NewYorkDave Mila Dual Tube Preamp Help Thread

[Animatic]

I'd flip the cap PS board 180 and move the power supply and switches over to the other side,
and then also move the DI side over with the mic input. The redo the face plate.

The OT and the PT can be in th same area as long as they are rotated
and their inductance fields are 90 degrees off angle, you have enough room there.
Move OT up front and PT to the rear and make sure they have the
iron's magnetic fields radiating at different angles. The PT should also
not have a field aiming at the tubes. Look at old Fender amp pictures
or old HiFi stuff for a reference.

DC heaters, good,
but also the grid plate cath wires crossing is not good. Good eye kh.
Not near as bad as the In Tranz covered with AC.

Don't feel bad tube amps are another world, and not necessarily intuitive.
I came to Prodigy because I met PRR at Hoffman BBS, he mention this place.
It's a space specifically about building tube amps
http://www.el34world.com/Forum/index.php
Lots of useful info and examples about tube amp design.
Mostly guitar amps, but high gain preamps are real hard to get right,
so it applys to your project. Maybe go check it out and look at some
of the links to finished amps.
Good amp dress is religion there. :grin:

Note on NYD's unit, if I remember correctly, his PS isn't even in the same box.

Don't worry you will get this running well and you WILL be happy.

I have the parts for 2 of these, but I am waiting for the custom iron to be made. And hoping they can do a input tranz properly here too.

 

[skipwave]

[quote author="khstudio"]#1 - Your input Tranz (Even if it's shielded) should be as FAR from the Power TX as you can get it :!: This could cause = HUM 60-120Hz :!: Especially with a Non Toroidal TX. [/quote]

I think Kevin is right on, look at your input transformer. Where have you grounded the case and core connections? The case should go to the nearest chassis connect. You could be dumping EMI into your circuit ground if it is grounded there.

I need to track down a little buzzzzzz issue in the NYD 6SN7 pre that I built. I remember a wise forum member saying a rule of thumb for starting to track down sources of noise: Hum means too many grounds. Buzz means too few.

 

[khstudio]

I posted this in the One-Bottle thread too:

Well... I searched for Dave's recent post & he hasn't even been here since (Sat Jan 12, 2008 11:03 pm)

It looks like CJ pissed him off in this thread. :sad:

http://www.groupdiy.com/index.php?topic=308906&highlight=#308906

WTF? I don't normally get to involved in this shit but CJ is just not cool sometimes. :evil:

Here we are building these NYD pre-amps that Dave shared with us & can't get his help & advise when we need it... probably because of that crap.

I like Dave, his circuits & advise :!: :!: :!:

I just finished a One-Bottle & could really use his input right now.

If anyone knows or has seen Dave, PLEASE let him know there are a lot of use around that appreciate him & could use his help. :green:

THANKS!!!!!!!

 

[Animatic]

I sent NYD a PM saying hey come back,

but who knows if he looks in or gets notices now.
No email or anything else but the in house PM.
We'll see.

 

[MasonAtom]

Here's my plan:

1) I'm thinking of pulling the PT and mounting it on the back of the case a la UA LA-2A, then running the AC power lines on the side of chassis, making sure they are tightly wound and clear away from the XLR input and the mic transformer.
2) Move the tube circuit board and the input transformer to the right by about 2 inches each.
3) Rewire the heaters and route these tightly wound wires in a manner so they don't intersect with key signal carrying wires.
4) Redo the lead dress on the tube sockets.

Sound like a plan?

I really don't want to redo the panel as I kinda want it as a reminder of what I had to do to figure out how to do the engravings, learn CAD, G-Code, backfillings, etc. It represents a lot of work to me... kinda like sentimental value I guess. It wouldn't be hard at all for me to redesign it and make a new one... but I'd just rather not.

I'm thinking one of the big issues is that the PT is not shielded so having it next to the mic input transformer was totally assinine on my part. Putting it on the outside separated by a metal partition should help, no?

Mason

 

[Animatic]

Sounds like a plan.

I understand the sentiment idea,
but sentiment and memories can hang on the work room wall.
My 1st priority is always
'Does the tool work properly.'

If you are going to leave the AC along the side,
you might think about putting a small 45 degree
electrostatic screening panel over them.

A doubling of the distance causes a 30% lowering on inductances,
I think it''s something like that, I will have to check a book.. forgot.
But you get the idea. PRR will know it from memory...

Also if you turn it on and turn the mic transformer 90-180 degress
with the output to a console or such and a speaker turned up loud,
you will likely find a place with the least hum, so then tighten it down there.
The coils should be at 90 degrees from a possible receiving unit.

If you can crank it up with that transformer freein hand, you can find
the best spot for it or if you have a spot that works at all.
But don't forget it's leads can have some inductance too.
You may want to put them in a small metal tube shield to ground also.

Heater wires could run on the floor at the front bottom edge
and only rise at the tubes Similar to Marshal style. See below.

Cathode and anode and grid all like a bit of separation
a centimeter at least, a half an inch better. Short as possible,
but enough space to clear each other. See below.

Power supplies are grounded at one end of the chassis,
and preamp tube stages are grounded at another point on the chassis,
to separate the grounds just a bit. Another anti-hum method.

Here is a Carolina Blues Express amp from Tubinit a moderator at Hoffman.
Credit where credit is due for nice lead dress.

He uses the up and over fender heater style,
But notice his tube in and out wires and his separation of sections in the chassis.
http://i28.photobucket.com/albums/c216/tubenit/CarolinaBluesExpresschassisinterior.jpg
His amps sound lovely. A good example.

And some amp porn from DWP
http://i304.photobucket.com/albums/nn191/dwpatte/howlin40002.jpg
The white low level cables are shielded/ chassis grounded at pots or plugs
since they cross over a lot of space. But all of it pretty DC. Shielded at one end anyway.
Same for the deep red, which I think is a dual ganged master volume.
And a black ground wire wrapped around it no less.

And something Marshal style, in close quarters like yours, from Blind Lemon
http://www.el34world.com/Forum/index.php?action=dlattach;topic=3717.0;attach=2023;image
http://www.el34world.com/Forum/index.php?action=dlattach;topic=3717.0;attach=2024;image

All these examples can help you. Tubes is another world, but a cool one.
I thank these men for posting publicly on www.el34world.com
so others can see how it's done well. Sure has helped me.

Hope this helps you.
:grin:

 

[MasonAtom]

[quote author="Animatic"]

Also if you turn it on and turn the mic transformer 90-180 degress
with the output to a console or such and a speaker turned up loud,
you will likely find a place with the least hum, so then tighten it down there.
The coils should be at 90 degrees from a possible receiving unit.

If you can crank it up with that transformer freein hand, you can find
the best spot for it or if you have a spot that works at all.
But don't forget it's leads can have some inductance too.
You may want to put them in a small metal tube shield to ground also.

:grin:[/quote]

I plan on using Stephen Keller's "Hum-o-phone" trick for the transformer placement, which is really just a variation of what you mention here.

Thanks for the help, it is really appreciated. I will post update as I continue to work on this.

Mason

 

[timmygrimm]

Just got back up and running in DIY land (after a couple of moves around the country!) and am revisiting my MILA. Still wondering about the question below. I'm also very interested in wiring up an output "fader" style pot so I can control how hard I'm pushing the tubes vs. how hot the output will be.

I realize these are somewhat complex questions which require some explanation... perhaps there is a link to another thread regarding wiring up a fader (or PS issues), or an external link?

I'll also take this opportunity to thank everyone at groupDIY for getting me into this whole thing. I've learned a ton!

Tim


[quote author="khstudio"][quote author="timmygrimm"]Hello Dave, etc...

I asked this question earlier but may have been buried... If I am building a two channel unit, do I need to build a separate power supply for the plate voltage of each channel (sharing nothing but the power transformer)? Forgive my lack of experience with such topics. This question was addressed in another thread but no conclusion was reached.

Thanks everyone for your help!

Tim[/quote]

Bumping this because I'm building a "Stereo Unit" too.

Should the (Individual Plates) B+ to the Left & Right channels be decoupled :?:
& if so... Could someone explain how best to do it??? :green:[/quote]

 

[Animatic]

PRR is the BEST reference for this but I'll try.

OK, If the power transformer can handle the combined amperage of ALL tubes,
then you can use separate IC chains for each preamp.

Each IC chain will create an isolation from the other :
cap / resisiter, cap / resister etc, Two lines of same in parallel.
and thus the PT recharges the TWO cap chains,
but the caps feed the tubes as needed.

I have seen similar in a couple of guitar amps, and also dual voltage systems
where you have a 2nd diode bridge and cap chain for a SS unit and one for the tube unit.

email PRR and you will get an definitive answer I am sure.

 

[timmygrimm]

Thank you Animatic. I'm assuming that the benefit of isolation is that if one channel is working extra hard at a given moment, it will not cause a power sag on the other channel. I would certainly like to achieve this... are there any downsides (other than more parts, build time, etc)?

As far as the fader situation goes, I've searched a bit and the best thing I can find is using an attenuator like this:

http://www.twin-x.com/groupdiy/albums/userpics/600ohmatten.gif

Also from our designer! Wondering if this is the best solution... or if there is a better (or easier) solution involving a single rotary switch or pot. It seems easy enough to convert NYD's design into a single rotary design, but not sure if steps would wind up too dramatic (or subtle?)... maybe 5 10dB steps would suffice?

Thanks again,

Tim

 

[Animatic]

PT / rectifier / then 2 identical cap resister bridges as explained in his plans.

Yes no sag is the goal. If there is enough CAP charge capacity then the PT won't really notice.
And mic preamps don't want ANY sag or interaction. IMHO

I suppose if there is a REALLY overbuilt single chain 2 pres running off it will hardly be noticed,
but cheaper to use normal sizes and do two chains.

Another way is an initial big cap or cap pair,
and then two smaller cap lines off that for each preamp tube.
Usually the output tube gets a PS cap,
and then the input tube gets a PScap
which isolates BOTH tubes within the same pre amp section.

Are you trying for pre T1 attenuation? Like extending SW1 20db pad for more control?
R5 could become a 1 meg pot, with wiper to grid stopper resister.
or a pot to ground next to or before R5 would reduce signal to V1a.


I don't think you need a balanced solution except before t T1.
NYD has a simple increase of resistance there with SW1
Making SW1 triple throw would add extra attenuation choices.

Or maybe I don't understand your aims.
Is this a 1st stage drive control your wanting?
NYD just has a 220R pad across input of T1
Any drive sound will be from V1B
not V1a since the signal will likely be too low there.
Hence VR1 being after V1a.

My attenuation problem is even deeper.
I want REMOTE attenuation of 1st to 2nd stage gain controlled from 30 feet away.

I have found some Vishay voltage controled resisters VCR2n4n7n
and will try to install them like pot VR1, but with a +/_ 5v control voltage
going up a separate line to the control room and a pot there.
I will be bread boarding that after I finish my Dumble clone amp.

But I also have some Analog Devices resister CMOS chips that might do the same thing.
I am planning both SS and tube pres.

Also I am waiting for some Cinemag Input T's that got sent to Boston
not Thailand...

 

[khstudio]

Hey Tim,
I actually built a ruff version of the One-Bottle pre & really like it.

It's built in a pretty large chassis so I could try the MILA circuit too to see which I liked better. Since the Mila & 1-Bottle have the same voltages, the same PT works for both.

About the Power Supply:
I think you can just build one supply BUT drop (R1) 2.2k.
Not sure what to, maybe 1.2?
Dave (or someone) mentioned (in the 1-bottle thread) dropping the same resistor in the PS for running 2 channels.

Anyway... did you try the circuit yet?

I think I may just build a dual 1-bottle. I like the character it adds... I "think" the MILA would be on the "Cleaner side & I think Dave even mentioned this.
I've got enough High gain pre's.
The DI is KILLER on keyboards, etc...
& it does wonders for adding a nice/fat tube character to my U87 on vocals.
3 of my friends picked the 1-Bottle with the 87 on vocals every time.

Should rip on guitars & snare too... ALL of these don't need a lot of GAIN so... you get the point.

Just thought I'd share.

:green:

 

[Animatic]

Hey KHS thanks for more input.
And clear feedback on working units.

NYD has simple and clean designs that's why I gravitated to them.

Got almost all the part for two Mila's now, waiting on Cinemags still,
but also have some Oxfords coming.

Here is some of my musings on Mila as a near mic preamp but controled from afar.

 

[Animatic]

Not to be exactly on line here is something from PRR from Hoffman BBS
about another amps power supply. But he's da man for these explanations,
so I add it to your information trove here.

PRR from Hoffman BBS.

> looking at ripple current because I'm led to believe this is
an important feature in a power supply.

The first cap has high ripple current.

"Generally", audio amps don't have trouble. You need a "large" capacitor
to get ripple voltage low enough for clean audio.
When you do that, the cap is usually physically large enough to throw-off
the heat caused by ripple current.

In this case I compute ~~150mA ripple for 47uFd, and that series' 47uFd 160V-200V part
is rated 270mA ripple current.
When we go 470uFd, rms ripple current is 270mA,
but the 470uFd cap is good for 1,400mA ripple.

And ripple rating is at maximum temperature.
Yes, sitting near a 50C5 may push your luck.
Small hot chassis like these do present challenges...
but OTOH, we're talking 1-buck caps and ten bucks for all the electronic parts (at 1960 prices).
So when stuff blows, you replace it.
But I'd expect these 1-buck caps to give decades of service.

> And ESR.

Not a big deal for 50/60Hz audio supplies. The cap-makers know what we are doing.
They hold ESR down low enough that filter-action is not impaired.
ESR impacts ripple-current rating, but that's covered in another spec.
(It does become an issue in high-frequency supplies and some more
delicate audio-apps than a 50L5 cooker/amplifier.)

> I don't know if the specs are up to snuff,

Ten times better than anything we coulda got when these amps were made
(and 30 times better than what Alamo et al put in there).

> I wonder what the specs are on the original caps found in some of these old amps?

"Low price!!! Not TOO much hum or smoke for 30 days.
Can you do better on the price???"

The un-told history of electronics is the penny-pinching.
After the initial rush to radio (or e-guitar), the only way to get and keep
a foot in the market was to sell for $27.99 List something comparable
to the other guy's $29.95 List model.
(Or offer 27% sales markup when the other guy could only do 25%.)


> go easy on the first filter cap after that 35W4. I think it wants not more than 40uf.

Depends on the series resistance. There is some value of R at which
the C can't hurt the rectifier. This actually was a problem when we
plugged RIGHT to the wall-outlet (no PT at all). You sometimes see part
of the rectifier heater in series with the path to first cap (cheapest "resistor" around!).
But the 35W4, and these low voltages, don't really give trouble.
Even at 470uFd. And "trouble" is $7 trouble, not worth a tear.

It has some of the logic of why certain caps were used.
Maybe not perfectly on mark, but useful.

 

[timmygrimm]

Thanks Kevin and Animatic!

That is good news about one supply. I am using the Hammond 270CAX which has 70ma current - more than double the rating of the Stancor used in the single channel.

I have yet to build the circuit as I am still w/o output transformers... but I've got most of it going on perfboard... got the Beyer i/p iron and the tubes, case, so I'm almost there. Will definitely post some results once I am finished.

Kevin, if you happen to see this, I noticed you used some software for doing a frequency sweep on your LA2A in a different thread... what software is this? PC/MAC? Seems it would be interesting to check out... though obviously I'll use my ears too!

Gonna try to cram it all in a 2U case too, with tubes on the back a la LA2A. If anyone sees a major problem with this please let me know.

By the way, the attenuator I am talking about would go after the output transformer (I believe) and would be to lower the signal level just before hitting the DAW, to allow me to dial in tube crunch without digital overs. I love distortion on drum room!! Something like NYD's attenuator would seem to work:

http://www.twin-x.com/groupdiy/displayimage.php?pid=1714&fullsize=1

I would prefer a pot-type situation (that's potentiometer, folks)... but don't know if I'd need to drop dollars on something like:

http://www.surplussales.com/Equipment/AudioAttenuator.html

And are those above even balanced?... and am I going on too long in this post? Yes... Too much soldering today... Sleep...

Thanks again everyone!

 

[Animatic]

Not sure if you can cram that Hammond into 2U...
Serious overkill, you could run 4 Mila's off that I think.

One of the reasons the torriodal types get used is hieght restrictions.
And with those separating the AC section from the DC / preamp section
with a grounded partition seems good practice.
Depends on the transformer of course.
Good luck with the whole thing.

 

[timmygrimm]

BIG version : http://www.bravadousa.com/testing/tim/images/MILA2.jpg

Here is my MILA in it's current form. Don't laugh too much at the layout (or at least let me know what to improve). It's the first time I've done one myself.

Fired it up without tubes and no explosions thus far. My B+ voltage is showing 370 volts (again, without tubes installed)...

Is it safe to put in the tubes with this plate voltage? The maximum rating on the datasheet is 300v.

 

[Animatic]

Hey Tim you're moving dfaster than I am!
Looks good so far. I need a case too.

Always good to use a 100w lightbulb and lill switch
in series with power line for your 1st start ups.
if it glows bright kill the switch and look around for the trouble.
Once it lights you got a problem, but a little glow is not an iaause.


I have been in cabinet building mode and also woodshedding
for a jazz gig that flew back over the horizon. I did parts out
a Gyraff G9 pair, but gotta do the Milas first, with the VCR2N4N7N
variable resistor for remote gain control. At that point I'll know
how it should work with the Gyraffs too.

 

[MasonAtom]

BIG version : http://www.bravadousa.com/testing/tim/images/MILA2.jpg

Here is my MILA in it's current form. Don't laugh too much at the layout (or at least let me know what to improve). It's the first time I've done one myself.

Fired it up without tubes and no explosions thus far. My B+ voltage is showing 370 volts (again, without tubes installed)...

Is it safe to put in the tubes with this plate voltage? The maximum rating on the datasheet is 300v.

Tim, cool to see that you've gotten started on this.

370V B+ at the first node sounds about right without the tubes in. I think mine was in that vicinity +/- about 15 V without the tubes. Also, remember that the plates on the tubes won't be seeing 300V+ at the quiescent operating point, as there will be a voltage drop across the anode load resistors.

Mason

Mason

 

[MasonAtom]

Quick update on my Mila. I've had a bit of time here and there over the past couple weeks and have done some modifications to try to address the noise issue.

First thing I did was move the PT to the outside of the back panel of the chassis and route all the AC mains wires along the inside of the chassis as far away from the input transformer as I could (without having to move the input traffo).

That made a difference in the noise, cut the noise floor down by about 25%, but it was still rather noisy. I then rewired the tube sockets and redid the DC heaters. That helped a tiny bit more. I tested a couple dynamics and a condenser and noticed that there was A LOT more noise on the dynamic mics.

Sooo.... I took a close look at my phantom power circuit and noticed that *lo and behold*, I had a ground loop in that part of the circuit... so I'm a dumbass! Fixed that and it turned out to be the BIGGEST culprit in terms of noise. Who woulda thunk?

Still not 100% noise free but it sounds great on electric guitar cabinets or anything with a rather large input signal. I'm going to clean up the wiring around the tube sockets some more (shorten leads and make sure of no grid/plate wire crossings), and add shielded cable on the XLR input and first grid.

Almost there... just need it to be a tad quieter so I can use it on acoustic guitar.

Mason