Friday, October 14, 2011
Modding my Musical Fidelity V-LPS
I listen to music in the living area and prefer my hifi gears to be slim, compact and minimal in box-count. As I own an integrated amplifier, my idea of the ideal phono stage would be an inbuilt module (one component and a pair of interconnects lesser) but alas it is not to be. Enter the Musical Fidelity V-LPS phono stage that I bought 2 ½ years ago and in use since.
The circuit of V-LPS (http://www.vinylengine.com/turntable_forum/viewtopic.php?t=36631) is supposedly identical to those of its bigger and dearer siblings in kW and various versions of X-series phono save for some updates in later incarnations and power supply circuitries. It draws 12V DC unregulated input voltage from either a wall wart or transformer and regulates the voltage down to 9V DC single supply with a LM7809 regulator on the phono board and into a JRC5532DD op amp for rail splitting whereas the X-series phono stages require 12V AC input voltage regulated up to 15V DC for each positive and negative dual supply using other LM7815 and LM7915 regulators. The later version of X-series phono should technically sound better than V-LPS as op amps are operating at optimum voltage but at nearly 3 times the price of latter (during launch), it is clearly not value for money.
The sonic quality of V-LPS in stock form is pretty good, one that I really like and could easily live with had I not have heard what it could be transformed into by using slightly better parts. My version (PCB 2300 issue 02) has the 8-pins SSM2210 dual-matched NPN transistor at IC202/IC302 in first stage and purportedly operating in class A mode with signals passing through a push-pull circuitry before the main pre-amplification and op amp equalization (input of MM signals) at second stage. Residue DC voltages along the signal path are being coupled with electrolytic capacitors, a choice probably to keep costs in check. Some measurements noted from an online review: Noise Ratio (MM @ 5mV, 1kOhm): 84dB / (MM standard system): 81dB / (MC @ 0.5 mV, 20Ohms): 74dB ; Gain (MM): 42dB / (MC): 60dB ; Headroom (MM): 21 / (MC): 3.3mV ; Input impedance (MM): 50k+74pF / (MC): 99Ohms ; Output resistance: 440Ohms.
The intent to mod my V-LPS has been firmly planted in me since day one of its acquisition but I procrastinated until recently when a friend pestered me to replace just the JRC5534D op amps in the RIAA equalisation section with the highly-rated OPA627BP, claimed to be 'objectively one of the best op amp for Audio, especially in EQ stages, and among those which would subjectively deliver some of the best sound quality' and leave the JRC5532DD op amp in the power supply section intact. I later recapped the entire Jamicon polarized electrolytics with low equivalent series resistance (ESR), high ripples and longer lives Panasonic FM/FC, a choice largely influenced by their availability and cost factor.
I have the 220pF and 22pF capacitors at C207/C307 and C209/C209 respectively removed. The 220pF blue-coloured film capacitor, deduced to be polypropylene is a filter (suppressor) capacitor fitted between the positive and negative inputs of op amp to remove any hum and/or HF hash of JRC5534D but tend to distort if this op amp is replaced with other low noise FET type such as OPA627BP as in my case. An online forumer has suggested a 1 meg resistor be inserted between pin 3 and pin 4 of SSM2210 dual-matched NPN transistor in the event of overload with MC stage, obviously he did not remove these capacitors. The 22pF ceramic capacitor is again not required as unlike JRC5534D, pin 8 of OPA627BP op amp is not connected to anything within.
I ordered some Wima polypropylene capacitors together with the Panasonic FC/FM electrolytics but have yet to use them to replace the ceramic RFI filter capacitors at C206/C306 and C222/C322 for fact that ceramic capacitors especially the NPO/C0G ones possess excellent low impedance up to the MHz range and are technically better suited for bypassing duty. It would be entirely different had the ceramic capacitors been found in signal path instead. I will eventually try out these Wima capacitors although not expecting a marked audible improvement in sonics.
The modifications see a marked increase in resolution, definition, soundstage size and PRaT, essentially better in all areas but its overall sonic character remain exactly same. The only downside I observed thus far being rumbles from needle drops are more pronounced than before, a sign that some running-in are again required. I have yet to replace the non-polar output capacitors, a critical component influencing how V-LPS sounds and undecided whether to opt for an unregulated transformer-based power supply or otherwise but at this juncture (midway through the mods), I wish to thank my friends for all their help and the sharing of many technical pointers especially CM for the soldering job and tracing the root cause of my MC overload problem.
I shall update more after completing the remaining mods.