The AKG D112 had been bought through eBay as it has a very similar frequency response curve to the Neumann U47. The idea was to connect it ie its moving coil element to a step up transformer and then to the grid of an electron tube powered by battery, see Pleiades V6. Underheated electron emitting cathode like the VF14 electron tube in Neumann U47.
Of course proximity effect compensation, Fletcher Munson curves compensation and voice effort effect compensation is needed to reduce the bass heaviness and archive a flat frequency response from singer's vocal chords to listener's brain [Lowe, Morgan].
This post focuses on D112 and wave filters only. Mic is connected to passive wave filter. The output of the wave filter is then connected to a standard preamp. So no mic booster amp such as Pleides V6 is used on these experiments.
Signal path:
Male singing voice at 1-5 in - AKG D112 - passive filter - Sony TC-D5 Pro at XLR mic in - Sennehsiser HD580
A phase inverting XLR to XLR is also used before Sony as for some reason the particular one used phase inverts.
1 The D112 is connected directly to Sony TC-D5 Pro. Very bass heavy sound.
2 The Cinema Eng passive EQ is connected between the two. At -10 for Low Frequency (100Hz?) setting the sound is remarkably nice. Boosting a bit at 3KHz makes presence even better. Unfortunately there is hiss as the EQ has insertion loss (14dB?) [Tremaine]. It should be connected after a mic booster front end amp or a preamp.
3 The Pleiades (130Ω,40mH) gentle slope low cut passive wave filter is connected between mic and preamp. Just a 130Ω resistor in series with 40mH. The resultant connected in parallel ie to pins 2.3 inside a Neutrik Module XLR to XLR barrel. The sound is very nice and big. There is almost no hiss as the Pleiads filter just subtracts what is not wanted. Sound could be further improved by optimizing the R, L values. A variable box should ideally be built. Series C in parallel with variable R should be tried too as an alternative to gentle slope low cut.
Later today's addition. By connecting L=85mH and R(330R log pot) with cable clips a combination of 85mH is series with 100Ω sounded very nice and natural with D112 at 3-4in. The 85mH inductor is made by winding a small grey colored Siemens Epcos (Siemens Matshushita) ferrite ring core which fits inside an XLR female to male barrel adapter. The core type should be R 12.5.
Singing together with Yamaha PSS-14 synth strings chord backing, the Pleiades (130Ω,40mH) sounds more edgy, with more bandwidth on both sides of spectrum, with less mid range giving a sense of a smoking singer, while the Pleiades (100Ω,85mH) sounds more clear with more focused midrange while still having treble. Bass frequency spectrum content of voice is natural in both cases. The latter filter has more focused presence. Pehaps with the backing harmony the first example is more suitable as the edginess makes the voice stand out. Another interesting R value is 73Ω. More experiments are needed, for example varying R,L parameters while singing to a backing track. See also Pleiades R,C filter on tomorrow's post.
references:
Sound picture recording and reproducing characteristics - D. P. Lowe, K. F. Morgan - JSMPE
Audio Cyclopedia - Howard M. TremaIne
No comments:
Post a Comment