24 channel line attenuator ?

[2Low]

I´m looking for a solution to bring down the outputs of my MCI - JH16 Tapemachine
by 3 or 6 db . The output levels are to hot to transfer the tracks to Pro Tools


Of course it´s possible to lower the outputs of the machine ( sync , repro )
but i would prefer to build an attenuator box

I checked the Uneeda website, but im not sure whats the best pad for the XLR ( + 4 dBu ) outputs of the MCI ...

Thanks !

 

[Holger]

Can you describe what levels come out of the tape machine that can cause trouble in a modern A/D converter?

 

[Tubetec]

What you could try is an unbalanced connection at the A/D end , if you keep the distance short it should work fine and give -6db level compared to balanced . Inline xlr attenuators are available
https://www.parts-express.com/Inline-XLR-Attenuator-Pad-10dB-240-412#lblReviews but at the end of the day you could easily make these up yourself with custom values . The XLR male to female barrels can be found on ebay for around $5 each .

Heres a better calculator which takes into account input as well as output impedence ,
http://www.nu9n.com/tpad-calculator.html
Im not sure if T or H pad is the most appropriate in this case , maybe someone else could chime in about the differences

 

[2Low]

Can you describe what levels come out of the tape machine that can cause trouble in a modern A/D converter?

F.x.hot recorded drum tracks above 0 VU on the machine will clip the AD of the Pro Tools interfaces

 

[Bo Deadly]

Unfortunately it is not possible to make a passive attenuator that attenuates less than 20dB or so without the series resistors becoming quite large which will potentially increase noise and distortion (albeit probably insignificant compared to a tape machine).

However, note that the noise floor of the tape machine is going to be at least 30dB above the noise floor of a vaguely good converter. So you could make your attenuator 20dB and then apply gain digitally after with no impact on performance. In fact, it would perform better than an attenuator with larger series resistance.

Also, if you make a passive attenuator, you should put it right next to the converter so that the line is very low impedance and less susceptible to noise pickup. So a box with some DB25 ports. Or, have a bunch of cables coming out of a cable gland. For one-offs, that can be a little cleaner.

 

[nielsk]

Change the reference level setting in Pro Tools?

 

[2Low]

Change the reference level setting in Pro Tools?

I´m using three 96 I/Os, so the input level has only two settings ( +4 dBu / -10 dBv )

 

[Bobby Baird]

Found some cool little db-25 boxes to make an 8ch pad with fixed resistors.

 

[Bo Deadly]

I´m using three 96 I/Os, so the input level has only two settings ( +4 dBu / -10 dBv )

Both dBu or dBV?

Compare the noise floor and THD of a 0dBu reference tone of a) what you have now with no pad and the input level at presumably +4dBu and b) a simple -20dB L pad with 10K:600 characteristic impedance and with the input level at -10dBV.

I think you might find that the -10dBV setting could be marginally better because internally converters can't handle high signal levels and so the +4dBu setting is actually just engaging a pad. So just use -10dBV and use your own pad specific to your application.

For 24 channels you might also make a little PCB with edge connector fingers that line up with a DB25 and a metal shell big enough for the PCB with the pad resistors. Those shells can be pretty big. Might be big enough for through hole but SMD on both sides would be best IMO. That saves you a whole lot of wiring and minimizes mistakes.

Make one 8 channel XLR -> DB25 snake and use that just for drums to see how it goes.

 

[2Low]

Both dBu or dBV?

Compare the noise floor and THD of a 0dBu reference tone of a) what you have now with no pad and the input level at presumably +4dBu and b) a -20dB H pad with 10K:600 characteristic impedance and with the input level at -10dBV.

I think you might find that the -10dBV setting could be marginally better because internally converters can't handle high signal levels and so the +4dBu setting is actually just engaging a pad. So just use -10dBV and use your own pad specific to your application.

For 24 channels you might also make a little PCB with edge connector fingers that line up with a DB25 and a metal shell big enough for the PCB with the pad resistors. Those shells can be pretty big. Might be big enough for through hole but SMD on both sides would be best IMO. That saves you a whole lot of wiring and minimizes mistakes.

Make one 8 channel XLR -> DB25 snake and use that just for drums to see how it goes.

Thanks

 

[hitchhiker]

For my 3M M56 I just turn down the machine output level by 6dB. That is usually checked and adjusted each time you use your MRL tape.

 

[abbey road d enfer]

Since the input impedance of the 69I/O is high enough (>10k), and the output impedance of the MCI machine low enough (,<100 r), a good attenuator can be designed that will maintain good performance.
An L-attenuator is sufficient. For 6dB I would suggest 1k in series and 1.1k shunt. For 3dB, 560r series and 1.6k shunt.
That would result in 96I/O seeing a source impedance of about 500r and the MCI seeing a load of about 2k, which is enough to linearize the transformer's response and let the output stage breathe.
Now if you want to do an H-attenuator, you must have 510r in series in each leg and 1.1k shunt for 6dB, or 270r is series in each leg and 1.6k shunt.
The choice is yours, but I don't think there is a significant difference in performance.

 

[mik]

I´m using three 96 I/Os, so the input level has only two settings ( +4 dBu / -10 dBv )

Pusch -10dbv you shoud see -14 on your DAW wen the Meter of the recorder hit 0Vu

 

[Bo Deadly]

Pusch -10dbv you shoud see -14 on your DAW wen the Meter of the recorder hit 0Vu

I don't think that's how that setting works. If you set -10dBV that means that the signal level coming in is -10dBV so trying to feed +4dBu into that will just overload the input even more (by ~+12dB).

 

[mik]

I don't think that's how that setting works. If you set -10dBV that means that the signal level coming in is -10dBV so trying to feed +4dBu into that will just overload the input even more (by ~+12dB).

No, because Who Is wrote the manuale Is not Who has build the device, so Is the exatly the contrary. If you pusch the -10dbv Button the device doesn't add any Gian as a metter of fact. Try to yourself. I know that device too well.
M.

 

[abbey road d enfer]

No, because Who Is wrote the manuale Is not Who has build the device, so Is the exatly the contrary. If you pusch the -10dbv Button the device doesn't add any Gian as a metter of fact.

When you engage the -10 setting, the input gain increases and the output level of the interface decrease.
That means that when it receives a nominal +4 from the tape machine, the input of the interface will be overloaded, by ca. 12 dB, because there is a 2.2dB difference between dBu and dBv.

 

[Bo Deadly]

No, because Who Is wrote the manuale Is not Who has build the device, so Is the exatly the contrary. If you pusch the -10dbv Button the device doesn't add any Gian as a metter of fact. Try to yourself. I know that device too well.

Well I don't know that device at all but based on general experience with devices that have that feature, it would be very strange if it was backwards like you say. I would think that engaging -10dBV would not add gain but would attenuate less which is the same thing. The end result would be that the level hitting the ADC would be 12dB hotter which is non buono for a +4dBu signal.

 

[2Low]

No, because Who Is wrote the manuale Is not Who has build the device, so Is the exatly the contrary. If you pusch the -10dbv Button the device doesn't add any Gian as a metter of fact. Try to yourself. I know that device too well.

My 96 I/Os are working like they were supposed to
Switching to -10 dBv increases the levels in Pro Tools ...

 

[abbey road d enfer]

add two resistors 680 ohm on each +/- of the output (ideally before the audio output transformer if the machine has output transformers.)

What would it do? About 1.5dB attenuation ... That is not enough

 

[abbey road d enfer]

My 96 I/Os are working like they were supposed to
Switching to -10 dBv increases the levels in Pro Tools ...

That's correct. Exactly what Bo and I said.
You said the contrary in post #13: "Pusch -10dbv you shoud see -14 on your DAW wen the Meter of the recorder hit 0Vu"
And it's actually 11.8 dB