during record, signal will not exceed -4dBFS

[oldmanpotter]

Just started a job at a new studio. Using an 888/24 and PT v.5.1.3, when recording a signal using any analog input, the signal is being clipped (or limited) prematurely, not coming anywhere close to 0 dBFS (-4dBFS was an approximation). The same thing happens no matter which input I select, and even happens on a second I/O, which leads me to believe there's a software setting somewhere (designed to prevent digital clipping?) and the threshold is way low. But where oh where do I adjust this setting? Or is there something else going on here? The fader on the record track is at unity gain, and I have the simplest signal path possible, with no plug-ins. When using the digital inputs, the problem disappears. I've tried using both 16 and 24 bit sessions with the same results. The answerbase mentions a problem with 888/24's connected to a Farm Card, but I'm positive this I/O is connected to the Core.

Peter

[Richard Fairbanks]

It sounds like the analog console feeding into your 888 is clipping. Either the console outputs are not designed to pump out a full +24dbV, or the input gain setting of the 888 is too low. Take a look at the spec sheets for the analog device feeding into your 888.

[oldmanpotter]

Thanks for the quick reply. The clipping definitely seems to be occurring at the input to the 888/24. I discovered the problem after trasferring some stems directly from a DA98. Just to be sure it was not somehow being caused by the source machine, I took a 1kHz sine wave from a signal generator in ProTools and routed it back into the I/O. I even bypassed the patchbay, connecting the output directly to the input on the back of the interface. Same result, same clipping.

When you mention the input gain being set too low, do you mean the +4/-10dB setting? This occurred to me, but if it was in fact inadvertently set to -10, I assumed the output level from the 888 would be low as well. But the output is routed to a Multimax, and the Dorroughs attached to the MM meter outputs show I have plenty of signal strength at that end. Therefore, I have not checked the operating level status. (Didn't the older 888's have a switch on the back to change operating level, or did I dream that?)

btw, congrats on the success of your Syncheck! What a shame we'll no longer be able to blame loose sync on our monitors!

Peter

[Richard Fairbanks]

There is a trim adjustment for each input and each output, those which are on the 888/24's front panel. You mention that digital signals into the unit do not have the same problem, which definitely points to analog A to D converters not receiving signals hot enough to make 0dbfs. The 888's input trims and/or +4/-10 jumpers are not set to match the analog signals feeding into your 888 inputs. The +4/-10 adjustment you mention is done with jumpers inside the 888's case. Which are you set for? The jumpers get you into the right ballpark, (are you interfacing to "pro" gear with +4 signal levels, or are you interfacing to "semipro" gear with -10 signal levels) and the trims allow you to make considerable variations to fine-tune both input sensitivity and output level of the 888. I think you should get your hands on a meter that can read RMS AC voltages in order to verify the trouble. For most "pro" setups, reference level (-20dbfs in the USA) sinetome coming from the signal generator plugin directly to a single output should read 1.23VRMS (AC). Rather than repeat setup procedure here, you should be able to find a good guide in the Pro Tools manuals. I can't recall if it is the thick manual or an 888 setup manual. Good luck.

[oldmanpotter]

Problem solved...the I/O's were set to -10 on input, +4 for their outputs. I honestly do know how to calibrate a recording device. (I came of age in the days of 2" 24 tracks. Oy!) But I was fooled by thinking that the inputs and outputs could not be independently switched. As you're probably aware, not only can they be independently switched, individual channels can be either -10 or +4. Thanks for your help Richard.

Peter

[Richard Fairbanks]

Judging by the hour of your post, you've had a good end to a very long day! You're welcomed.