Korg Forums

[panrixx]

I have three keyboards and two rack mounted sound modules. The sound modules are a Roland JV-2080 and a Yamaha Motif-Rack XS. I also have two patchbays - a P&R PB48 and a DACS MIDI patchbay.

I am successfully using the DACS MIDI patch bay to easily patch any of the three keyboards to either of the two sound modules. So far, so good.

I have configured the P&R PB48 so that the first 12 pairs are de-normalled and the second 12 pairs are (as factory set) half-normalled. So, the first 12 pairs (24 individual connections) are straight through connections with no link to the upper or lower.

Rather than plugging my headphones into the relevant sound module being used each time, I thought I could link each headphone out (from the sound modules) to a seperate 'de-normalled' input at the rear of the patchbay. I would then simply connect my headphones to the relevant output at the front of the patchbay.

Although this sort of works the quality of the sound is adversly affected. The volume is reduced and the clarity is degraded.

I know the headphones cable is un-balanced stereo. I have tried the link between module and patchbay with both un-balanced cables (TS) and balanced cables (TRS). The headphones I have tried are Sennheiser HD 515 (50 ohm impedance) and Audio-Technica ATH-911 (600 ohm impedance). Obviously, the latter give even lower volume for the same set point.

Is it even possible to do what I am attempting or am I missing a basic mis-matching of headphones and patchbays?

Thanks in advance for any advise on this topic.


EDIT:

I just connected a TRS cable at the front and back, allowing me to test for continuity with a multi-meter. Both Tip and Sleeve have continuity but not the Ring. So, the patchbay is an un-balanced (TS) version and presumably why I am getting a degraded sound when the un-balanced stereo lead of the headphones is connected.
_________________
Rhodes 73 Mk1; Kronos 88; Synthex; Electribe EMX-1; Motif XS Rack; Yamaha DX7IIFD; Yamaha MM8; Roland Alpha Juno-2; Roland JV-2080; Tascam FW-1884.

[xmlguy]

That's completely the wrong way to attempt to monitor your gear with headphones. Headphone outputs are amplified audio, not line level outputs. Your patch bay is for line level devices only, not for patching amplified audio. You could even damage the patch bay, if you haven't already done so, by patching amplified audio through it. The power could easily burn up the internal connections or passive circuits.

Line levels are in milliwatts. Headphone outputs can be in the watts range. The current for headphone output can be much higher than line level, and excessive current can heat and burn up conductors and components not made to handle the load.

To monitor your gear with headphones, there's a device already made to handle this task: a mixer. You plug your headphones into the mixer, and it can hook the input of its headphone amplifier to monitor whatever it allows. Mixers let you monitor various channels or busses, before the fader (prefader listen PFL) or afterwards (post fader-main buss). More advanced mixers give you detailed control over what you can monitor. There are even mixers designed specifically for monitoring, which have a much larger number of output busses to give musicians their own individual monitor mix, for example.

Patch bays are for patching line level components together, not for mixing or monitoring the components. So you are using the wrong tool for the job.

[panrixx]

OK, thanks for the clear explanation. I'll make sure I use the patchbay for line level ONLY in the future.

I have only tried the headphones on one bank (top and bottom) so the maximum damage is 42 way, rather than 48 way.

However, I assume that as my post test with the multi-meter gave continuity, front to back, on both the Tip and the Sleeve things may be un-damaged. Is that logic correct?
_________________
Rhodes 73 Mk1; Kronos 88; Synthex; Electribe EMX-1; Motif XS Rack; Yamaha DX7IIFD; Yamaha MM8; Roland Alpha Juno-2; Roland JV-2080; Tascam FW-1884.

[xmlguy]

Instead of measuring continuity, measure resistance. If the resistance is low, then it's undamaged. If it's high, then something probably got cooked. It would probably be fixable though by tracing down any point of high resistance to replace whatever connection got overheated.

A small diameter wire will exhibit low resistance to current flow within its current handling capability, but will show higher resistance/impedance to current above that point. A wire can act like a high pass filter, until the point that the current fries it, at which point it becomes an air insulator when the smoke is released after the conductor get vaporized. One of the ways you can tell that a conductor is undersized for the audio power is that the highs will sound fine while the lows distort/disappear, until you smell the putrid smoke of electronic death. This is what happens when someone mistakenly uses a mic/line cable to hook a power amp to a speaker instead of a large gauge speaker cable. This happened rather frequently when we used XLR and 1/4 TS connectors on amp main outputs and passive speaker inputs. That is one reason why Speakon connectors were developed, to make them physically incompatible with audio level cables so that this confusion is eliminated.

[panrixx]

Makes a lot of sense but being a mech engineer rather than an elec engineer could you give me an idea of what would be classed as low resistance (un-damaged) and what would be classed as high resistance (cooked)?

You are expanding my knowledge with every post - thanks again.

EDIT:

Just tested the impedances, using a 3m long ultra-low impedance copper core mono jack lead in the back of the patchbay and a very short patch cable in the front.

Both top and bottom connections read more or less the same:

Sleeve to Sleeve = 0.5 Ohms
Tip to Tip = 0.9 - 1.0 Ohms

Does that seem OK or not?
_________________
Rhodes 73 Mk1; Kronos 88; Synthex; Electribe EMX-1; Motif XS Rack; Yamaha DX7IIFD; Yamaha MM8; Roland Alpha Juno-2; Roland JV-2080; Tascam FW-1884.

[xmlguy]

Glad to help. Just measure the resistance of some of the other connections that you didn't use that way, as a point of reference. If the resistance is consistent, then you're OK. If the resistance is much higher, it was cooked.

1 ohm seems rather high, but if it's consistent, then I wouldn't worry about it. If the other patch connections are in the milliohms, then the 1ohm ones are cooked. Be sure to use the milli or micro ohm range on the multimeter, and ensure a solid contact of the probe/clips to the testing surface. Oh, I just read that you used a 3m cable. That's too long for this kind of test. I would make some 1/4 TRS testing plugs, with 3 inches of speaker wire per conductor, then attach the multimeter with alligater clips to the speaker wire. You want to test the internal resistance of the patch bay, not the overall resistance of the patch bay plus long cables. Short patch cables would work ok too, if you use alligater clips to get a good contact with the probe. The total resistance will be the sum of the total circuit length plus the contact resistances. To calculate the internal resistance, you would need to subtract out the resistances of the external cables, and if those are disproportionately longer, then the precision of the internal resistance will be degraded.

The internal conductor length inside the patch bay is probably only a few inches, so you want to test with Short cable -> patch bay -> short cable to be able to keep the patch bay as a large proportion of the overall circuit length. Long cable -> internal -> long cable will be mostly testing the resistance of the cables, not the internal resistance of the patch bay.

You get more precision weighing 3 mice, then subtracting out 2 mice, rather than 2 elephants and 1 mouse, subtracting out 2 elephants, if you see the analogy.

[panrixx]

I will need to make up those shorter test leads and clips but, in the mean time, I used the same configuration on two other (previously unused) connectors and the readings were consistant.

Take my first reading of 0.9-1.0 Ohms for the Tip to Tip, via the patchbay.

I measured the leads and got the following readings (Fuke 112 multimeter):

Short patch lead (200mm) = 0.2 Ohms
3m low impedance, mono, jack lead = 0.5 Ohms

So circuit resistance , say, 1.0 Ohms less leads resistance = 0.3 Ohms. Good contact was made with plenty of pressure applied to get an accurate reading.

My worry is that when measuring the Sleeve to Sleeve impedance I got around 0.5-0.6 Ohms, so the maths comes out negative resistance !!!
_________________
Rhodes 73 Mk1; Kronos 88; Synthex; Electribe EMX-1; Motif XS Rack; Yamaha DX7IIFD; Yamaha MM8; Roland Alpha Juno-2; Roland JV-2080; Tascam FW-1884.

[xmlguy]

Your multimeter doesn't have the precision to measure short wire lengths. It is accurate only to 0.1 ohm, while the resistance of a 3" length of 24 gauge stranded copper conductor is about .006 ohms. But this doesn't really matter. If you measure the resistance to be about the same, then they are ok. The problem would be if the resistance was many times higher, like 100 ohms when the rest were .5-1 ohm. So i think you're good to go.