PhilGood
Juice box hero
Wouldn't multi chambered core interfere with flux? I thought that was the whole purpose of good iron with some nickel in the core was to allow the bass below 200hz to flow through without saturation.
Wouldn't multi chambered core interfere with flux? I thought that was the whole purpose of good iron with some nickel in the core was to allow the bass below 200hz to flow through without saturation.
Marik
Pro Microphone Design
PhilGood
Juice box hero
That's what I said...
Marik
Pro Microphone Design
That's what I said...
OOOOPS
There is definitely some confusion here... Since we deal with that stuff a lot, or at least use words like Lundahl, Cinemag, for better understanding the meaning behind, it would be nice to make a little excursion into transformer theory... I'll try to do it as simple as I can.
In the transformer we have many contradictory parameters. In the nutshell, we need to get both, LF and HF extensions, where in order to get good low frequencies we need to have good inductance, i.e. whether lotsa turns, or good core permeability. At first, seems no problem, as we could easily do either. But the catch is... we also need to get a good HF, and here we have two enemies--leakage inductance and capacitance. So the more turns, the more both of those we get. OK, then what about just use higher permeability material (i.e. higher Ni)--no problem, but then... the rule is the higher permeability, the lower saturation point. So in fact, the higher Ni, the more saturation we get in the bass.
For example, if we get for an audio transformer silicone steel as a core our saturation point would be very high, but we also will lack inductance and LF response, as a result... I even don't mention the distortions will be little bit crap.
With high Ni we can go around and by increasing the area of the core (stucking more lams) raise the saturation point, but then we also increase the length of the wire, thus increasing winding resistance and associated losses. In short, there is something to think about
. Special winding techniques deal with HF extension improvement, where in multichambered windings (as in Oliver's transformers) we divide the bobbin into a few sections and then winde them separately (as opposed to just winding along the whole bobbing's length). It is used to reduce winding capacitance and thus extend HF bandwidth.
Another winding technique called "interleaving", when instead of pri-->sec (like stock 6802 transfomer), we wind it as pri-->sec-->pri (or we could do even pri-sec-pri-sec-pri). This arrangement reduces leakage inductance.
Those both techniques can also be combined together, but it is very time consuming in manufacturing and would mean MUCH more labor, and you cannot do it as a completely automated process we find in the cheap "stock" transformers.
Hopefully it is more clear now and puts some light on the things.
Best, M
Last edited:
RightOnMusic
asshat
you got it.
Marik
Pro Microphone Design
Last edited:
Marik
Pro Microphone Design
The PSU top:
http://home.comcast.net/~markfuksman/TNC6802_PSU.JPG
and bottom:
http://home.comcast.net/~markfuksman/TNC6802PSU_PCB.JPG
Looking at that I couldn't help but chuckle, wondering: "What were they thinking?"
Pay attention free holes on traces and hand drilled holes--the capacitors were originally supposed to be in different location. Ones realized that the XLR connectors are in a way, they had to move the caps
.
Notice the diode accross the board and cut trace under it?--the usual trick to rase the 7806 regulator voltage.
Remember, I wrote we have 6.54V on filaments? Thank this diode
Remove it, and restore the trace. We will get about 5.92V on the filament--just about right for the tube long and happy life.
Best, M
http://home.comcast.net/~markfuksman/TNC6802_PSU.JPG
and bottom:
http://home.comcast.net/~markfuksman/TNC6802PSU_PCB.JPG
Looking at that I couldn't help but chuckle, wondering: "What were they thinking?"
Pay attention free holes on traces and hand drilled holes--the capacitors were originally supposed to be in different location. Ones realized that the XLR connectors are in a way, they had to move the caps
.Notice the diode accross the board and cut trace under it?--the usual trick to rase the 7806 regulator voltage.
Remember, I wrote we have 6.54V on filaments? Thank this diode
Remove it, and restore the trace. We will get about 5.92V on the filament--just about right for the tube long and happy life.
Best, M
Last edited:
Marik
Pro Microphone Design
Sorry, double post
Last edited:
antichef
pornk rock
awesome!!!
man, I wish I could do that.
I
ittam
New member
No pictures here with win explorer?
Matti
Matti
Marik
Pro Microphone Design
OK, edited as links
dgatwood
is out. Leave a message.
Notice the diode accross the board and cut trace under it?--the usual trick to rase the 7806 regulator voltage.
Remember, I wrote we have 6.54V on filaments? Thank this diode
Remove it, and restore the trace. We will get about 5.92V on the filament--just about right for the tube long and happy life.
If I'm looking at it correctly, you could just replace the diode with a jumper. Easier than rebuilding a trace.
Marik
Pro Microphone Design
Sure! Matter of personal preference, I guess.
RightOnMusic
asshat
The PSU top:
http://home.comcast.net/~markfuksman/TNC6802_PSU.JPG
and bottom:
http://home.comcast.net/~markfuksman/TNC6802PSU_PCB.JPG
Looking at that I couldn't help but chuckle, wondering: "What were they thinking?"
Pay attention free holes on traces and hand drilled holes--the capacitors were originally supposed to be in different location. Ones realized that the XLR connectors are in a way, they had to move the caps
Notice the diode accross the board and cut trace under it?--the usual trick to rase the 7806 regulator voltage.
Remember, I wrote we have 6.54V on filaments? Thank this diode
Remove it, and restore the trace. We will get about 5.92V on the filament--just about right for the tube long and happy life.
Best, M
pardon me for seeming noobish here, but once i remove the diode, what is involved in restoring the trace?
tarnationsauce2
Welcome to the jungle.
Just solder a wire in place of the diode.
Or if you want to get fancy, you could connect a toggle switch to short the diode. that way you can choose heater voltage.
I'm not sure of the sonic differences. Only difference I know of it tube life.
Marik, why is it that they did this? Is there a sonic difference? Or was that band-aid fix for some other shortcoming?
I just don't see the point in raising the regulator voltage because it will kill the tube sooner and not to mention it increased cost of parts and manufacturing.
dgatwood
is out. Leave a message.
Just solder a wire in place of the diode.
Or if you want to get fancy, you could connect a toggle switch to short the diode. that way you can choose heater voltage.
I'm not sure of the sonic differences. Only difference I know of it tube life.
Marik, why is it that they did this? Is there a sonic difference? Or was that band-aid fix for some other shortcoming?
It's a tradeoff. With the diode, you get 5% over the desired voltage, and without it, you get 5% under. I'd prefer to get closer to 6.3 than that, personally, but subbing in a variable regulator would be a little harder mod. I might be tempted to leave the diode in but use a pair of resistor to divide the voltage down.
Marik
Pro Microphone Design
Marik, why is it that they did this? Is there a sonic difference? Or was that band-aid fix for some other shortcoming?
Well, it is a touchy subject. Generally, I prefer underrheating to overheating. On the other hand different tubes react differently.
If you study some old Neumann/Gefell/AKG schematics you will find that for sake of noise the VF14 was used underheated, on the other hand 6072 feels happiest at exactly 6.3V, but again, the EC92 has better noise parameters underheated.
Since the 12AT7 is the same (only twin) tube as EC92, and if you deside to keep it in the mic, it should be used underheated. In this case replace that damn diode with a jumper already!
Or else, if you want to hang out little more and see what else can be done, don't do anything, yet, to see where our jorney can bring us...
Oh, and BTW, today is a scheduled capsule delivery, so I better hurry to finilize some preliminary things with the mic...
So stay tuned...
Somnium7
Noise Criminal
The diode trick with those regulators also subtracts about 6dB from the ripple regulation ability.
Taking out the diode will make the output from the regulator slightly cleaner.
Taking out the diode will make the output from the regulator slightly cleaner.
Marik
Pro Microphone Design
You don't need two resistors--one should do.
We need to drop 6.54-6.3=0.24V accross 0.3A filament current (for 12AT7, 12AU7, 6072, etc), so we will need 0.24/0.3=0.8 Ohm.
But wait a second, the 0.3A is for two filaments. Since we use only one half of the tube, and only one corresponding filament, our current is only 0.15A, so we will need a 1.6 Ohm resistor, which we might very well later use, should we deside to go with 6072 tube
. Best, M
I
ittam
New member
Slightly out of contects: what hapens if
put a 1:5 transformer in instead of a 1:10 or so
intended for that circuit, I want more gain without
side effects. Teach me if you will
Cheers
Matti
put a 1:5 transformer in instead of a 1:10 or so
intended for that circuit, I want more gain without
side effects. Teach me if you will
Cheers
Matti