Hello

I am trying to record from tape decks into notebook computer (t40) using ECHO INDIGO IO sound card. The signal is too low (-30db max). Recording works very fine with other cards like M-audio (but I really do not like USB interface for getting so easy interrupted).

Can someone please let me know whether is there some simple preamp that can be used to amplify the signal from tape deck to the level acceptable by Echo Indigo Soundcard?

Can I use preamp like Audio Buddy from M-Audio which only has guitar and microphone amps?
I cant find any other preamps being sold except Mic, Guitar and Turntable preamps.

Thank you

Petr Yao Mares

Hi there,

Are you using a line out or a head phone socket. Line outs have a lower ohmn
value and are able to give you a much stronger and stable signal. If there are
only a headphone socket try turning up the gain untill it get usable signal.

Regards
Aaron

Thank You

The problem is that there is only one completely unamped line-in in the echo audio pcmcia soundcard. And I need to record from unamped line-out in tape deck.
Recording from heaphone socket of tape deck give signal strong enough but I there is too much distortion and some equalization from that headphone amp. I would very much like to avoid using the headphone socket of tape dack and go directly from line out of tape deck into line-in of the sound card putting some hi-fi preamp in between.


I think the only solution seems to be to buy some kind of hifi preamp and put it in between. But I am not sure mic preamp or guitar preamp will do. Can anyone advice on some solution?

Thanks

Petr

Hi yao;

I don't recommend a mic preamp or guitar preamp... Why??
Because the mic or guitar preamp is for very very low level signals.. LINE OUT tape deck signal is a lot more strong than mic or guitar signals.. as consequence the tape deck signal applied to one mic or guitar preamp causes a lot of distortion...
I believe, what the solution for your problem is a normal signal preamp, to boost for example, +5 or +10 dB.

Thanks a lot for the advice, but searching through the net I found nobody selling anything like normal signal 10db preamp - or what is being offered is mic, guitar or turntable preamps. Do I have to build one myself? Can you reccomend on where to get it? Or how to build it?

I am curious nobody have ever come to this problem.. I did not find it mentionet anywhere on net although I tried really lot.

Thanks again

Petr

-Hifi preamps won't help you here, because they usually don't amplify the signal. The preamp of Hifi amplifiers or receivers (combination with tuner) only switches the signal from the right source to their outputs or to the inputs of the main amp. Turntable preamps have an extra amplifier, because a turntable's signal is so low; but they also apply some EQ-processing. Thus it is no recommended solution.
Don't use headphone outputs. The plug is different for a reason.
I think the best thing you can do is
-either leaving the signal that quiet (if signal-to-noise difference is no problem, most likely the cassetes' noise is already much louder than the additional noise caused by the transferring process)
-or leading the signal through an additional hifi device of higher (or the same, at least) quality like the tape. I mean a second tape here, DAT or CDR would be ideal. Put them into record pause, then you can adjust their record level (=output level) to a higher value.

Here's a thread with very similar topic:
http://www.hydrogenaudio.org/forums/index....=ST&f=1&t=29658

Hi yao, in this moment I am very busy, tonight I will be here again with one design of preamp for you

If you have a 24bits ADC (ECHO INDIGO IO should have it), why not boost it digitally? With 30db of headroom, you still have 19 bits of resolution; which is quite good, especially if it ends in 16 bits.
With an analogue amp, adding noise and distortion, you will decrease the dynamic. It shouldn't be the case in the digital domain (well in a more controlled way).

Dear friends

Thank you for the helpfull sugestions, It seems I have just realized what the problem is:
It seems there is different operating level for the "profi" and "non-profi" gear.
"Profi" equipment runs at +4dB while "non profi" runs at -10dB voltage.

In the "Intruduction to Sound Recording from Geoff Martin"
(http://www.webtopus.co.il/jonathan/textbook.rar)
I have found:

"If you take a piece of consumer gear outputting what it thinks is 0 dB VU
(0 dB on the VU meter), and you plug it into a piece of pro gear, you’ll find
that the level is not -14 dB but -11.79 dB VU... The reason for this is that
the professional level is +4 dBu and the consumer level -10 dBV. Therefore
we have two separate reference voltages for each measurement.
0 dB VU on a piece of pro gear is +4 dBu which in turn translates to an
actual voltage level of 1.228VRMS. In comparison, 0 dB VU on a piece of
consumer gear is -10 dBV, or 0.316VRMS. If we compare these two voltages
in terms of decibels, the result is a difference of 11.79 dB."

I am afraid this is the problem I had run into. As I think I can't change
the input level on Echo Indigo from +4 into -10db, neither the line-out
of tape deck into +4db the only solution seems to be some kind of simple
line amp or transformer installed in between the card and the tape deck.

Can someone of you advice on where to buy it or how to build it?

I am really curious that there are almost no people running into problem of connecting some "consumer" gear with "profi" for some "transforming" gadget to be widely available?


Should such home-build device bring less distortion than recording in 24bits maximizing, requantizing and dithering through L2 or similiar IDR and going back to 16bits as suggested by Precisionist and Marcan?

If the digital gaining in 24 bits is better solution then using home-build preamp; should that be
applied immediately after the recording?
Or should it be the last process I do before finalizing the recording?

Will the Waves L2 be the adequate digital software processor to use?

Thank you all for your help

Yao

Uh, I don't understand all those levels... dBV, dBu, dB UV...

I didn't suggest higher resolution than 16bit, Marcan did it...Anyway, recording and working (editing) in higher bit resolution is always a good choice. Converting up to high resolution should be the very first step while converting down to lower resolution should be the very last step. (Which doesn't mean that the higher resolution neccessarily causes an audible difference.) The question is if the soundcard & software actually record analog->24bit, or if only the first 16 bits are used.

For the digital boost, a simple gain or normalising should make the trick. L2 is a limiter, so it's not adequate here unless you want to reduce the dynamic. IDR or another dithering could be used if you want to end in 16 bits. Maybe this last step is not necessary, the source coming from a tape, it has already a natural dithering and the digital dithering might just take care of the tape's noise and therefore revealing it.
The best ideae would be to record first. After you can apply a normalisation in order to know the exact gain for all your recording. Finally you can downsample it in 16 bits with a good dithering tool (like IDR or Foobar).
With a 24 bits recording, don't hesitate, it's the way to go, it will be better and cheaper. What do you need more? If you want to add noise and to spend money, go for an analogue gear…

marcan,
the soundcard probably has a fixed noise level caused preferably by the computer's other equipment-no matter what the input level is. By increasing the input level, while taking care to not distort/clip the signal, the SNR can be maximized.

Your right, it might be a problem whether the SNR of the soundcard is less than 10 db better than the preamp. The level of the noise should be checked with a sample of the soundcard without any entry. The sample should be boosted digitally with the right level. This noise should be compared with the noise of the preamp with the same amount of gain. According to my experience with a laptop (if it’s an Indigo) you can drastically increase the SNR if you unplug the charger.
Anyway, a good idea could be to make a sample of the noise from all the chain (tape+conversion) and retrieve it with Waves X-Noise.

I am sorry for my poor English and rather complicately worded questions on what is probably obvious.

As the echo Indigo has a true 24bit ADC and as the resulting file I am aiming at is 16bit my question now comes to:

Will the decrease in the quantization levels caused by low line-in signal (-11.79dB) be completely compensated by recording in 24bits and requantizing to 16bit (as compared to recording in 16bits and reaching zero dB VU)??

What I feel is
(24 bits * 6 dB per bit) – 3 dB = 141 dB
(16 bits * 6 dB per bit) – 3 dB = 93 dB
difference is 48dB,
so even while the recording level in 24 bit will be 11.79dB and more bellow zero (or even if being 48dB bellow)- after requantizing into 16bits it will in fact be as if I have recorded in 16bits with level reaching zero dB.
Is that right calculation?
Does that mean that noise of the card itself will not play bigger game due to actual voltage level difference?

Resulting question will be

Should I apply X-noise, x-hum, equalization, c4, and normalize on the 24 bit or on the quantized 16bit file?

Thank you for the patience

Yao

Hi yao , This is the design what i promise...



I believe what this schematic is a probably solution a your problem... but I'm not sure.. It's a audio signal preamp (one channel). The gain is adjustable from 0dB to +20.8 dB. The chip is a NE5532, is a low noise opamp with high slew rate (high response) and high bandwidth (10Mhz unity gain).
The schematic shown only one channel (L or R), duplicate the circuit for to drive both channels..
In the schematic shown one half NE5532; this opamp are two individual opamp in a package..
The frequency response is almost flat, from 20Hz to 20Khz [+0.5db, -0.5db]
Supply power: I recommend any comercial 12volts DC, 1000 mA.

Excuse me my late reply, I was very busy yesterday and today.

Yes

After the gain(normalise), you will increase the noise of your card by 11.79 dB.

As written (did you read me?), Normalise and X-noise might be usefull, the others are not necessary unless you want to change the original sound.

Thank you all for the help.

Thank you for the especiall patience Marcan, I did read you - what I did not mentioned was that the tapes and more often "reels to reels" are some old Classical Carnatic (South Indian) music played by excellent musicians but recorded by amateurs. They definitely need improving- often they are overcompressed and strangely equalized - so I feel I have to apply at least C4 and Q10 to make them sound more close to natural.

I am trying to record them now in Wavelab 5. I do succeed to record them in 24/44 - then I run the record through X-Hum - X-Noise - Q10 - C4 - L2 (type1 dithering, noise shaping, 16bit), all in 24bit chain rendering it into 16bit. I run it through L2 just to apply the dithering which I hear as better, than the internal dithering of wavelab as well as better than the "natural noise of the tape dithering".

I thought that simple requantizing from 24 to 16 will have have effect on the 48db headroom. Can you please tell me why the 48db room does not dissappear in the wavelab when changing the file from 24bit file into 16bit. It still appears after the dithering to have the same -11.79 dB room (so do I have to apply the gain to the 24 bit file before dithering and requantization after the effects?

I am probably missing some very obvious link between requantizing of 24bit-16bit headroom and the gain-normalization. Can you please enlighten me on that?

Thank you

P

When you requantize from 24 to 16, you just loose the less significant bits, so you don’t increase the level but you loose precision. You must apply a gain (or normalising) of around 12 db BEFORE requantizing. If you want to “master” it, personally I would use waves L3 to make all the process (gain, eq, compression, requantize, dithering) except X-noise which you should use before all other treatment. Fore X-noise, you should use a sample of the noise tape.

yao,
24bit or 32bit integer (that means non-float) don't add headroom at the 'top' of the audio - they add footroom at the bottom (the most quiet area) of it.
Don't do any process which is not really neccessary. I think the 'natural' dither, as already mentioned, should be enough. And please: no dynamic range compression ! if it's that what you mean.

Hello

I really appreciate your kind advices. It would be much harder for me to come to the 24bit solution in recording without the hydrogenaudio forum help.

Precisionist, what do you mean by "no dynamic range compression !" - Is that an advice to not to uncompress the overly compressed files? In fact some of the tapes I try to restore have so distorted sound that leaves me completely confused as to what tool to apply. I installed the L3 now but I still have a problem "help" the recordings into more natural sound. I came to all this just because I love this Carnatic and Dhrupad music so much and having access to some old archives I wish to restore and preserve them. My knowledge of digital audio is therefore very limited and stems only from working on this individual project.

Would any of you be so kind and hear few ten to fifteen second samples with some problems I try to "clean" from the recordings and advice me on what tools (pref. of waves plugin) should I use?

I can upload some 10 to 15 sec. pieces on hydrogenaudio website or elsewhere in mono flac, ape or whatever format you prefer or mail them to you.

Thank You

Yao

Sure.
Put it at 24 bits without any treatment. You could also try it with and without the charger of the laptop plugged, so we can check the difference. Let also a blank (of the tape’s noise) at the beginning of the sample.

No. Maybe I misunderstood you; dynamic range compression is considered purely evil by me.
If your audio is dynamically compressed, you'd need to apply dynamics expansion. In practice it's impossible to do this properly. (I tried with some test samples - even if the compression ratio hasn't been infinite, it's too hard to find the correct settings.)
Most of the software you find out there is to limit/compress audio - the opposite of what you want (like L2). I have not yet found a specific "uncompressor", "hard expander" etc.

I have had hard time to find a place online where to letthe few samples available. I have finally succeeded.

I am very glad for you offer to look at them and give me some advice on how to make them more listanable. If anybody else will give me advice I will truly appreciate it.

If you can open this address

http://www.mbizgroup.com/file/ShowFolder.a...e&shareID=31252

And you will find on two pages about 15 very short samples in mono flac 24b in there. Two are just the card with plugged and unplugged notebook. I do not think there is any difference. Definitely not such that will metter compared to the quality of the sound of this old music.

The card itself does not add any sound in my opinion (it is around -100db) sound starts as soon as i plug the tape in. it goes to about -60dB even if it is not actually playing. That is still OK with me.

What I would like you to help me is if you can point me to a particular chain of Waves plugins and their setting that will make the sound more detailed, bright and possibly add the harmonics that had been cut due to poor tape quality in India in 50's. Also very often the music was overamplified while recorded - can I help that in any way?

It will be best if you hear the samples and just give me some opinion as what improvement you suggest.

I normally use chain of X-Noise (with picked sample of tape noise) - X-Hum - C4 (for too compressed or strangely equalized) ending with L2 for not compressed to boost them little.

One file is speech (I also digitalize library for blind people in Czech Republic) having very strong "esss". I tried to use waves "de-esser" without any success. Do you think the "esss" can be eliminated somehow?

Thank you for the kind help

Yao

I need your User name, your Share Password and your Share Name in order to access to those files.

please log on as yao with password yao that way you became admin and can do anything there - I hope