PSU +24V +48V issues...need fast help

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

RayBlack

Well-known member
Joined
Oct 13, 2009
Messages
50
Hi there,

I'm experiencing several problems with my PSU design, that I combined from this source http://www.barthman.de/rampedphantom.html
and a friend's advise.

The psu is used for racked V276a,V376b or V672 preamps.

My schematics looks like this:
http://www.baziany.de/DIY/PSU_Schematic_030110.pdf
http://www.baziany.de/DIY/V276a_Input.pdf

I'm supplying 2 preamps, with +24V of this psu and also the six relais for 48V, PAD and Phase, plus the LEDs.
The +48V goes to the Mic Ins and is blocked with to 220uF Cs in the direction of the mic pres.


Problems:
- I get a constant 50/100Hz hum and I think what you call HF oscillation on the mic inputs, when using both 24V and 48V circuits.
- Yesterday I connected chassis ground to 0V of the psu and smoked a diode I think.

Questions:

- Is it even possible to get +24V and +48V out of this xfmr ?
- What kind of wiring would you suggest for the two circuits ?
- Would you connect chassis ground, signal shield, ground of preamps and 0V to get rid of the hum ?
- My xfmr supplies 416mA. Is that too much for the 48V phantom ?


Would appreciate some thoughts on this, as it is my first DIY project and I'm very insecure about many things.


Thanks !
Ray
 
Sorry if I was unclear, but I've already finished the pcbs and want to use them.

Next time I'll look for the JLMs but this time I need to get this working.

Ray
 
RayBlack said:

I would strongly recommend against using this scheme. As shown it takes seconds for Phantom to ramp up, and not all mikes will like that. Whether the circuit works at all also depends on hard-to-control factors like capacitor leakage current.

There are a few discussions linked from the Mic Pre META on whether slow-start P48 makes sense at all; the (oversimplified) bottom line would appear to be no for anything more than a ~100R resistor charging a ~100u buffer cap.

RayBlack said:

Try to imagine what happens when you have the two grounds connected together.

My suggestion for a quick fix: change your circuit into this, but leave out all the -18V-related parts (ie the bottom 1N4002, the 7918 and related capacitors).

(There are quite a few errors in your schematics, but I'm not sure if those are drawing errors or actual bugs in the circuit. If I have to mention one thing, 47uF is much too small as a decoupling cap for the 7824).

JDB.
 
jdbakker said:
(There are quite a few errors in your schematics, but I'm not sure if those are drawing errors or actual bugs in the circuit. If I have to mention one thing, 47uF is much too small as a decoupling cap for the 7824).

And do NOT lift safety ground (M) from case! If Load (L) wire comes loose and touches the case, you have a LETHAL 230V on your case (and possibly on everything connected to it). Ground switch should separate audio ground from case/chassis.

Good luck!

/Dave
 
@JDB
I guess I have read every thread in the phantom power section of the Mic pre Meta, and I am familiar with the different POV about ramped phantom usefulness.
I was under the impression that though not justified for some, proper ramping couldn't hurt...
But you wrote: "...it takes seconds for Phantom to ramp up, and not all mikes will like that"
Does it means that ramping could actually damage some mics?
I'd be glad if you could explain or point me to the right thread...

Axel
 
mad.ax said:
Does it means that ramping could actually damage some mics?

I don't think it'll cause actual damage (although I'm obviously not giving any hard guarantees here), merely 'unexpected results'.

For simple, traditional phantom-powered condenser mics not much will happen. These mics use the P48 directly to polarize the backplate, and slow-ramping will simply cause the mic the mic to have a slowly-increasing sensitivity. With the circuit linked to earlier this might mean that you have to wait a minute to get full performance but that's usually no big deal.

More modern condenser mics have a switching regulator for polarization, in part to be less dependent upon the exact phantom voltage. This is where things can get interesting. To oversimplify things: many regulated SMPSes present a constant-power load to their input: less voltage=more current (or, roughly speaking, negative impedance). While these often have undervoltage lockouts, they are usually designed to 'see' an input voltage slew rate of maybe a volt per millisecond, or 50-100ms to get to full P48, as you'd get with a simple RC filter. The circuit as linked has a ramp rate that's about a hundred times slower, which can confuse such switchers. This 'confusion' can manifest itself as a burst of oscillation at startup, sustained oscillation or complete lockup until power is removed and re-applied.

(It is worth noting that 'smarter' regulated switchers tend to suffer more from this than simple transistor/CMOS-gate charge pumps).

Does that help?

JDB.
[on my way out the door, so apologies if the story's a bit unclear]
 
It sure does help!

See, I became interested in ramped phantom out of concern that abrupt switching may damage vintage transformers such as the V76's input...
I never thought about modern SMPS possible issues until now. Many thanks for pointing that out!

Axel
 
I changed the circuit yesterday, by replacing the 47uF to 1000 uF /470uF and getting rid of the ground lift.
I also changed it to seperate xfmrs, 1 for +24V and one for +48V which has only 63mA current.
www.baziany.de/DIY/PSU_Schaltplan_050110.pdf
www.baziany.de/DIY/V276a_Input_050110.pdf

Now the strange thing is that I have constant 48V on my mic Input...why is that so ?

The PSUs and vintage preamps shall be used with historic ribbon mics such as r44 or r88dx, so that's my reason for the ramp.
Thanks to JDB for the detailed answer, I'm beginning to think about the ramp in a different way now.

Also my question: Do you use phantom blocking caps for vintage preamps, such as the V276a and are the applied properly in my schematic ?



Thanks
Ray
 
RayBlack said:
Now the strange thing is that I have constant 48V on my mic Input...why is that so ?

If your circuit matches your schematic, then your Phantom on/off relay is wired incorrectly. I would suggest that you remove the 2,7k resistor and connect the common contact of the relay to the base of the TIP41C. In that case you might as well remove the MPSA06 transistors, as the ramping will be implemented through the first pass transistor.

(If your circuit matches your schematic, then I don't quite understand how you see 48V at all. Where are the MPSA06 transistors getting their base current?)

RayBlack said:
The Psus and vintage preamps shall be used with historic ribbon mics such as r44 or r88dx, so that's my reason for the ramp.

Ramped phantom only gives a measure of protection if such a mic is connected to the pre when phantom is on (and even then there is debate whether it's necessary at all). What happens when you plug such a mic into a pre with P48 already enabled? ("Don't Do That Then" -- same goes for applying phantom to those channels in the first place).

JDB.
 

Latest posts

Back
Top