| Brian |
Can a power amp be built with two output transformers per channel? A big one optimized for accurate bass going to the woofer, and a small one optimized for clear and sweet mids and treble going to the wide range driver.
The output from each tranny would be mostly full range so a crossover circuit on the speaker would still be required. I am not seeing this as any kind of biamping solution, only an optimization. |
|
|
| Bas Horneman |
Actually Doede Douma has something just like that...
http://dddac.de
Well almost ..it is one transformer with 2 secondaries...1 for bass 1 for treble with silver wire.. |
|
|
| dhaen |
Hi,
This has been suggested before, either here or AA.
It's the primaries that have differing requirements.
I think it would be best to put the 2 transformers' primaries in series, with the HF transformer connected to the anode.
Cheers, |
|
|
| Bas Horneman |
Hi John,
Got the EY81's and topcaps today! Thank you very much!
Here is the link to the transformer page...on Wil's bifilar windings..one with silver wire and one with copper...so not really with 2 secondaries... but nevertheless an intersting concept..
http://www.dddac.de/tp05.htm
I don't understand why you would want the dual primaries? Could you elaborate...?
Cheers,
Bas |
|
|
| dhaen |
Hi Bas,
Glad they got there! International post can be a bit dodgy.:dead:
Doede's 2 Headed dragon:
Interesting concept. I'll have to have a think about what's happening. It's not clear from the pictures, what the winding detail is, but I have some suspicions.
Dual primaries (this principle is for SE, but could be adapted to PP):
I meant 2 transformers: One for HF with low inductance and low capacitance.
The other for LF with High inductance and high capacitance.
In series:
The HF primary is low impedance to LF,
The LF primary is low impedance to HF.
Both transformers get what they want out of life;)
By putting the HF transformer nearest to the anode we ensure that the HF is not lost to ground through the LF transformers primary to secondary capacitance.
Just an idea....
Cheers, |
|
|
| dave slagle |
| quote: | Originally posted by dhaen
In series:
The HF primary is low impedance to LF,
The LF primary is low impedance to HF.
Both transformers get what they want out of life;)
Cheers, [/B] |
the two primaries in series will form a voltage divider based on the inductance in parallel with the loads, with a perfectly linear inductance and a resistive load it could be worked out, but in real life i suspect you will get a mess... (that might just sound great)
dave |
|
|
| dhaen |
Dave,
We're just dreaming:idea:
Any idea what's happening with Doede's dual secondary transformer?
I was wondering if the HF secondary was underneath the LF secondary.
Cheers, |
|
|
| fdegrove |
Hi,
| quote: | | the two primaries in series will form a voltage divider based on the inductance in parallel with the loads, with a perfectly linear inductance and a resistive load it could be worked out, but in real life i suspect you will get a mess... (that might just sound great) |
Yes...a big mess...unless...:
You parrotfeed each OPT...well that's how I dreamt it up anyway..:cannotbe:
Pfffffttttttttttt,;) |
|
|
| Circlotron |
| A top-end-only transformer might be a good thing for a biamped or triamped system. Less turns for less capacitance as someone already said, and maybe the option of fancier core material seeing you arent struggling for inductance at low frequencies. It would be small too so not very expensive. Much easier to diy one too. |
|
|
| Brett |
| quote: | Originally posted by fdegrove
parrotfeed |
<i>rawwwkkkk!</i> |
|
|
| Brian |
| Thank You for all the replies fellas. It's great to have somewhere to turn for answers to audio questions. |
|
|
| dhaen |
Hi,
I gave this idea some tought overnight (sag git), and realised that it would be completely load dependant. The load on each secondary would affect the primaries impedance and hence the proprtion of signal they got.
Certainly a no-goer.:dead:
Cheers, |
|
|
| planet10 |
I started another thread http://www.diyaudio.com/forums/show...=&threadid=5655 which involved 2 OPTs -- not in the same manner thou -- in a single channel.
I also have seen a number of inexpensive stereo amps which introduced a 3rd transformer which was used to drive a bass speaker, (snippet of the circuit trace below). I am still not completely comfortable with my understanding of what is happening here althou i have my suspicions. What happens if you leave off the bass speaker or leave out the big OPT completely?
There was also a Crowhurst schematic published in a recent audioXpress that used dual OPTs -- it has me completely baffled.
dave |
|
|
| dhaen |
Dave (P10),
That is pretty much what I was thinking of. The word "inexpensive" doesn't fill me with much confidence about it's prospects. [Though it'll have Joel salivating]:nod:
And yes, it all falls apart if you disconnect the big speaker.
Cheers, |
|
|
| fdegrove |
Hi,
| quote: | | And yes, it all falls apart if you disconnect the big speaker. |
Even to the point of destroying the OPT when signal is present...
Much easier to build another set of amps and use the passive filtering at the input, I reckon.
Cheers,;) |
|
|
| planet10 |
| quote: |
Even to the point of destroying the OPT when signal is present... |
Thanx.
OK. I'll either leave it out if i ever rebuild it... make sure there is a bass speaker, or add another couple tubes and use it as a separate summed bass channel.
This one isn't high in the queue, but i was curious about it so had traced the circuit (i have learned much by doing this)
dave |
|
|
| Brett |
Designed and optimised for a specific application, by a <i>good</i> engineer (like Crowhurst), it could be made to work but it could be a real dogs breakfast otherwise.
Sakuma-san did a couple of designs with seperate secondaries on the same core iirc. |
|
|
| NickC |
| ok guys u mention a transformer with separate windings for lf and hf, how about 2 transformers connected in serial or parallel. Output impedance should be higher in serial i think double and half when in parallel rite. This way we can use small output trans for higher output wattage. Small output trans should have better hf response compared to its 1 single opt of the same wattage of the 2 connected together. I think parrallel is the one where the wattage is double, i thinking of using it. Am i going in the right way? looking at the 12ese they have multiple connections for different impedance ratio 10000,5000,2000 i think but only 80ma dc current. When planning for a se with kt 88 or even 300b the current wouldn't be enough. Hence an idea of using 2 12ese connected in parrallel, their ratio would be 10000 to 8. Hence when connected parallel togehter it should be 5000 to 8 roughly rite? Please advise, i am not aiming for excellent sound just for getting my feet wet in se design but need the power hence kt88. The current should be double the capacity to 160ma. of course in real world the trans would be excatly the same resistance so real current sharing would not exist so maybe one trans might draw more current. Please advise me on this? |
|
|
| jlsem |
Sorry, late to this thread. Paul Klipsch actually designed a dual output transformer 2A3 PP amp ca. 1946. The really interesting part is that a passive crossover appears in the schematic between the 2A3's and the two output transformers. One transformer was wound to advantage for LF and the other for HF. Both transformers had an impedence of 5K.
John |
|
|
| Brian |
| Thank You John, that is what I had in mind. |
|
|
| Kuei Yang Wang |
Koinichiwa,
| quote: | Originally posted by Brian
Can a power amp be built with two output transformers per channel? A big one optimized for accurate bass going to the woofer, and a small one optimized for clear and sweet mids and treble going to the wide range driver.
|
A way in which I can see this work is to have a LF optimised transformer which can have high leakage inductance and thus very low parasitic primary capcitance and to use the second transformer in parafeed with a small, high quality core (Nickel etc...)and your favourite wire. That might be interesting...
Another option is double up the output stages and give each it's own OPT.... ;-)
Sayonara |
|
|
| planet10 |
The snippet of a Grundig map shows 1 1/2 OPTs per channel with a common bass channel (i also have a Motorola amp with a similar scheme). The HF OPTs are quite small, the common one dramatically larger.
dave
BTW: Labtec just got a patent which -- as much as i can tell -- is essentially this scheme (but SS with no OPTs). |
|
|
| Kuei Yang Wang |
Koinichiwa,
| quote: | Originally posted by planet10
I also have seen a number of inexpensive stereo amps which introduced a 3rd transformer which was used to drive a bass speaker, (snippet of the circuit trace below). I am still not completely comfortable with my understanding of what is happening here althou i have my suspicions. What happens if you leave off the bass speaker or leave out the big OPT completely?
|
Well, a few things here. These schematics do interesting things. They are early "Sub/Sat" systems. The circuit works by having one channel inverted both in the drive to the output valve and at the output transformer. This makes the Amp still provide the correct polarity, but the anodes now operate out of phase, just like a PP Amp.
So, if you use a big fat PP OPT for bass (and note that the transformers primary sections are ALWAYS bridged by a capacitor) and a small and skinnly lowish inductance Transformer for midrange/treble, voila, you have an active Sub/Sat system with 4 times the LF power and a series X-Over.
It is an interesting concept for a small "phrugalphiles" Valve System. Take a pair of cheap 300B's, a pair of Hammond 125SE outputs and a nice spare 5k A-A Z PP Transformer. Set the X-Over as it falls and drive your Lowthers or whatever of the small Transformers suitably mismatched to have low primary inductance.
Again, the principle can be applied to single channels using one large "bass" Opt and one small "treble" OPT....
Sayonara |
|
|
| Brian |
| Kuei Yang Wang, thank You. These sound like good ideas. I will have to learn more about transformer design before I can implement them as I am still very much a beginner. |
|
|
| Kuei Yang Wang |
Koinichiwa,
| quote: | Originally posted by Brian
Kuei Yang Wang, thank You. These sound like good ideas. I will have to learn more about transformer design before I can implement them as I am still very much a beginner. |
Well, lets just do this a little bit theoretically.
First, lets decide upon a sensible Crossover frequency for our 2 Transformer series arrangement. Lets say we stick to the classic 80Hz X-Over to a bass. Lets use a 300B and a lot of people tend to run 3K nominal impedance Transformers on a 300B.
So, what we really want is to calcualte a series crossover at 80Hz for a 3000 Ohm impedance, where the HF transformers primary inductance forms also the inductance for the highpass Filter. We then want to know what value capacitor we need in parallel to the LF transformer.
A spreadseet to calculate (among others) 1st order series X-Overs is here:
FRD Consortium Passive Crossover Design Calculator
Lets load this and punch in our numbers....
This suggests 3H primary inductance for our HF transformer (meaning we can use a large airgap and a small core) plus a 1.33uF Capacitor across the LF transformer.
IN FACT, we likely could even make an absolutely outstanding "Fullrange" output transformer by simply connectingthe two secondaries in series as well....
Anyway, this should give enough ideas for now.
Sayonara |
|
|
| janneman |
| quote: | Originally posted by planet10
The snippet of a Grundig map shows 1 1/2 OPTs per channel with a common bass channel (i also have a Motorola amp with a similar scheme). The HF OPTs are quite small, the common one dramatically larger.
dave
BTW: Labtec just got a patent which -- as much as i can tell -- is essentially this scheme (but SS with no OPTs). |
Dave,
This is interesting; it seems the two small xformers play a major role in providing the feedback to the cathodes at the input side, thereby defining the closed loop.
Where do the wires go to at the right bottom side? Or did I miss that diagram?
Jan Didden |
|
|
| planet10 |
| quote: | Originally posted by janneman
Where do the wires go to at the right bottom side? Or did I miss that diagram?
|
They go off to the speaker plug-ins. (I have the entire map if anyone wants it -- Grundig Werke Furth)
dave |
|
|
| planet10 |
And the actual OPTs
dave |
|
|
| dhaen |
Grundig always were pig-awful at schematics.:eek:
They sounded good though :cool: |
|
|
| Brian |
Thank You Kuei, I have decided to make my main amp a little 1-1/2W job. So I guess I need to biamp to avoid the bass eating all my power. Still, it is good to learn of these transformer alternatives. Perhaps I can try one of them on my next project.
Brian |
|
|
|
