneroid is fucking pig.
> I said there wasn't a load match for 4 ohms,
** MASSIVE LIE !!!!!!!!!!!
" Damn it, there is no 4 ohm connection "
> but only because it is to me a
> lousy load match because it is so lossy.
** To the insane Turneroid shit is chocolate and chocolate is shit.
> I don't know anyone who uses Quads set to 4 ohms.
** LOL .
Fuck off you evil bloody imbecile .
.............. Phil
Patrick Turner
Phil Allison wrote:
Wrong again Phil.
We are discussing Quad amplifiers.
You are talking nonsense.
Patrick Turner.
Phil Allison
"Patrick Turner"
Phil Allison:
>>
>> The subject is the Turneroid's insane dogma and irrational prejudices.
>>
>> The result of an inflexible mind besotted with self inflicted rote
>> learning
>> and possessing no technical insights.
>
>
> Wrong again Phil.
** The subject * IS * the Turneroid's insane dogma and irrational
prejudices.
The result of an inflexible mind besotted with self inflicted rote
learning
and possessing no technical insights.
> We are discussing Quad amplifiers.
** The subject * IS * the Turneroid's insane dogma and irrational
prejudices.
The result of an inflexible mind besotted with self inflicted rote learning
and possessing no technical insights.
> You are talking nonsense.
** Everything * REAL * is nonsenss to the Turneroid as a result of his
inflexible mind besotted with self inflicted rote learning and possessing no
technical insights.
A geriatric, RDH4 regurgitating, autistic bloody bricklayer !!
............. Phil
Jon Yaeger
in...@turneraudio.com.au wrote on 2/15/05 11:38 PM:
But Patrick, if you forget your meds and swear enough and have a temper
tantrum and make an ass of yourself, it magically becomes true!
I thought you knew . . .
Jon
John Byrns
<in...@turneraudio.com.au> wrote:
> I don't think Phil has the facts quite right about the secondary windings on
> the Quad II OPT.
Hi Patrick,
I think you hit the nail on the head with that! I can't say I understand
what may have possessed Phil to resurrect this old thread after nearly two
months, I guess he wanted to stir up a little controversy so he could call
people names?
> He certainly fails to tell us of the facts, and just says I am wrong, which is
> typical Phil.
How True!
Agreed.
It was stated earlier that the total DC resistance of the windings when
connected in series for the 16 Ohm connection is 1.6 Ohms. I am curious,
did you measure the resistance of the individual sections to get the
partial resistances from P-Q, Q-S, and R-T? I am wondering if the same
gauge of wire was used in all the sections, or if the P-Q section was
wound of heavier wire than the two shorter sections.
Regards,
John Byrns
Surf my web pages at, http://users.rcn.com/jbyrns/
Phil Allison
"Jon Yaeger" = a NG zombie
** No functioning brain - but still moves.
........... Phil
Patrick Turner
John Stewart wrote:
I have never cut open a Quad II OPT or and Hammond OPT.
Exactly what they have done inside remains concealed to my inquiring mind.
But both share the same poor technique of offering various load matches because
winding losses are unnecessarily high, current densities in windings are not
uniform for all selected load
matches, and leakage inductance will vary depending on which load match is
selected.
The Quad and Hammond OPT winding layout is the principle reason why
I would never buy either type of transformer for myself, because I can do a
better for myself by myself.
For those not so dedicated to building amplifiers better than those made by
Quad, Leak, Dynaco, Radford, Luxman, and just about every other brand on the
market now or from the past,
the Hammond solution is an excellently cost effective one, and I wish them
well.
I have never used a Hammond, although I at present do have a SET transformer
set given to me by a customer
for his revised 2A3 amps. Without NFB, I expect them to be just fine.
With NFB, I am not so confident......
I will get the chance to discover the facts about these types at least.
I have always believed the correct way to do it is to have all windings fully
utilised at all times.
Doing it my way as mentioned at my website means that there are many more
connections to
strap correctly for the main 3 different load matches available, 4.5 , 8.0, and
18 ohms,
or any other set of 3 loads with the same relative values.
Hammond and Quad underestimated the capacity for people to get more than one or
two link
changes correct when changing load matches.
They also like keeping costs low by using simpler winding geometry and less
winding sections.
Its lowest common denominator engineering.
Those that believe in tapped or unused portions of windings believe in
mediocrity.
A colleague in Sydney says that when using Hammond OPTs in PP amps,
it is harder to make the amp unconditionally stable when global NFB is used.
Some of the transformers wound in the past by Partridge and Sowter,
and here in Oz by A&R and others make the Hammond/Quad techniques look crude
and rude.
Patrick Turner.
Phil Allison
"John Byrns" = criminal scum bag
"Patrick Turner" = Quad hating autistic bricklayer scumbag
** Watch while they publicly perform mutual arse licking.
How utterly disgusting.
........... Phil
Patrick Turner
> But Patrick, if you forget your meds and swear enough and have a temper
> tantrum and make an ass of yourself, it magically becomes true!
>
> I thought you knew . . .
Its alright, we have smooth level flight, all 4 engines humming,
clear skies ahead, and moving on with life.
I dunno about the other guy, I think he hit a mountain....
BTW, you get to see that movie 'Aviator', about Howard Hughes?
Patrick Turner.
>
>
> Jon
Patrick Turner
John Byrns wrote:
I have never stripped a Quad II transformer down.
Pehaps your questions could be answered just by measuring what is there.
But its easy to measure the total winding resistance
with the amp functioning with a sine wave.
Use a 400 Hz signal voltage without a load with say 50 vrms a-a.
There will be some signal voltage at the cathodes = to about 11% of anode voltage.
Anyway, measure the total voltage a-k of the two tubes, it will be about 112 vrms.
Measure the unloaded secondary voltage.
Then connect a load.
Use 16 ohms connected to whatever load match is set up,
and do not allow the thd to exceed 1 %.
Increase the amp signal to get the same V a-k for the unloaded condition.
Measure the output voltage with the load, and this should be less than the
unloaded voltage.
Work out the load current.
Work out the difference between unloaded and loaded voltages.
The total P+S winding resistance seen at the S = Vdiff / load current.
Total winding loss % = 100 x above Rw / ( RL + Rw).
The winding losses are the % of power lost out of the total power from the tubes.
Phil is trying to make a big deal of my mistake that there
was no load match for 4 ohms.
There is actually a match, but it is lossy.
a pair of KT66 with 10% CFB as in Quad II,
and when dissipating 22 watts each for class A can just make 20
watts of pure class A at 45% plate efficiency.
But not when using a Quad II transformer when set to 4 ohms.
Its more like about 16 watts.
Better than no watts at all.
Its enough power for a lot of ppl, but I prefer 50+ watts AB1 from a pair of KT88,
and using an OPT with 5% losses.
The Quad 4 ohm connection would be good for somebody with sensitive
speakers.
When connected with say 8 ohm speakers, the effective winding
losses halve over whatever they are with 4 ohms, but the maximumm po is limited to
say 10 watts.
Patrick Turner.
Phil Allison
"Patrick Turner" = congenital, autistic LIAR !!!
>
> But its easy to measure the total winding resistance
> with the amp functioning with a sine wave.
** It is much simpler to measure the secondary resistance values for
various settings with a DC ohm meter.
They are, as previously stated, 1.6, 0.85 and 0.6 ohms for the 16, 8 and 4
ohm settings.
The primary resistance is obviously invariant.
> Phil is trying to make a big deal of my mistake that there
> was no load match for 4 ohms.
" ** Damn it - there is.
The "Q" link on the OP tranny is a 4 ohm tap.
Gives 15 - 18 watts at 4 ohms, no problem. "
** Err - where am I making a big deal ?????
> There is actually a match, but it is lossy.
** More Turneroid bloody criminal lies.
> The Quad 4 ohm connection would be good for somebody with sensitive
> speakers.
** My Quad Mk 2 ( s/n 8734) delivers 21 watts into 4 ohms at 0.22% THD @
1kHz.
It gives 15 watts ( ie rated power ) with 0.11 % THD.
Set to 8 ohms it gives 22 watts at 0.23% THD.
Once again, the TURNEROID tells more and more LIES
!!!!!!!!!!!!!!!!!!!!
> When connected with say 8 ohm speakers ..... 10 watts ....
** Hoooooold on a minute - that is a deliberate 2:1 mismatch.
More criminal lies and deliberate distortions from the evil Turneroid. The
underlying problem is the Turneroid is full of insane dogmas and irrational
prejudices - the result of an inflexible mind besotted with self inflicted
rote learning and possessing no technical insights.
Long may the public menace rot in hell.
............ Phil
John Stewart
luc.de...@free.fr wrote:> Ok the conclusion is, Walker did an amazing good job.
> In fact i am in the "don't change nothing" posision for capacotirs and
> resistors.
>
> I have now another problem :
> Woddside is not available to do the transformers at the moment (selling
> business ?), and Sowter says that they are not doing perfect replicas
> (instability ?).
>
> Who knows where to buy good quad transformers at reasonable price ?
> Thank's
> Luc.I sure don't, but I would roll my own if I was suitably bribed
to perform the task.I'd try to find a way of improving on the horrors of the Quad OPT
with its high winding losses when connected for 8 ohms.
It is presumptuous of you that you would criticize Quad while freely using their cathode FB output circuit. Got a better idea?
Well Patrick, we all have some 20-20 hind sight. It's cheap & certainly easy for us at this late date to criticize what has been done in the past by others. Then throw money & time at the problem & we are apt to make an improvement (or not). If Quad & others like Hammond & Sowter had found an economical way to combine all of 4, 8 & 16 ohm taps while using all the copper in the secondary winding all of the time they would have shown us how by now. Sowter manages to do that by using a 4-winding secondary, but their cost is more than double that of a similar Hammond. But even while using a 4-winding secondary the outcome is not perfect. What should be the 8 ohm connexion actually results in 10 ohms.
And Hammond managed to do theirs & deliver a product to the market at a reasonable price point, something overlooked by many. Something called 'Business Acumen', I guess. The secondary winding as used by Quad appears to be the same as found in the Hammond 1600 Series OPT's, so probably manages 4, 8 & 16 ohms without difficulty.
In some of your posts you mention the possible difficulty of getting stability while using NFB with the OPT's of others. A quick look at your circuits shows them to be festooned with many phase correction networks. It seems your OPT's may be in that same category as those you have criticized.
Now, here are the numbers for the secondary resistance for a Hammond
60 watt OPT.
While connected for 4 or 16 ohms, 86% of the secondary copper is used.
The 8 ohm connexion uses all of the secondary copper.
For the 16 ohm connexion the resistance in the used windings in series
with the load is 388 milliohms.
For the 4 ohm connexion the resistance in the used windings in series
with the load is 96.6 milliohms.
For the 8 ohm connexion the resistance in the used windings in series
with the load is 178 milliohms.
All measured by applying a constant current & measuring the resulting voltage with a precision DMM.
Before you dismiss this as shoddy you had better to be ready to match it or do better for 88.64USD a copy & still make a profit sufficient to stay in business.
The Sowter U066 is especially interesting since cathode FB similar to that used in the Quad is possible. To me it looks like it could be substituted into a Quad if there was no other alternative available. I almost bought one a few years ago for experimental use since it has many windings & possibilities on the primary.
See the U066 at http://www.sowter.co.uk/
Damn it, there is no 4 ohm connection.
And the method of changing Z involves different current densities
in windings, not good practice....
Please tell us why in 50 words or less. And how much different are the current densities in your opinion?
Patrick Turner.Cheers, John Stewart
John Byrns
<in...@turneraudio.com.au> wrote:
> > It was stated earlier that the total DC resistance of the windings when
> > connected in series for the 16 Ohm connection is 1.6 Ohms. I am curious,
> > did you measure the resistance of the individual sections to get the
> > partial resistances from P-Q, Q-S, and R-T? I am wondering if the same
> > gauge of wire was used in all the sections, or if the P-Q section was
> > wound of heavier wire than the two shorter sections.
> >
> > Regards,
> >
> > John Byrns
>
> I have never stripped a Quad II transformer down.
> Pehaps your questions could be answered just by measuring what is there.
True, all that is needed is a simple DVM to measure the DC resistance of
each of the secondary sections. I have the DVM, but not the QUAD II.
> But its easy to measure the total winding resistance
> with the amp functioning with a sine wave.
Using a sine wave to do a simple job like measuring the DC resistance of
the secondary winding sections sounds like overkill to me!
Regards,
John Byrns
John Byrns
<phila...@tpg.com.au> wrote:
> "Patrick Turner" = congenital, autistic LIAR !!!
>
> > But its easy to measure the total winding resistance
> > with the amp functioning with a sine wave.
>
> ** It is much simpler to measure the secondary resistance values for
> various settings with a DC ohm meter.
Agreed, AC testing is overkill just to get the DC resistance, and it will
throw other losses into the measurement.
> They are, as previously stated, 1.6, 0.85 and 0.6 ohms for the 16, 8 and 4
> ohm settings.
The 1.6, and 0.85 Ohm measurements for the 16, and 8 Ohm settings were
previously stated, if the 0.6 Ohm measurement for the 4 Ohm setting was
previously reported, I missed it.
At any rate, thanks for the complete set of measurements, it seems to
suggest that the P-Q or "4 Ohm" winding section is wound with heavier wire
than the other two sections.
Regards,
John Byrns
shiva
"John Byrns" <jby...@rcn.com> wrote in message
news:jbyrns-1602...@216-80-74-216.d.enteract.com...
Hi John - looked at your 25L6 amp was this another case of "jeesh, I've got
these 25L6's, and I'l be darned if I can't mske somethin' reasonable out of
them (usually how my projects start out).
Here's a question: did you try playing with the values of the 47K
resistors across the input transformer's secondaries & do tone sweeps? Or
Nobel (sp?) networks? I've had weird experiences with small signal
transformers (I've been pulling them from old mixing consoles& recording
amps forever), where they would resonate at weird freqs, depending on the
output load. Perhaps that's where your "one note bass" was coming from?
-dim
shiva
Or wound inside the other windings, making the wire length per turn shorter,
making total length shorter. I'm sure people's gear is better than mine,
but when it gets to sub - Ohm readings, one almost has to solder the test
leads on to the probes, - the margin of error jumps right up. Perhaps
that's why Patrick suggested AC testing? (didn't read the rest of the
thread).
-dim
John Byrns
> "John Byrns" <jby...@rcn.com> wrote in message
> news:jbyrns-1602...@216-80-74-216.d.enteract.com...
> >
> > The 1.6, and 0.85 Ohm measurements for the 16, and 8 Ohm settings were
> > previously stated, if the 0.6 Ohm measurement for the 4 Ohm setting was
> > previously reported, I missed it.
> >
> > At any rate, thanks for the complete set of measurements, it seems to
> > suggest that the P-Q or "4 Ohm" winding section is wound with heavier wire
> > than the other two sections.
>
> Or wound inside the other windings, making the wire length per turn shorter,
> making total length shorter.
Good point, but I suspect there is more than that going on, especially as
Phil's measurements suggest that the other two secondary sections have
similar DC resistance at 0.5 Ohms each.
> I'm sure people's gear is better than mine,
> but when it gets to sub - Ohm readings, one almost has to solder the test
> leads on to the probes, - the margin of error jumps right up. Perhaps
> that's why Patrick suggested AC testing? (didn't read the rest of the
> thread).
I don't see how AC measurements would help with the problem you bring up?
AC measurements would also include other losses, which would add even more
confusion factors to divining if one wire gauge was used for all secondary
sections. I guess I need to find someone who has cut a dead QUAD II
Output transformer apart to see how it is wound.
Regards,
John Byrns
John Byrns
> Hi John - looked at your 25L6 amp was this another case of "jeesh, I've got
> these 25L6's, and I'l be darned if I can't mske somethin' reasonable out of
> them (usually how my projects start out).
Actually I don't think I even had the 25L6's when I set out to build the
amp. I had all the other parts except the tubes and the main filter
capacitor. The chassis, which turned out to be a perfect fit for the
project, I had bought at a ham fest still wrapped in its original brown
paper, but the bottom of the paper had been punched in and someone had
used it as an ashtray. For reasons I now forget, I had been theoretically
considering a similar design for several years, but didn't have any reason
to build it. Then back in 1998 someone in this newsgroup said it was
impossible to build an amplifier with direct rectification of the power
line, without ending up with tremendous hum. I took exception to that
idea, and ended up building the amplifier to prove that it could be done.
I ended up with hum and noise 80 dB below 1 Watt, without any negative
feedback, which I consider a pretty darn good result. The biggest hum
problem was the layout of the heater wiring near the driver tubes, which I
had to rearrange to achieve the desired results. Even better results
could be achieved if the heaters of the driver tubes were powered with DC
from the cathodes of the output tubes, although that would require using
different driver tubes with 150 mA heaters.
> Here's a question: did you try playing with the values of the 47K
> resistors across the input transformer's secondaries & do tone sweeps? Or
> Nobel (sp?) networks? I've had weird experiences with small signal
> transformers (I've been pulling them from old mixing consoles& recording
> amps forever), where they would resonate at weird freqs, depending on the
> output load. Perhaps that's where your "one note bass" was coming from?
I used the 47k resistors because the transformers were speced for a 100k
secondary load, and I figured the response would be flattest with that
load. The high frequency response extended to something like 80 kHz. I
would expect that improper loading would cause weird response mainly at
the high frequencies, the most I would expect at low frequencies would be
a possible early roll off. I believe microphone input transformers are
often run unloaded, for reasons I have forgotten.
I triple diddled, that is I changed three things at once, and all I know
is that the one note bass and the grainy sound both disappeared after I
made the changes. Two of the changes were in the amplifier circuitry, but
I now suspect that neither of them had any effect on solving the
problems. I suspect that the problems were due to the cheap Radio Shack
bridging transformer I first used to match the 600 Ohm input of the
amplifier to my preamp. I never bothered to go back and try the Radio
Shack transformer again in place of the RCA Broadcast Quality bridging
transformers I have been using the past 6+ years. It would be easy enough
to try the Radio Shack transformers again to see if they were the cause of
the problems. They simply plug into the line between the preamp and power
amp, but I haven't felt inspired to try it, the amplifier has been working
so well all these years, with no failures, and still operating on the
original tubes.
Phil Allison
"shiva"
"Phil Allison"
>> Agreed, AC testing is overkill just to get the DC resistance, and it will
>> throw other losses into the measurement.
>>
>> They are, as previously stated, 1.6, 0.85 and 0.6 ohms for the 16, 8
>> and 4 ohm settings.
>
> Or wound inside the other windings, making the wire length per turn
> shorter,
> making total length shorter.
** But not 40% shorter !!!
> I'm sure people's gear is better than mine,
> but when it gets to sub - Ohm readings, one almost has to solder the test
> leads on to the probes, - the margin of error jumps right up.
** FYI - I used a bench PSU plus a known 1% accurate, 2 ohm, 25 watt
resistor to measure the resistance figures.
The bench PSU voltage was adjusted to give exactly 2.00 volts DC across the
2 ohm resistor when wired in series with the Quad amp's output terminals.
The same (4.5 digit ) DC meter was then used to measure the voltage drop
across the output probing the OP transformer terminals inside the amp
itself. Given that the current is 1.00 amps, any voltage drop corresponds
with resistance value in ohms.
Inductance effects preclude the use of AC for measuring transformer winding
resistance.
Also - be aware of the tempco of copper is + 0.004 per degree C.
.............. Phil
Mike Coatham
or 60 cartons of w/shop stuff to unpack but I think I can lay my hands on it
in the next day or so. From memory, the secondary sections all have the same
number of turns and same guage of wire but I'll confirm one way or the other
very soon. Cheers
Mike
Patrick Turner
Phil Allison wrote:
> "Patrick Turner" = congenital, autistic LIAR !!!
I delete the rest of the repeated BS which fails to address Phil's
misunderstandings of what I have said.
For some silly reason, Phil cannot accept that somebody could
easily make an amp with better engineering involved than Mr Walker did all those
years ago.
I refuse to debate the subject with you unless you refrain from insults
and refrain entirely from snipping my posts.
So please fuck off. You ain't welcome here, and you are acting like a shithead,
but all the ppl here who have killfilled you
already know that.
Please notice that I am the only one to have the guts to stand up to an
arsole like you, and just remember that every time you start a new thread
which insults me, you insult yourself, and you make your credibility a POS,
and you make it very unlikely anyone else will ever reply to you.
Patrick Turner.
Phil Allison
>
> "Patrick Turner" = despicable, congenital LIAR !!!
>
> For some silly reason, Phil cannot accept that somebody could
> easily make an amp with better engineering involved than Mr Walker did all
> those
> years ago.
** The Turneroid sub human uses yet another desperate debating cheat by
changing the subject.
Finding it impossible to explain away or face his FUCKWIT errors he now
tries to divert the topic to a new and irrelevant one.
" The real problem * IS * the Turneroid's insane dogma and irrational
prejudices. The result of an inflexible mind besotted with self inflicted
rote learning and possessing no technical insights. "
> I refuse to debate the subject with you unless you refrain from insults
> and refrain entirely from snipping my posts.
** LOL - the Turneroid sub human now goes into pompous asshole mode.
He complains that I will not allow him go down sidetracks or reply to his
copious verbal diarrhoea.
He can go rot in hell before I do that.
> So please fuck off.
** Fuck off yourself - you evil, posturing bag of excreta.
> You ain't welcome here, and you are acting like a shithead,
** The Turneroid does not hold any ownership over RAT.
There are plenty of RAT posters both now and formerly that despise the one
eyed evil pig more than I.
............. Phil
Patrick Turner
John Stewart wrote:
> Patrick Turner wrote:
>
>> luc.de...@free.fr wrote:
>>
>> > Ok the conclusion is, Walker did an amazing good job.
>> > In fact i am in the "don't change nothing" posision for capacotirs
>> and
>> > resistors.
>> >
>> > I have now another problem :
>> > Woddside is not available to do the transformers at the moment
>> (selling
>> > business ?), and Sowter says that they are not doing perfect
>> replicas
>> > (instability ?).
>> >
>> > Who knows where to buy good quad transformers at reasonable price
>> ?
>> > Thank's
>> > Luc.
>>
>> I sure don't, but I would roll my own if I was suitably bribed
>> to perform the task.
>>
>> I'd try to find a way of improving on the horrors of the Quad OPT
>> with its high winding losses when connected for 8 ohms.
>
> And what would they be?
I measured a total of about 17%
> What about the core losses?
Core losses in most OPTs, even in Quad II, are negligible, because the
inductance of the primary is a high impedance at most signal F.
But Quad II also saturate at least twice the frequency I would ever
consider acceptable.
> It is presumptuous of you that you would criticize Quad while freely
> using their cathode FB output circuit. Got a better idea?
You bet I have. I like the Quad principle, and i apply it in my best PP
amps, but I use
OPTs with far lower losses and leakage inductances.
I also prefer my drive amp ideas. EF86 belong in preamps IMHO.
I refuse to produce amplifiers which comply with 1955 lowest common
denominator
standards. Sorry. I don't give a rat's arse what other folks are doing
with their amps,
I have my standards, and that's that.
> Well Patrick, we all have some 20-20 hind sight. It's cheap &
> certainly easy for us at this late date to criticize what has been
> done in the past by others. Then throw money & time at the problem &
> we are apt to make an improvement (or not). If Quad & others like
> Hammond & Sowter had found an economical way to combine all of 4, 8 &
> 16 ohm taps while using all the copper in the secondary winding all of
> the time they would have shown us how by now.
They consistently have refused to honour the best tube craft methods as
outlined in RDH4.
If they can't afford the extra 15 minutes of labour to make their
transformers comply with
the humble expectations of ppl like myself, then so be it, I will wind
my own trannies.
I shouldn't have to.
> Sowter manages to do that by using a 4-winding secondary, but their
> cost is more than double that of a similar Hammond.
My trannies are alsp more expensive than Hammond who wind trannies by
the thousand.
Its entirely unfair and unreasonable to expect a one off order for 2
OPTs by
anyone not doing volume to produce better quality at the same price.
I recently wound a single 400 watt OPT for a guy in NZ.
He couldn't find a better deal anywhere.
Couldn't find better performance either.
> But even while using a 4-winding secondary the outcome is not perfect.
> What should be the 8 ohm connexion actually results in 10 ohms.
How do you figure that?
>
>
> And Hammond managed to do theirs & deliver a product to the market at
> a reasonable price point, something overlooked by many. Something
> called 'Business Acumen', I guess.
They could be 25% more expensive, and could have the right way of doing
the secondaries.
They'd have the quality I'd expect, that that others would expect, and
get more business.
To continue to make sub standard gear at a budget price is plain
stubborn.
> The secondary winding as used by Quad appears to be the same as found
> in the Hammond 1600 Series OPT's, so probably manages 4, 8 & 16 ohms
> without difficulty.
I think you will find the Quad method differes from the Hammond.
Both are sub-standard, IMHO.
>
>
> In some of your posts you mention the possible difficulty of getting
> stability while using NFB with the OPT's of others. A quick look at
> your circuits shows them to be festooned with many phase correction
> networks. It seems your OPT's may be in that same category as those
> you have criticized.
Nearly every amplifier I ever examine has the same HF phase correction
circuits.
Those that don't usually always oscillate at HF with a 0.22 uf on the
output.
All have a badly rising response with a peak around 30 kHz.
Few have the same LF networks, and thus are only marginally stable at
LF,
often with poor recovery after over load.
The use of the phase correction networks improves a tube amp and the
sound.
For eg, my 300 W OPTs have bandwidth from 18 Hz, at saturation at full
power
to 270 kHz.
I can't use all that bandwidth, so I have a step network to reduce open
loop
gain, and so at higher than 30 kHz, the amount of applied FB is reduced.
You simply don't need a lot of applied FB at 50 kHz and above!
And nor do you want lots of applied FB at below 16 Hz.
Tube amps are not direct coupled, and there should be limits on the
applied NFB
at extreme ends of the bansdwidth.
Correctly done, such networks improve the sound.
> Now, here are the numbers for the secondary resistance for a Hammond
> 60 watt OPT.
> While connected for 4 or 16 ohms, 86% of the secondary copper is used.
> The 8 ohm connexion uses all of the secondary copper.
> For the 16 ohm connexion the resistance in the used windings in series
> with the load is 388 milliohms.
>
> For the 4 ohm connexion the resistance in the used windings in series
> with the load is 96.6 milliohms.
> For the 8 ohm connexion the resistance in the used windings in series
> with the load is 178 milliohms.
The S losses would be 2.36%, 2.34%, and 2.17% respectively, and this
looks better than Quad II.
I still won't buy a Hammond.
When they start using all the windings, I might, and only if they have
at least
4 separate sections in the secondary, with 5 primary sections, ie, like
I have done and as I have
described at my website.
WTF do you think I won't buy them? Its because I like my own better.
> All measured by applying a constant current & measuring the resulting
> voltage with a precision DMM.
I assume you used DC.
>
>
> Before you dismiss this as shoddy you had better to be ready to match
> it or do better for 88.64USD a copy & still make a profit sufficient
> to stay in business.
I didn't say its shoddy, you said that. I just say below the high
standards I think are reasonable.
I don't care if they sold at 8c, I still wouldn't buy them. Not ONE.
People attribute what they hear from my amplifiers to what I make, not
what Hammond make.
I have my standards, and its unreasonable for you to expect that I make
something alternative and cheaper
as a small volume concern.
I have had enquiries from chinese ppl spamming me with offers of their
services to wind transformers, so I point them to my website where the
information is sufficient
to allow anyone with knowledge and who can read english to be able
to make superlative OPTs for tube amps.
I have told the two ppl concerned I'd like to see a sample of what they
can do soon,
but not a single person has offered to wind something like I wind for
the amps I build,
and thus save me from the donkey work of winding bloddy transformers.
They see the detail I want which was routinely available by makers in Oz
and the US and the UK
in 1955, and they vomit and they go away knowing its all too hard,
and they are peeved that I don't want 10,000 of their samples of
absolute muck,
if they did things their way.
I will do what Hammond and the chinese won't, and what Quad didn't.
No other Oz makers I know can wind anything I want, including mains
trannies.
Don't get me started about the local mains tranny winders;
They are mainly a bunch of ignorant do-little-for-a-lot goons who know
SFA
about transformers. They all spew up when I tell them how to wind
transformers.
So I wind my own.
Mains transformers for tube amps shall be inaudible, and run nice and
cool,
despite handling hundreds of watts into a rectifier with large C and SS
diodes.
I get fed up having to repair the Luxmans, the Phase linears, the crap
with noisy transformers.....
Since a bowl of rice in Oz costs a shirtload more than in china,
its going to cost a lot more if I wind anything for you.
I have months of work ahead of me, and I don't want to specialise in
making OPTs for people with no idea about quality, performance,
and who can't pay the right price.
Nobody else seems able to wind quality at a reasonable price.
BTW, an OPT worth USD $88 in the US will be aud $122 here plus
the freight, and so the importer here charges about aud $200+.
I couldn't get any sense out of this fellow when i tried to pin him down
about
stocks he actually had, and a price.
So the customer of mine who had asked me to get a price from the Oz
distributor went elsewhere, but the prices were not cheap.
S I guess the Oz hammond distributor sells very few OPTs, and stocks
almost
nothing. He mainly sells tubes.
But folks say they love tubes until they have to pay for the
transformers that go with them.
I don't see you offering your services below a figure you could live on.
>
> The Sowter U066 is especially interesting since cathode FB similar to
> that used in the Quad is possible. To me it looks like it could be
> substituted into a Quad if there was no other alternative available. I
> almost bought one a few years ago for experimental use since it has
> many windings & possibilities on the primary.
Yes but the other complaint I could make about all the transfromer
makers and amp builders
is that almost none tell us exactly what is in their tranny.
Ian Iverson is quite happy with the CFB trannies he ordered from Sowter.
But I don't eaxctly know WTF is actually inside what they do.
What have these guys got to hide?
Are they frighened some bastard will steal their intellectual property?
They must be crazy, because whoever winds anything has to sell what they
wind....
Hardly anyone winds their own after making a lathe and practising...
You must have seen the Williamson OPT specification in RDH4.
That spec has been there for 50 years, and the ideas behind it dismissed
by
most mass makers.....
From the spec, we can work out all the parameters of the OPT, load
matches,
winding losses, leakage inductances, shunt C, primary L, saturation F,
primary wire dia, secondary wire dia,
exact winding geometry, insulation material and thickness...etc,
but we don't get any info from these other makers.
I have included such info at my website, and the OPT No1 is a typical
sample of anything
found in any amp I may build; but often I do better.
>> Damn it, there is no 4 ohm connection.
>> And the method of changing Z involves different current densities
>> in windings, not good practice....
>
> Please tell us why in 50 words or less.
If you cannot see the point of what I am saying, you need to read at
least more words
than 50 of mine to understand.
RDH4 has a bit on it, but try read well beyond that book as well.
The spec of the U066 OPT at sowter's site falls well shot
of the info I would expect.
Where have they placed the CFB winding with respect to the anode and S
windings?
They appear to have used 4 sections of secondary, but does this mean
there are two sections of two layers of S windings, or 4 sections of one
layer each?
What are the turns in each winding shown?
How are they combined to get different load matches?
Sorry Mr Sowter, but you will have to do better than you have when
providing information!
So I won't buy any of their transformers either. Its no use asking these
ppl.
They just say they have a reputation for quality. Not good enough for
me.
They fail to convince me they know what they are doing.
I make no apologies for being hard to please.
Besides, Sowter prices are higher than Hammond, and freight is a killer,
and it makes it worth to wind my own, and to hell with the lot of them.
> And how much different are the current densities in your opinion?
Lots different.
I don't believe in asymetrically arranged OPT secondaries.
There are no unused portions in any OPTs I wind.
There are no paralleled sections of windings where the current
density varies between any wire used in the secondary.
Do it right, or not at all, imho.
To those not worried by current density variations and wasted portions
of windings
and highish winding losses, enjoy your music and don't start worrying.
But as soon as you do start worrying about the details, and the methods
used in OPTs,
then pretty soon you start wondering why they do it the way they do.
Patrick Turner.
Patrick Turner
John Byrns wrote:
The easiest way to measure the winding resistances other than
as I have suggested, which tells you what the actual working resistance is
would be is to apply 10v dc via 100 ohm resistance to the sec, and measure the
sec coil voltage. This is the CCS method.
Primary R can be measured easily with a DVM, although probably
with the secondary shorted, lest the inductance of the primary prevent
the resistance measurement.
My Fluke refuses to measure my own OPTs due to so much inductance
lest I shunt it by shorting the sec.
But Phil Allison in between/during his latest tantrums has provided us with
winding
resistances for Quad II secs.
BTW, to measure the actual Ro of any amp, set it up
with a 400 Hz signal at say 3 volts of output, no load, and then
apply a load, and measure the new output signal, and find the V change.
Ro = Vchange / IRL.
Ro becomes greater when the amp works in the B part of AB, and will be different
at 10 Hz, 400 Hz, and maybe 20 kHz..
shiva
> In article <cqKQd.59403$g16.44283@trndny08>, "shiva" <help...@666.com>
wrote:
>
> > Hi John - looked at your 25L6 amp was this another case of "jeesh, I've
got
out of
> > them (usually how my projects start out).
>
> Actually I don't think I even had the 25L6's when I set out to build the
> amp. I had all the other parts except the tubes and the main filter
> capacitor. The chassis, which turned out to be a perfect fit for the
> project, I had bought at a ham fest still wrapped in its original brown
> paper, but the bottom of the paper had been punched in and someone had
> used it as an ashtray. For reasons I now forget, I had been theoretically
> considering a similar design for several years, but didn't have any reason
> to build it. Then back in 1998 someone in this newsgroup said it was
> impossible to build an amplifier with direct rectification of the power
> line, without ending up with tremendous hum. I took exception to that
> idea, and ended up building the amplifier to prove that it could be done.
> I ended up with hum and noise 80 dB below 1 Watt, without any negative
> feedback, which I consider a pretty darn good result.
Hi John -
Not sure why someone would say that 'bout he hum for series fils on AC
Mains - perhaps 'cos everyone's so DC-crazy (Good DC fils are a rarity,
unless good SS regulation is used (not for regulation - for *filtering*, and
then - it stops being a toob amp. I'd rather have a pile of differential
noise of a set freq. with no harmonics than a mixture of common &
differential mode noise with *tons* of harmonics.
I'm sure the fil the wiring layout is quite a bit more critical (I usually
use 2 dif. color wires for fils - I've seen many PP sections where fils were
wired out of phase - depending on the heater construction, this interferes
with hum cancellation in a PP circuit), though it certainly could be done.
You might have to "wiggle" a few heater wires after changing a toob... I
usually stay away from series fils (even though they make plenty if sense)
simply 'cos I 'd have to think more about tube selection, no last-minute
change - of mind about the tubes used, etc., etc.
> The biggest hum
> problem was the layout of the heater wiring near the driver tubes, which I
> had to rearrange to achieve the desired results. Even better results
> could be achieved if the heaters of the driver tubes were powered with DC
> from the cathodes of the output tubes, although that would require using
> different driver tubes with 150 mA heaters.
>
You should check out some heater DC schemes, like (If I remember right)
Bogen's DB 30 (the one with cathode feedback OT's), where he gets DC for a
preamp tube in a kind'o weird way by shunting a HT choke, through a pair of
low-voltage / high Mf caps with the tube's fils as the resistance. I could
be mis-remembering, though...
> > Here's a question: did you try playing with the values of the 47K
> > resistors across the input transformer's secondaries & do tone sweeps?
Or
> > Nobel (sp?) networks? I've had weird experiences with small signal
> > transformers (I've been pulling them from old mixing consoles& recording
> > amps forever), where they would resonate at weird freqs, depending on
the
> > output load. Perhaps that's where your "one note bass" was coming from?
>
> I used the 47k resistors because the transformers were speced for a 100k
> secondary load, and I figured the response would be flattest with that
> load. The high frequency response extended to something like 80 kHz. I
> would expect that improper loading would cause weird response mainly at
> the high frequencies, the most I would expect at low frequencies would be
> a possible early roll off. I believe microphone input transformers are
> often run unloaded, for reasons I have forgotten.
>
In my head, it goes something like this (this is the opposite of a lecture -
just thinking this through in my barely - awake head... please correct me
where my reasoning goes astray): The transformer's winding could be seen
(to simplify from an infinite ladder) as RL shunted by C (dc res,
inductance of the windings, and inter-winding capacitance for the shunt C).
The L & C parts of the thing make it a basic resonant circuit above which
you should have a pure RC roloff, but it's a ladder, of rl shunted by c's,
so it's not quite that...
The circuit could be made to oscillate (welll, resonate) at any freq.
below resonance by changing the C of the load (bigger C = lower res. f. &
earlier roloff, higher reflected R into the primaries at high freq's.), and
by changing the load R (dropping reflected resistance to the primary, thus
cutting the high freq's on the input side (imperfect - presumably RC/LRC,
whatever), dropping the coupling efficiency (thus droppinghe couplin'
efficiency & causing low -end distortion & roloff )...
Oh, got to go and make $$$ - I'll send this unfinished, please point out
the mistakes ( my mind is in a fog now, and I'm not leafin' through ref.
books to prevent blunders).
-dim (off to do some blue-collar work...)
shiva
"Phil Allison" <phila...@tpg.com.au> wrote in message
news:37ibr1F...@individual.net...
>
> "shiva"
> "Phil Allison"
>
> >> Agreed, AC testing is overkill just to get the DC resistance, and it
will
> >> throw other losses into the measurement.
> >>
> >> They are, as previously stated, 1.6, 0.85 and 0.6 ohms for the 16, 8
> >> and 4 ohm settings.
> >
> > Or wound inside the other windings, making the wire length per turn
> > shorter,
> > making total length shorter.
>
>
> ** But not 40% shorter !!!
>
Hi Phil -
Look, I'm not even following this thread, as I've mentioned. I don't even
know what the debate's about...
I also don't know how thick the secondary wire is, how the windings are
interleaved, how badly they're wound, how small the radius is (circumference
bein' 2Pi r, if r1's approachin' 0, than you can have as many turns as you
wish without any appreciable length of wire, but r2's gonna be 0+(thickness
of secondary wire), unless you're stacking right away & without
tape)+(thickness of any tape), so the next layer's gonna be ... well, a
whole *bunch* longer... Just an extreme case, but you see my point, right?
So, off the top of my head, 40% for secondary winding res. diff doesn't seem
out of bounds. It's obvious that the 4ohm winding would be the closest to
the primaries - has to be to improve the couplin', with the fewest turns
being the worst offender, so it's got to be shorter. 40% is fine by me....
>
> > I'm sure people's gear is better than mine,
> > but when it gets to sub - Ohm readings, one almost has to solder the
test
> > leads on to the probes, - the margin of error jumps right up.
>
>
> ** FYI - I used a bench PSU plus a known 1% accurate, 2 ohm, 25 watt
> resistor to measure the resistance figures.
>
> The bench PSU voltage was adjusted to give exactly 2.00 volts DC across
the
> 2 ohm resistor when wired in series with the Quad amp's output terminals.
How good are the terminals?
> The same (4.5 digit ) DC meter was then used to measure the voltage drop
> across the output probing the OP transformer terminals inside the amp
> itself. Given that the current is 1.00 amps, any voltage drop corresponds
> with resistance value in ohms.
>
Hey! I knew that ohm's law stuff would come in handy someday!
> Inductance effects preclude the use of AC for measuring transformer
winding
> resistance.
>
Inductance could be easily measured, especially given such low DC R values.
Also, the inductance should be rather small, if all the other taps are open.
At 50 /60Hz, which is what I assume Pat suggested (did he? Using the
mains?), you'd get pretty close. I wouldn't think of using AC for *DC*
resistance measurements of trannys, but who knows - i didn't read the
thread...
> Also - be aware of the tempco of copper is + 0.004 per degree C.
That's nice. why, did you get a major readin' drift while the winding was
under test? Maybe youu were undone by the 25W resistor's temp. co.? (being
silly here)
Hey, maybe you have a shorted winding, and maybe the winding is not 4 ohms
at all, perhaps it's 3.2, and conveniently labeled as 4 'cos that's what the
consumer expects to see? Thhe power loss of a mismatch like that is
negligible, and the THD usually improves...
Anyhow, what *is* this thread about?
-dim (clueless)
>
>
>
>
> .............. Phil
>
>
Jon Yaeger
Phil Allison wrote:
"Patrick Turner" = congenital, autistic LIAR !!!
Je parle,
Phil Allison = killfilter's raison d'etre
Au revoir, Philippe!
- Jon the Zombie
Phil Allison
"shiva" = a drug addict shithead
"Phil Allison"
>>
>> >> Agreed, AC testing is overkill just to get the DC resistance, and it
> will throw other losses into the measurement.
>> >>
>> >> They are, as previously stated, 1.6, 0.85 and 0.6 ohms for the 16, 8
>> >> and 4 ohm settings.
>> >
>> > Or wound inside the other windings, making the wire length per turn
>> > shorter, making total length shorter.
>>
>>
>> ** But not 40% shorter !!!
>>
>
> Hi Phil -
> Look, I'm not even following this thread, as I've mentioned. I don't even
> know what the debate's about...
** Then you ought to fuck off.
> I also don't know how thick the secondary wire is, how the windings are
> interleaved, how badly they're wound, how small the radius is
> (circumference
> bein' 2Pi r, if r1's approachin' 0, than you can have as many turns as you
> wish without any appreciable length of wire, but r2's gonna be
> 0+(thickness
> of secondary wire), unless you're stacking right away & without
> tape)+(thickness of any tape), so the next layer's gonna be ... well, a
> whole *bunch* longer... Just an extreme case, but you see my point,
> right?
** Go check the the shape of the core of a real transformer - shithead.
> So, off the top of my head, 40% for secondary winding res. diff doesn't
> seem
> out of bounds.
** But your head is pointed and nothing will sit there.
> It's obvious that the 4ohm winding would be the closest to
> the primaries - has to be to improve the couplin', with the fewest turns
> being the worst offender, so it's got to be shorter. 40% is fine by
> me....
** Who cares what is "fine" by an admitted fuckwit.
>> Inductance effects preclude the use of AC for measuring transformer
> winding resistance.
>>
>
> Inductance could be easily measured, especially given such low DC R
> values.
** How would a total fuckwit know ??????
> Also, the inductance should be rather small,
** It is huge even at 50 Hz - you complete ass.
> if all the other taps are open.
> At 50 /60Hz, which is what I assume Pat suggested (did he? Using the
> mains?), you'd get pretty close. I wouldn't think of using AC for *DC*
> resistance measurements of trannys, but who knows - i didn't read the
> thread...
** What a drug fucked ass.
>> Also - be aware of the tempco of copper is + 0.004 per degree C.
>
> That's nice. why, did you get a major readin' drift while the winding was
> under test? Maybe youu were undone by the 25W resistor's temp. co.?
> (being
> silly here)
** Maybe you were an abortion that crawled out of the bucket ??
> Hey, maybe you have a shorted winding, and maybe the winding is not 4 ohms
> at all, perhaps it's 3.2, and conveniently labeled as 4 'cos that's what
> the
> consumer expects to see? Thhe power loss of a mismatch like that is
> negligible, and the THD usually improves...
> Anyhow, what *is* this thread about?
> -dim (clueless)
** Piss off moron.
............. Phil
shiva
Dear Phil.
I do not use recreational drugs, haven't for years, unless you consider a
couple of beers on weekends. Do try to keep your bitch ass from making
slanderous statements.
Yes, i do make money in electronics, and a cunt like you is just beggin'
to be sued, but, somehow, I'd feel a little soiled...
While it is self-evident that my cats have finer debating techniques than
you, I thought I'd mention it nonetheless. Things which are obvious to
folks with IQ's in low double digits may not be as clear to you, who's few
remaining neurons are relyin' on syphilis spirochetes for neurotransmitters.
Don't loose heart, Phil - a human brain is an incredible machine, even the
one you've been 'blessed" with. What's so darned hot about upper brain
function, anyhow, right? Why should you care if your bile-filled verbiage
is birthed in grotesque neuropathology & has no reflection of, or pertinence
to, reality?
Just take pride in the fact that you're a *mammal*, one of the most
sophisticated things roamin' this planet. No matter how *lacking* a mammal
as you may be.
Life itself is a mmiracle, and *you*, (lucky you!), are *alive*.
If you *still* don't feel that mere *existence* and the exclusivity of
mammalhood is enough, modern medicine has come a long way, dear friend, and
you may find out that your so-called "deficiencies" are caused by something
as trivial as a football-sized brain tumor, or a neglected cranial fracture
(check now - a festering hole, oozing puss & spinal fluid, could explain so
much!). All this may be remedied by your friendly neighborhood MD, or even
a barber...
And Phil - you're not in my killfile (I ain't even got one). I *enjoy*
knowing that folks like you exist, and when I feel down, i remember you, who
(by contrast) make me feel truly blessed.
Now, take a deep breath, and think *happy thoughts*!
I'm afraid I'll have to go now, Phil. Chin up, Champ!
-dim