DIY Mic Pre Kits, Lunchbox Mic Preamp Kits, Power Supply Kits

Hi,
I was done assembling SC-1 preamp, PSU-1848, and 1U power transformer into an 2-mm thick aluminum box. The picture is included in this post. I was able to get a decent sound from it using my Shure SM57. Still I am a bit concerned about the noise.

According to the 5th post of this thread, it reads



When I turned both gain and trim knobs fully clockwise with nothing connected to the input, I got the noise at around -28 dBFS, which was quite high. This is actual noise that I captured. Should I be worried about this? (Note that for my vocal, I only need ~18 dB of gain in order to get the vocal level to be around -18 dBFS. The noise floor is around -60 dBFS.)

I also notice something strange. With gain knob fully counter-clockwise (no gain), when I turned the trim knob from its fully clockwise position, the noise started to get louder and become the loudest at the knob position halfway between two extremes. With further knob turning, the noise decreased and faded away when the knob was fully counter-clockwise.

In summary, I am wondering whether there is something wrong with my preamp or not. If so, I ran out of idea to debug it. At some point, is it possible if I send you the preamp to test?

Thank you very much for your help,
Nirand

I was having some weird problems like that. Turned out to be an IC chip. Which could add to the noise problems. Owel suggested in this thread:

http://www.fivefishstudios.com/forum/vi ... .php?t=370

That i should basically just try swapping/taking out each chip one at a time and come to find out, one of the chips was the problem.

By the by... i can't help but notice the umich address... are you a student there too? What are you studying?

Hi guitarguy12387,
Thank you very much for your help.



I am not sure this is the thread since it does not seem to talk about IC chip. Do you mind directing me again to the information?

The other problem (due to my ignorance) is that I soldered IC chips (instead of IC sockets) onto PCB directly. In this case, it might be possible that my IC chips was damaged during the soldering. I guess I need to send it back to Owel to try to get them out using his de-soldering tool.

Again I really appreciate your help,
Nirand

Yes, I am a student at U of M. I study Engineering. Are you a current UM student too?

Nirand,



I listened to the clip. I can hear some hum and RF noise. At extremely high gain, there will be white noise. At max gain, we're talking about 72dB, which is 4000x amplification. Any little noise, will be amplified 4000 times! So you can see how it is very important that construction and wiring be as clean as possible.

Your hum seems loud... but I also see that your power transformer is near the preamp. See some tips below:

Also, how did you record these? From the preamp out, did it go to a Mic input of your audio interface or the line input? Is it a balanced or unbalanced connection? Is there any "gain" being applied by the DAW software or audio interface?

Here's a sample of my 72dB, 150ohm terminated input, well grounded case, 1u rack, PSU and transformer far from the preamps... sound clip.
http://www.fivefishstudios.com/audio/DCDC72dB150R.mp3

Some suggestions based on the photo you sent.

1. Ideally, you want the power transformer and AC lines as far away as possible from the preamp. Otherwise, any slight hum will be amplified... especially at 4000x.

But since your case is small, I suggest tighter and cleaner wiring. For example, I see the DC power supply wires (CABLE4) going to the pre almost touching the power transformer. There's plenty of room to reroute that... move those wires away.

2. Did you ground your case? I mean truly grounded? It looks like a painted case and I wonder if you removed any paint to ground it. Otherwise, the paint will act like an insulator. Did you ground the XLR tab and pin1 on the XLR jack too?

My hunch is your case isn't well grounded...

3. In real life, you won't be running the pre with the inputs hanging. There will be a mic connected to it. Have you listened for noise with the inputs terminated with a 150R resistor? to simulate a mic connected to it? This should result in a lower noise floor compared to the above recorded sample. (Also, take note... your above recorded sample is 72dB gain!)

4. The trim knob should be set at max clockwise... that's the 0db mark, i.e. no trimming done. Only at high gains where you want an "in-between" setting should you turn that trim down a bit.

If you really want the best performance out of the preamp, try a bigger case with the PSU section, as far away as possible from the preamp. Otherwise, hum and noise brought by the AC lines will be to near proximity to your preamp.

5. Have you tried the rotating toroid trick? I.e. loosen the bolt of the transformer, and turn it around slowly and listen for the sweet spot where noise dramatically goes down. Sometimes a few degrees of turning have a big effect on lower noise floor. WARNING: This needs to be done with power ON so you can listen to it... otherwise, to be safe... turn power on/off, adjust toroid, power on/off listen... repeat.


6. I personally don't use shielded wire for the XLR in/outs. Instead I use stranded wire and twist it tight. In my experience, it performs better than using shielded wires for this short run. Experiment if the noise goes down if the wires are right against the case floor, or elevated.

Hope this helps... there is no ONE magic thing to do to ensure a low noise build. You need to be meticulous on every aspect of the build as every little thing can add up to a big noise. Hope the above pointers help you.

_________________

Oh yeah? Thats cool. What kind of engineering? Yeah, i'm an EE.

I missed this part...

>> If so, I ran out of idea to debug it. At some point, is it possible if I send you the preamp to test?

Sure... but first, I'd like you to take a stab at troubleshooting it. You'd learn A LOT more about stuff when you're trying to debug something.

You have to understand NOISE, differentiate the kind of noise, understand how each kind behave or manifests itself... then based on the type of noise, do some reading about it and you'd figure out what step(s) required to defeat it.

_________________

I have a question about how you terminated the input with 150R resistor. My understanding is that you have to put 150R across two pins of XLR input jack. However, I don't know which two pins should I put the resistor across. Is it between pin 1 (ground) and pin 2 (hot) or is it the other combination?

Thank you very much for your help,
Nirand

I believe across pins 2 and 3... check out this thread

http://www.studio-central.com/phpbb/vie ... p&start=25

Yup, pins 2 and 3.

_________________

Hi,
After juggling with SC-1 for almost a week, I think I figured out the problem with the SC-1 being noisy. The problem is that I didn't set up the correct test situation, being the following:

1) I left the input XLR open. (Not plugging anything to it.)
2) I set the input of my interface to -10 dBv.

So when I terminated the input with 150R and reset the input of my EMU 1820 interface to +4dBu (as well as moving wires around, trying different outlets, toroid tricks, moving SC-1 away from the transformer), I got the noise at about -65 DbFS in my DAW (as opposed to -28 DbFS from before) with maximum gain of SC-1. Although this is not as good as -72 DbFS from Owel's test, I am satisfied with this. I did some vocal recordings with SC-1 and the noise level was way below the recording anyway. There is no need to fuss about the minor difference.

So, I guess that there was nothing wrong with the SC-1 in the first place, the problem is my inexperience.

Also, I would like to thank Owel for taking his time to inspect my SC-1. He didn't have to do this and I appreciate it. I couldn't ask for more for the price I paid in terms of product quality and customer service.

Sincerely,
Nirand

Hi everybody. I am experiencing te same type of problem.
I have 6 SC! mk1 mic-pre:s
ALl of the have this problem???

I also have two SC1Mk2, And they dont behave the nois when turnig up the trim. I also have 2 X12:s and no problems with hiss-noise whe turning up the trim-pot.

I can see in the scematics that SC1 mk1 dont have the OP-amp buffers between the input Chip and the trim-pot, and between the THAT1646 (output amp) as the MK2 has.
This hiss is regardless of the level of gain in the first step. and is the same whether or not I have any phantompower turned on. or if I bypass C18 cap. I dont use the instert connector, so I cannot test to insert anything there.

There was some discussion about a faulty THATchip somewhere, but I have similar bahaiviour on all my SC1:s

There are 2 controls on the original SC-1. The Gain knob and the Trim knob.
The trim knob basically attenuates the signal.

In normal use, the TRIM knob should be set to fully clockwise setting (0dB trim), i.e. no attenuation. You adjust the preamp gain using the GAIN knobs.

But at the higher gain settings, 48dB+ and higher, it may be hard to find a middle spot where you want more amplification but not too much. (at the higher gain setting, you have the equivalent of 500x, 1000x, 2000x times amplification) This is when you adjust the Trim knob and dial it back down a little.

The mk2 is the 2nd generation and improved design of the SC-1. I found out that some people use the original SC-1 by setting the gain real high and then adjusting the trim pot down (instead of just leaving the trim fully clockwise).

The original SC-1 have jumper headers, and connections/pads for an INSERT. One can remove the jumper here and insert a buffer, located on an external PCB). This could be a DIY project as I don't have kits for these.

_________________

So there should be a buffer between the trim-pot and the Output Chip?
Does the noise go away when, you insert something there, lets say a compressor in the insert-point, or a simple buffer, before the THAT1646?

Or do you think I should replace the Trim-pot with a 100K resistor, to avoid the option to set the pot midways?

For the SC1:s In this particular case, where I dont have the insert-jack soldered, I have a ballanced compressor after the SC1, and set the SC1 trimpot to max.
My following compressor has a threshhold control, and an output gain control, which makes it easy to trim the signal a bit.
Even if you set the SC1 gain for 30-40 or even 50 dB gain.

In the case of the SC1mk2 and the X12:s, I have solderred the insert break jack (drilled on the frontpanel) , and using an (unballanced) RNC inserted into the insert-jack. then the micpre:s trim works as an output control sending the signal to the recorder.

On the first generation SC-1, if you're going to leave the Trim knob at fully clockwise setting most of the times, you don't need a buffer. But you can use a simple buffer at the insert point. The second generation SC-1mk2 has this.

The trim knob attenuates the signal before the 1646, so normally you want that setting at zero attenuation (fully clockwise).

_________________