Bravo 12AU7 Mods and Appreciation thread

[Deleted]

It appears a rather large current has passed through the red LED and 2N3906 transistor (it being 'fused together' is the give away).
I suspect there is something wrong with the LM317 and/or MOSFET.
The tube is unlikely to be the culprit and also not damaged by this.

If it were mine I would replace these components before firing it up again:
red LED
2N3906
MOSFET
LM317

and check:
output cap (if it is not shorted)
7R5 resistor near LM317 (for continuity and resistance value)
220 Ohm resistor between 2N3906 and Gate MOSFET (for continuity and resistance value)

Since modding has been performed check soldering for 'shorts' or bad joints.

[suloveu]

Mar 6, 2011 8:02:41 GMT @solderdude said:

It appears a rather large current has passed through the red LED and 2N3906 transistor (it being 'fused together' is the give away).
I suspect there is something wrong with the LM317 and/or MOSFET.
The tube is unlikely to be the culprit and also not damaged by this.

If it were mine I would replace these components before firing it up again:
red LED
2N3906
MOSFET
LM317

and check:
output cap (if it is not shorted)
7R5 resistor near LM317 (for continuity and resistance value)
220 Ohm resistor between 2N3906 and Gate MOSFET (for continuity and resistance value)

Since modding has been performed check soldering for 'shorts' or bad joints.

Thanks solderdude,

anyway,
I think I found the cause of this problem.
When I replaced the PS caps a month ago,
I intentionally left the pin uncut, and I just bend it over to a "safe" place.
Before problem occured, the last mod I've done was replacing the output caps.
The accident probably happened when I put back all the cases together, and the uncut pin (which is also connected to red LED (+) pin) shorted with the switch pin, I think the red LED in this way was directly connected to the power supply

So,
I already ordered the 3906 and I already have some red LEDs,
I will check the output caps and resistors for the resistance and continuity also.
Any trivial way to check if the IRF510 and LM317 are still OK or not?

Many thanks

[Deleted]

hmm.

In that case the amp would have turned off and nothing would have been broken as the + of the cap is connected to the red LED already and also to the switch so nothing would have changed, you just wouldn't be able to turn it off through the switch.

I still suspect something with the output cap/ LM317/MOSFET, but could be wrong...

Since the MOSFET and LM317 cost very little I would simply replace them, just to be sure.

The MOSFET could be checked but not with a multimeter.
The same goes for the LM317.
LM317 does have overcurrent protection and temp. protection but still could become faulty.

[suloveu]

Mar 7, 2011 7:16:29 GMT @solderdude said:

hmm.

In that case the amp would have turned off and nothing would have been broken as the + of the cap is connected to the red LED already and also to the switch so nothing would have changed, you just wouldn't be able to turn it off through the switch.

I still suspect something with the output cap/ LM317/MOSFET, but could be wrong...

Since the MOSFET and LM317 cost very little I would simply replace them, just to be sure.

The MOSFET could be checked but not with a multimeter.
The same goes for the LM317.
LM317 does have overcurrent protection and temp. protection but still could become faulty.

Hmmm... in fact the amp intermittently switched on and off at the moment,
only after I took it off the plexiglass, the amp went steady but I lost 50% of the volume.
Well, I can be wrong on guessing the cause anyway

Things went worse when I realized the 3906 was shorted, one of the tube filament is not glowing anymore

I will order the LM317 and MOSFET then, the problem is I should wait until next week to get them because the local stores don't have any of the MOSFETs...but in the meantime I can use the 630 for testing.

Is LM317T or LM317BT OK for this?

Thanks dude

[jacjan]

Got one of these on the way from ebay.

Thinking of doing these mods:
- Replace the power supply cap with a 4700iuF / 35V nichicon
- Replace the two 1000uF units with 2200uF 105C nichicon units
- Yanked out the two small caps on the input side, right next to the tube and ran jumpers
- Replace the 630's with IRL530PBF mosfets
- And cut some traces and so forth per the crosstalk mod

Are these parts something I can get from Mouser?
If so, do anyone happen to have partnumbers for the recommended parts?
It`s my first tube amp, and my first mod project, so I feel pretty unsure of various brands and types of components.

Also, I believe I should replace the 12au7 tube the amp gets delivered with.
I got Grado SR-225`s, any tips to brand and variant of the tube that should be ok?

I have a usb dac I`m going to use aswell, but might get a better usb dac soon.

Thanks!

-j

[Deleted]

When you plan to mod it and roll tubes I suggest to replace the CCs resistors for 5k pots.
This way you can adjust the anode voltage.

just type in the parts numbers/description and they will pop up.
Also consider Panasonic caps (105oC)

[jacjan]

Mar 7, 2011 19:44:42 GMT @solderdude said:

When you plan to mod it and roll tubes I suggest to replace the CCs resistors for 5k pots.
This way you can adjust the anode voltage.

just type in the parts numbers/description and they will pop up.
Also consider Panasonic caps (105oC)

I`d probably manage to blow something up by changing to 5k pots

-j

[suloveu]

Solderdude,
I've changed the MOSFET to IRL530, brand new LM317s and 2n3906s...
MOSFET and LM317 also become hot which I assume they do have current supplied to them, dont know the exact measure though.
all caps have no shorted indication, all R also have their resistance...
but still no luck, the amp won't give me its original BIG gain... I have to turn the vol knob to around 3-4o'clock to have a bit amplification from the original source... and distort also

Anyway, I measured the red LED resistance and it shown me around 200kohm and 9.6V, is it the right value? I still concern about the red LED, since I don't have exact spec on it...

TIA

[Deleted]

the LED should have about 2V on it when 'on'
If there is 9V on it there won't be much sound and it will be distorted.

you cannot measure a LED on the resistance measurement mode.

I reckon the red LED does not glow ?
Both channels do the same ?
the blue LED is glowing normally, dim or off ?
both filaments glowing again ?

[suloveu]

Yes the red LED doesn't glow,
both channels sound balance volume-wise yet distorted.
Blue LED was glowing normally and both filaments also glow again...

So I think the red LED "consumed" too much power on it, so other parts of circuitry don't supplied sufficiently? Can I replace the LED with an R or diode? What is the spec should the R or diode be?

Thanks

[suloveu]

Solderdude,

My bravo has fixed and back to its fabulous and powerful sound!!!

I tried to swap the LEDs (bought around 20 red LEDs and 20 green LEDs last week), at the third try it suddenly magically glows up, measured the voltage drop on the LED, got +/-2V, plug in source and cheap cans to test the sound and yep it sounded just as good as before...even crispier due to IRL530 being used here.
Although I still dont know the root cause, but I'm happy enough to get this tiny amp back to its max performance... Cheap LEDs (got it for 5 cents per pc) have made the troubleshoot become much harder.

Many thanks to you sir! Really appreciate your helps


Here are the mods I've done anyway:

1. PSU caps: Nichicon KG Gold Tune 4700uF 50V
2. Output caps: Nichicon Muse 2200uF 35V
3. Input caps: Wima MKS4 4.7uF 63V
4. MOSFET: IRL530
5. Tube: '50s RCA Black plate 12au7 tube
6. Crosstalk mod by cutting trace and shorting path
7. Replaced 3k fixed R with Bourns 5k trimpot

[Deleted]

Glad you got it working again.

the LED is direction sensitive.. If you solder it in the wrong way around you COULD end up with about 9V voltage on it (and it not glowing) as it is close to the 'reverse voltage' of the LED.
long leg to the + of the circuit, short leg to the BASE of both transistors.

You can replace a LED with about 4 diodes (1N4148) in series if you don't want the red glow for instance.
Bias would need to be re-adjusted in this case as the voltage isn't the same.

the LED does have it's perks:
You can 'see' the part of the circuit works as aspected.
the CCS current is more 'stable' over the temperature range of the amp (and thus the bias voltage)
It is slightly less noisier then 4 diodes but in this design the tube is already FAR more noisier then the CCS circuit.

You can reduce 'pop on noise intensity' and the noise of the red LED by placing a 100uF/6V (or 10V or 16V) electrolytic cap directly across the LED (soundwise should make no difference).
be very aware of the + and - of this cap.
In the 'modding of a miridiy' pdf it is mentioned I thought.

[suloveu]

I would love to have the LED as the indicator for working circuit, so I dont think I will replace it. My previous question of diode (or R) as substitution was more as a hopeless option as I failed several times with the LED things
I have noticed the + and - of the LED too,
it's the first thing I pictured on my mind when I removed the original LED. But yes on my trials I've also tried to swap the + and - leg position.

Anyway,
I will try to add the 100uF caps,
I think I still have an idle nichicon 100uF/16V somewhere

edit: Add the amp's picture


Thanks

[tacoboy]

Any mods for Muse Amps?

[Deleted]

Muse is the same as the first Indeed, Bravo and Miridiy.

Modifications for the Miridiy (Indeed, Bravo e.t.c.) amplifier:
www.mediafire.com/?vzdwnrlnjjz

[tacoboy]

I''m currently using the Muse 6922EH/IRF630 model and looking to by a newer (with IRF510) headphone amp. is the 12AU7 considered a better tube over the 6922EH?

[PinkFloyd]

Apr 18, 2011 19:31:17 GMT tacoboy said:

I''m currently using the Muse 6922EH/IRF630 model and looking to by a newer (with IRF510) headphone amp. is the 12AU7 considered a better tube over the 6922EH?

6922EH is current production "6N23P-EB".... they are not at all bad..... I prefer the 6N23P (6922) to the 12AU7, there are more available variants..... 6N11, E88CC, ECC88, 6DJ8, 6922, 6N23P etc. etc...

The Electro Harmonix will sound a lot nicer (more transparent) with the IRF510.

[ilooli]

Hi Everyone,

I now own a little Bravo V2 and it is my first tube/valve toy.

I've been looking at oohms's mods and will slowly do all of them.

First I would like to reduce the noise that is coming from the PSU.

Does anyone know what the inductor is? It it an off the shelf thing or do you make it? If so what do i need to make one suitable for the Bravo?

Reference, circle thing under the switch

Thank you

PS: do tubes/valves glow? If i take out or dim the blue LED will the tube emit some kind of light?

[serenith]

Apr 22, 2011 13:05:24 GMT ilooli said:

Does anyone know what the inductor is? It it an off the shelf thing or do you make it? If so what do i need to make one suitable for the Bravo?

Reference, circle thing under the switch

I'd also like to know what this inductor is or if coiling a wire 30 times around a bolt or washer would make a satisfiable replacement?

Also, I bought a pair of IRF520 and IRF610 from my local shop and I was wondering (the only thing they had). Which model would you recommend I use ?

[Deleted]

IRF610

the inductor is self wound on a ferrite and will suppres RF and noise spurious above 100kHz.
Noise within the audioband will not be filtered.
It could theoretically help when you have RF poluting signals coming thru the mains.
Still won't filter anything coming in through the line input.

With most tubes you can see a faint glow when the blue led is off.
Leave it in there ... just short it (put a short wire over the legs).
When you remove the wire again the LED will function normal again.
Do not remove the LED, the amp wont work without it (unless you remove it and put a short in it's place)

You cannot use a bolt or metal washer.
It needs to be a ferrite and there are different types of ferrite for different bands to be filtered. (they peak at a certain frequency)

[ilooli]

Thanks solderdude,

Is this what I want? www.jaycar.com.au/productView.asp?ID=LO1242

It is 15mm wide, the hole is 8mm and the height is 6.5mm

I don't quite understand how this works, does the size of these things matter? does wire need to be wound around these a certain number of times?

[Deleted]

the physical size matters when higher levels of current (thicker wires) are used or bigger DC currents flow as saturation will occur.
For this application a small one will do but these toroid filters are more intended for AC line filtering then DC line filtering.
about 10 to 15 windings should do.. don't do more as saturation from the core (the DC running through the coil) is lurking and can completely eleminate the workings.

Here is a prebuilt one, jaycar: CAT. NO. LF1270 but are intended for AC power line operation and can be found in the mains line of dimmers, SMPS, motor controls etc.

these cores have ferrite and iron in it for a 'wider' filter range: CAT. NO. LO1230

I sincerly doubt it will do anything with this amp, most certainly not in the audible range and if I would want to apply one I would have used a so called common mode choke which has 2 windings.
One for the ground connection and one for the 24V DC so a choke with 4 connections instead of 2.
These also do not suppress noise in the audible band at all, they simply have no effect below 50 kHz and would only be there to prevent common mode signals from entering the audiochain through the headphone amp.
a single coil simply does not do that as the ground connection is not filtered ...

[ilooli]

Hi solderdude, i guess i won't bother with the coils.

So far i have only done the crosstalk mod. Am planning on doing the other mods but am having trouble sourcing IRL530's (still looking). I will probably do the other mods soon.

Thanks for your help

[PinkFloyd]

I am a great believer in ferrites / chokes and ALWAYS use them.... they certainly don't do any harm and a lot of people DO hear an improvement even though the measuring tools suggest nothing is happening.

Here is an interesting document to read:

freespace.virgin.net/mcg.mcg/ferrites.pdf

As always.... use your EARS to decide what is best for you.

Mike.

[ilooli]

Apr 27, 2011 0:30:16 GMT PinkFloyd said:

I am a great believer in ferrites / chokes and ALWAYS use them.... they certainly don't do any harm and a lot of people DO hear an improvement even though the measuring tools suggest nothing is happening.

Here is an interesting document to read:

freespace.virgin.net/mcg.mcg/ferrites.pdf

As always.... use your EARS to decide what is best for you.

Mike.

Hi Mike,

What configuration do you recommend for the choke?

Currently with no inputs plugged in i can hear a very very faint clicking/tapping which i assume is from the el-cheapO PSU. Will the ferrit/choke fix this?

I'm in the process of getting 35V capacitors for the 6800uf and 1000uf. Apart from the fact 25V is a little too close to the 24V going through it, will the 35V improve anything?

Thanks.