SE output transformers.
Jenya
I was wondering if there is a place where I could buy a good quality
transformers for my SE 300B. I checked triodeel. They have Magnequests for
$150 each. Are these good transformers? Is there a better way to shop for
them?
Jenya
mr_montreal
As far as I am concerned, the best SE transformers are made by Electra-
Print. Jack Eliano is a super craftsman.
Check them out at:
http://www.execpc.com/~n9zes/electra.html
Regards,
M
In article <8s7fgd$10og$1...@news.mainstreet.net>,
Sent via Deja.com http://www.deja.com/
Before you buy.
CERREM
>I was wondering if there is a place where I could buy a good quality
>transformers for my SE 300B.
Hi... Being an arogant EE and transformer designer I will of course say my
OPT's are the best....and have satisfied customers... I don't like or care for
what is on the market.
But I must confess and agree with the second post to this thread.... The
Electa-Print are VERY good OPT's and are the only OPT's other than mine that I
really like and respect....the designer has good design pholosophy.....
CHEERS
CM
Lou Baby
CERREM <cer...@aol.com> wrote in message
news:20001014002334...@ng-bj1.aol.com...
Mark Pearson
You already decided that the stock transformers that came with your kit
aren't good enough? I didn't know you had even finished building the
project yet. Wow, those old Zenith speakers must really be good if you have
already decided you need a higher quality output transformer than what came
with your kit!
Mark
Jenya <je...@securify.com> wrote in message
news:8s7fgd$10og$1...@news.mainstreet.net...
> Hi,
> I was wondering if there is a place where I could buy a good quality
Ole Christensen
Lou Baby <emai...@att.netREMOVE> skrev i en
nyhedsmeddelelse:EdZF5.28641$tl2.1...@bgtnsc07-news.ops.worldnet.att.net.
..
> I guess that's why Tango is going out of business?
>
> CERREM <cer...@aol.com> wrote in message
> news:20001014002334...@ng-bj1.aol.com...
bobcx
and I am looking forward to your explanation of the special design
philosophy you admire so much. And while you are at it can you please give
some details of YOUR special transformers.
CHEERS
BC
"CERREM" <cer...@aol.com> wrote in message
news:20001014002334...@ng-bj1.aol.com...
Ivan Johnston
"CERREM" <cer...@aol.com> wrote in message .... The
> Electa-Print are VERY good OPT's and are the only OPT's other than mine
that I
> really like and respect....the designer has good design pholosophy.....
I'm very surprised that you would say this. Do you believe that a primary
inductance of 20H is adequate for a 10k impedance OPT? Jack apparently
thinks so, witness the design of the KL10KB. How about 16H Lp for a 6.2k
OPT (E18B), or 16H for a 5.4k OPT (MT5KB)? These seem to be severely
lacking in inductance, unless they are intended only for driving a midrange
and/or tweeter.
Regards,
Ivan
CERREM
>inductance of 20H is adequate for a 10k impedance OPT?
Obviously I haven't looked at his OPT's in a while..... I have seen one a while
back that seemed pretty good... So next time I should keep my mouth shut until
I look over most of his specs again :(
BTW: With triodes...the inductance required is not really not a function of the
plate load...
Inductance required is = ( plate load in parallel with plate resistance) / (
6.28xfr)..
The fr = -3dB low frequency point...
So with triodes..the plate resistance will dominate when paralleled with a much
higher plate load...while with pentodes the plate loads are smaller than the
plate resistance, about 1/10 - 1/12......so the plate load dominates in the
equation...So when dealing with pentodes...and looking over transformer
specs...one can come to a close rough idea of the low end response by plate
load.....and with triodes...knowing the plate resistance will be usefull...
CHEERS
CM
Ivan Johnston
> BTW: With triodes...the inductance required is not really not a function
of the
> plate load...
> Inductance required is = ( plate load in parallel with plate resistance) /
(
> 6.28xfr)..
Yes, that goes without saying for most of us here. One would not typically
specify a 10k OPT for a low Rp tube, so using the load impedance as a
reference is good enough to illustrate how poorly these transformers would
perform, for a typical application. The 10k transformer in question is
specified for use with a 211, according to Electra-Print's data sheet. No
matter how much you polish this turd, the inductance is woefully inadequate
for any real-world application that would require a 10k load, unless you dig
elliptical load lines.
Regards,
Ivan
Alan Douglas
Ivan Johnston asked:
>I'm very surprised that you would say this. Do you believe that a primary
>inductance of 20H is adequate for a 10k impedance OPT? Jack apparently
>thinks so, witness the design of the KL10KB.
Primary inductance is one of those numbers, like Gm of a tube, that
appears easy to specify, but actually depends strongly on the
measurement conditions. You may find that the inductances are not so
different, when competing transformers are measured with identical
methods.
Cheers, Alan
Ivan Johnston
"Alan Douglas" <adou...@gis.net> wrote in message >
> Primary inductance is one of those numbers, like Gm of a tube, that
> appears easy to specify, but actually depends strongly on the
> measurement conditions. You may find that the inductances are not so
> different, when competing transformers are measured with identical
> methods.
That goes without saying, the measurements must be performed under the
conditions that the transformer is designed to operate. I've specified and
measured enough transformers to understand what the numbers mean. I also
own some EP transformers, and the measurements indicate performance that is
consistent with the specifications on the manufacturers data sheet. I have
also found that the inductance of the EP transformers does not vary a whole
lot as a function of signal level (probably due to a large air-gap and
conservative design), which is good. The inductance is just simply too low.
Regards,
Ivan
dave slagle
> Primary inductance is one of those numbers, like Gm of a tube, that
> appears easy to specify, but actually depends strongly on the
> measurement conditions. You may find that the inductances are not so
> different, when competing transformers are measured with identical
> methods.
hey,
i'm with alan on this one... for me the inductance number is worth
little more than the paper it is printed on... the true test of an
output is the in circuit performance, and in defense of jack... he lists
22hz as the minus 1 DB point of his 10K trannie... and i know of several
people who have happily built with them...
dave
Acrosound
>Primary inductance is one of those numbers, like Gm of a tube, that
>appears easy to specify, but actually depends strongly on the
>measurement conditions. You may find that the inductances are not so
>different, when competing transformers are measured with identical
>methods.
Your right here Alan. And if you don't specify the conditions under which an
inductance were obtained it is very much like a statement that an amplifier can
"deliver 50 watts of power"
in this case, we would want to know over what frequency range does this hold
and at what level of distortion does this hold. In years past, funny numbers
games regarding power output levels prompted the FCC to require that all power
claims be stated in a manner which provided both frequency and distortion as a
part of the claim.
Inductance numbers by themselves are not as useful as when a context is
provided for them. But this is not that dificult either.
An inductance claim or measurement should state what AC voltage level was
employed, the AC frequency and how much DC current (in the case of a
conventional airgapped SE transformer) was across the primary for the test.
Ideally the AC voltage level will be both at an incremental point (small
volts... say four or so) as well as at the maximum power output level specified
for the transformer.
The DC unbal current level should also be specified. There is no standard
shared methodology for the generation of these inductance specifications...
meaning many different manufacturers will use differing procedures to generate
their numbers.
So you would want to know and evaluate the inductances in relation to how the
AC volts across the coil, what frequency was used (a test done at 50 hertz for
instance as opposed to 1 khz imposes a flux level fifty times that of the 1 khz
test), and how much DC current was put through the coil.
but given this context it should be possible to compare inductance measurements
btwn different samples.
MSL
CM
Acrosound wrote:
>
> Alan Douglas wrote:
>
> >Primary inductance is one of those numbers, like Gm of a tube, that
> >appears easy to specify, but actually depends strongly on the
> >measurement conditions. You may find that the inductances are not so
> >different, when competing transformers are measured with identical
> >methods.
Agreed...
When I design OPT's I always supply the customer with all the specs they
want with the test conditions.... BEFOREHAND...
My standard is to give the L at FULL ratted DC current and at FULL power
output and low power output..
My standard is to rate the OPT at FULL power output @ 20Hz @ a specified
flux density....
Depending on the core material...for example..M6 Push-Pull design I will
be at about 10,000 to 12,000 flux density range at 20Hz @ full power.
I have plotted core distortion vs. Flux density at 20Hz and feel these
number to be good...keeping the core distortion below .5% at full
power..
There are lots of tranny companies that play games with the specs to
make thier power outputs look better....such as ratting the power
output with a flux density at 18,000 gauss for M6...this is Bull-S#$%
!!!
Since this is right at below the saturation curve for M6...and the core
distortion is rediculous.... But they will not tell you this...then tell
you everything is proprietiery.... which is another crock of sh#%$ !!!!
Or they get really cute and rate the full power at 400Hz or which does
the same thing as previously stated...
If do-it-yourselfers took it upon themselves to read and learn more
about the workings of transformers...they would start demanding more
stringent specs and data sheets from the OPT makers... then the good
ones would stay and the bad ones would get better or go out of
business...
I have some problems with some of the newer publications on amp building
and transformers...some of these writters are mis-informed and are not
mathamatically inclined and do not have the proper insight to inform the
public on such matters...what may happen is we get a new generation of
tube newbies who don't have a clue....
The good news is there a many new to the hobby that are digging up the
old classic engineering textbooks on the subject and learning it
properly from ground up while also experimenting....
CHEERS
CM
Acrosound
Here is an example of the testing points which I feel are useful in
relation to documenting an inductance claim in a conventional airgapped
single ended output transformer.
Subject transformer has a nominal primary impedance of 2500 ohms.
Subject transformer has a published DC current rating of 60 madc.
Subject transformer was designed principally for use with 2A3 tubes or
equivalents.
Subject transformer has a published primary inductance number of 18 henries
IIRC.
All inductance measurements were done at a line frequency of 50 hertz.
All are open circuit inductances.
here are the findings when you plot AC volts against DC current levels.
voltage level power level 60madc 70madc 80madc
(in watts)
1 .0004 15.7 14.8 14.8
6 .0144 18.2 17.8 17
10 .04 18.9 18.4 17.5
20 .16 19.8 19.4 18.4
40 .64 20.6 20.3 19.5
50 1 watt 20.8 20.5 19.8
100 4 watts 21.5 21.2 20.3
Interesting tidbits:
The 1 volt drive level represents a power level of four ten-thousandths of a
watt. Once you get to fourteen and a half one-thousandths of one watt output
the inductance level varies roughly only about 10 percent from this level all
the way up to 1 watt and only about 15 percent up to four watts in the 60 madc
unbal plate current category.
These tests were all performed by an outside agency which has mil-spec T27 lab
approvals.
Actually, Mil Spec T-27 is a very good independent source for constructing a
wide range of tests and their conditions. Information on these specifications
and procedures should be available somewher on the web or through a library.
MSL
Acrosound
you make. Some of your other contentions/allegations I won't respond to as
they seemed to be "baiting" and "provacative" in intent.
>My standard is to give the L at FULL ratted DC current and at FULL power
>output and low power output..
This is a good practice. See my post in response to Alan Douglas for an
example I think much like what you recommend.
>My standard is to rate the OPT at FULL power output @ 20Hz @ a specified
>flux density....
This too is a good idea and practice. Assuming that you "subtract" the
magnitude of DC flux from the maximum available flux if the unit is designed to
carry unbalanced plate current.
>There are lots of tranny companies that play games with the specs to
>make thier power outputs look better....such as ratting the power
>output with a flux density at 18,000 gauss for M6...this is Bull-S#$%
>!!!
whether bullshit or not... it would, in and of itself, be an improvement in the
sense of disclosing what the flux level was at the rated power level.
My sense would be just give the facts and let the interpretation of the facts
not be the subject (at a first level) of debate, criticism, or speculation.
There are many, many world class transformers that do operate above the 12
kilogauss level and there may be reasons designers do choose to operate at a
higher flux density level. And these reasons may be good reasons to boot.
But this takes us into "interpretation" and "evaluation" of data points when it
might be argued that getting the data points out is itself a first good step in
making a fuller disclosure of what a transformer's behaviour might be.
>Or they get really cute and rate the full power at 400Hz or which does
>the same thing as previously stated...
I agree that a power rating at say 400 hertz is not very useful. It represents
only one twentieth of the flux density that would be present in the transformer
than if you did the measurement or calculation at say 20 hertz. 400 hertz, is,
in short, IMO, not very useful.
But again, if someone would choose to state it this way... that is their
option. Meaning there is no law or requirement that they state the data in a
specific format or within specific parameters.
>If do-it-yourselfers took it upon themselves to read and learn more
>about the workings of transformers...they would start demanding more
>stringent specs and data sheets from the OPT makers... then the good
>ones would stay and the bad ones would get better or go out of
>business...
With the cavaet that a few (however many parameters we all agreed on)
specifications may or may not.... this being the subject of great debates and
controversy itself, be wholly determinative of a transformers use and quality
in real world applications.
Recall that many of these tests are done with a steady (non-dynamic) singular
frequency and into (in many cases) purely resistive loads.
We do not capture the impulse response behaviour of the device or the behaviour
of the device while playing complex (many notes at the same time) music.
It's the same cavaets, for better or worse, that not all the time does the
cartridge with the best frequency response always sound the best when it is
actually playing records. Or, who would purchase a pair of speakers just based
on numerical data points without a listen to the subjects first?
None the less, I too, feel that more data as opposed to less data on tranneys
is a good thing.
>The good news is there a many new to the hobby that are digging up the
>old classic engineering textbooks on the subject and learning it
>properly from ground up while also experimenting....
The RDH is the "bible"... it has very good discussion of tranneys and their
behaviour. Another excellent reference which I study a lot and have found very
useful is Dr. Partridges landmark research on magnetic cores and distortion and
which was published in a four part series in 1939 in England (Wireless World
magazine).
MSL
Acrosound
data points in a comparative (unit A versus unit B) manner.
This is so true. Here are some of the differing "funny numbers" games I have
seen published in regards to transformers.
1) Company A publishes a DC current rating " at the knee of the curve". They
say it is the "...maximum amount of DC current before saturation..." In the
real world if you use the "big" number they give you... you have already used
up all of the core's flux capacity before ever putting a single AC signal (of
any strength or frequency) through the trans.
It gives them a big DC current rating to publish and gives you absolutely zero
AC magnetic headroom.
2) Another maker uses a big number in their published DC current rating but
does all of their own in-house testing of frequency response and inductance
measurements at one half (or close to one half depending on the particular
model) of the published DC current rating and then only at one AC voltage level
(in this particular case 4 volts RMS) and at one frequency (50 hertz).
3) Another company published their inductance figures with a voltage level of
250 volts RMS across the primary and NO dc unbal current through the unit.
This will generate a big inductance number but much of this inductance may well
disappear with the onslaught of the music in the real world.
These are just three examples that I thought of quickly. I am sure there are
many more out there.
It is a prime example of when comparing quantitative data make sure that you
are comparing "apples with apples and oranges with oranges".
If you don't have an equal basis for the generation of the data then you have
zero ability to really do any comparative data anaylsis btwn differing units.
I can recall so many times seeing people get excited about a SE output because
it had a huge *published* DC current rating and them thinking that this was a
"figure of merit" for a transformer so that the trans with the largest DC
current rating was ipso facto better than one with a smaller published DC
current rating.
More interesting would have been how do these two differing DC current numbers
compare when acheived using the same methodology. How much AC magnetic headroom
do they allow in their designs? What is the distortion and phase shift of the
subject transformers at some stated low frequency and power level.
Bear in mind... though... that even my current discussion presupposes (only for
the sake of the point being made) that
DC current capacity is even near the top of what makes a good
tranformer a good transformer.
MSL
CM
Acrosound wrote:
>
> Hello CM: enjoyed most of your post and wanted to respond to several points
> you make. Some of your other contentions/allegations I won't respond to as
> they seemed to be "baiting" and "provacative" in intent.
I have seen this NG get really boring lately ....so I thought I would
throw in the grenades :)
Although in person I am nicer and act professional and cordial...
> >My standard is to rate the OPT at FULL power output @ 20Hz @ a specified
> >flux density....
>
> This too is a good idea and practice. Assuming that you "subtract" the
> magnitude of DC flux from the maximum available flux if the unit is designed to
> carry unbalanced plate current.
My earlier posting mentioned Push-Pull for that example... But for SE
applications I define full power output as the sum of the AC flux
density and DC flux density to be limited to a specified MAX. total flux
density...This is where I establish full power output...
The DC flux density is at full DC current...and the AC flux density is
at full output wattage intended and at 20hz... Now the proportion or
ratio of AC/DC flux is also a matter of choice up to a point depending
on the optimization in mind.
The desicion on "WHERE" to put the maximum flux density is strictly a
matter of descretion on my part... my main concern is to maintain and
limit the core distortion to a predefined point... Now with P-P this is
straight forward since the gap is goverened by the space between the
interleaved E's and is fairly constant and consistant... With butt
stacked cores..the gapping will and increase the available flux headroom
by means of linearizing the BH loop... Now I am able to push the total
flux density higher BUT still maintain my crieria of max allowable core
distortion... I prefer to target flux density and not L in P-P OPT's
since I always wind up with a heck of a lot more L than required...
With SE OPT's I target the Max total flux and required L simutaneously.
>
> There are many, many world class transformers that do operate above the 12
> kilogauss level and there may be reasons designers do choose to operate at a
> higher flux density level. And these reasons may be good reasons to boot.
Agreed as well... as I stated above.. I assume we are leaving this
discussion open exclusively to OPT's.....
> Recall that many of these tests are done with a steady (non-dynamic) singular
> frequency and into (in many cases) purely resistive loads.
>
> We do not capture the impulse response behaviour of the device or the behaviour
> of the device while playing complex (many notes at the same time) music.
I have included elliptical loads in my analysis...but because every
speaker is different...I being the power amp designer have fullfilled my
requirements... it is the speaker designer that shares the blame... I
do compensate for this in some way provided I am allowed a amp with
feedback... Open loop amps are at the mercy of the LOAD...
The step(transient) response is totally predictable based on the
frequency response with use of Inverse Transforms.......
I feel the biggest problem with OPT's is the capacitance being dynamic..
The leakage is pretty much locked in place...the inductance is
incremental based on gap...But the capacitance is all over the place...
I have plotted the capacitance vs. power output for many classic
OPT's...as you may know the AC voltage gradient changes so does the
dielectric constant..in simple terms... The worse case offenders are the
ones with ORGANIC dielectrics....
> The RDH is the "bible"... it has very good discussion of tranneys and their
> behaviour.
Personally the RDH is written for techs and hobbiest and is not a
serious engeering book...even though I like the book and read it over
several times in my teens.... The section on transformers I feel is
introductory and good starting point. The book was intended for design
engineers with lower math skills.... Associate level..ect..
Most of the hard-core EE's were doing solid state research, RF and
vacuum science....
Another excellent reference which I study a lot and have found very
> useful is Dr. Partridges landmark research on magnetic cores and distortion and
> which was published in a four part series in 1939 in England (Wireless World
> magazine).
I read these articles years ago as well... as informative as they were
in their day...it's old hat....it's up to us to take it to another
level...the cores Dr. Partridge worked with were from another time...
I have done research with M6 and Supermendur and have analyzed and
graphed the core distortions produced...these have been converted into
polynomials and through linear systems derived the coefficients to make
a single equation for distortion... This is what I use to assist in OPT
design...
CHEERS
CM
Koenraad De Bie
Jenya wrote:
> Hi,
> I was wondering if there is a place where I could buy a good quality
James Bond
I hear that these transformers are great!
If you can't get a hold of one. Try the Hammond 300B. They go for $111 (I
think from www.angela.com) It is specifically designed for 300B tubes. The
model number speaks for itself!
That is what I am using right now with my new amp. I was using the 274BX
before and it got really hot after many hours of use. Now it is barely warm
to the touch. If terms of sound, I dunno. I didn't break-in and/or listen
to the 274BX and the whole amp before I replaced it with the 300B xfrmr.
Koenraad De Bie <koen....@pandora.be> wrote in message
news:3A036601...@pandora.be...