Modifying the Bellari VP-129/VP-130 phono-pre

[Brinkman]

Hello.

I thought I'd share some experiences I've had modifying my Bellari VP-129.

Here are the specs:
I/ O Connectors: RCA, 1/4" TRS Stereo
Input Sensitivity: 30 dB gain @ 1kHz, 47k W Input Z
Input Impedance: 50k Ohms
Output Impedance: 8 Ohms Headphone, 100 Ohms RCA
Max Output Level: >10dB @ 1kHz at at .1% dist.
Equalization: RIAA +/- 1.5 dB, 20Hz to 20kHz
Rumble Filter: 20Hz
THD .02% @ 1KHz
S/N Ratio: >80 dB unweighted
PWR: VP-129: 12 - 16 VAC (PS12 or PS12E)
VP-130: PSU115 15 V DC 700Ma
Size: 6"w x 2.5"h x 5.4"d
Weight: 1 lb
Indicator: 1 Power LED

Here is a link to a photo of the PC board, courtesy of VE member Jazzbeq.

Here is a link to a .pdf of the schematic, courtesy of AVforums.

And here is a link to the datasheet for the 12AX7LPS

PREAMBLE

So there was a bit of murmuring out there about this unit when it first came out, some of it praise, some of it derision and most of it a little uninformed. So a little clarification is needed.

In my estimate, the very complimentary Fremer review in Stereophile stirred the pot up quite a bit. Critics speculated that add revenue *gasp* biased the review and wrote it off without hearing it. And why would they want to? Bellari (owned by the Rolls corp.) is well-known for manufacturing budget-level rack-mount mic preamps, processors, and compressors of questionable merit, so how could they design anything to audiophile standards?

Easy. They didn't. Sort of.

In promotion, the VP-129 is claimed by Rolls to be based off a "tube design from 1957." But what's not mentioned (and this does warrant criticism) is that the design is actually a hybridized iteration of an old circuit. Essentially, all but one of the active devices from the original circuit (valves) are replaced by op amps.

[At this point I'll be referring to the schematic, so it might be helpful to have it open in another window.] The circuit is a single stage RIAA EQ (U1A/U2A) with an bypassable subsonic filter (U1B/U2B), an output buffer for headphones (U3B/U4B) and an output buffer for normal RCA out (U3A/U4A). From the schematic, there's two dedicated op amps per channel rather than four op amps shared by both channels, which means swapping an op amp is done in pairs rather than the singular. In other words, whatever op amp is buffering the RCA out is also buffering the headphone out for a given channel.

So if it's basically your run-of-the-mill transistor-based phono pre-amp, what's the tube for?

Before the schematic was available lots of grumpy people were speculating that Bellari had just done the typical tube-as-output-buffer marketing gimmick that's not all that uncommon. Well, what this design does is a little different than that. V1A/V1B is there to provide input impedance as well as loading capacitance, with R1 and R18 (47 kohm) acting as both cartridge loading resistors and grid leak resistors.

The spec'd 12AX7 is being utilized in what's called a "starved plate" design. Typically the plate of a 12AX7 has around 250-300VDC on it, but in this application it's just 125VDC. Because the plate is being "starved" of it's design-centered voltage, it's subsequent output current is less than you'd see on traditional tube circuitry. However, because op amps operate at much smaller voltages this is of little consequence. Unlike some starved plate designs, the voltage is enough that there's adequate headroom and tube rolling makes an audible difference. Unfortunately, the voltages aren't enough to create that tube glow most have come to expect.

MODIFICATIONS

The first modification can be made from looking at the specified components on the schematic. Specifically, the type of 12AX7 tube. The schematic specifies a 12AX7LPS, the LPS meaning Long Plate Spiral (the plates are longer than the average 12AX7 and thusly have more area, and the filament is wound in a spiral which lowers noise). My stock Bellari came with a Electro-Harmonix 12AX7. After a bit I rolled in a Ei brand gold pin 12AX7 and didn't really like it, so I went back to the E-H. It wasn't until after I had tried and settled on the Sovtek 12AX7LPS that I noticed it had been specified on the schematic all along.

The second modification I made was to replace some resistors. Namely, the cartridge loading resistors (R1/R18). Before I yanked them, I went into the Bellari with my DMM and measured them. The tolerance was within 5% of the specified 47kohm value but I knew I could get closer, so I ordered ten 47kohm PRP brand resistors (1% tolerance) and hand-matched a pair to within .5% of 47kohm. With this same brand, I also replaced the plate load resistors (R3/R20), the output resistors (R17/R33), and some RIAA EQ resistors (R7/R40). Further possibilities would include the VB+ step-down resistors/voltage dividers (R5/R22 & R6/R23), and R39/R21 and R4/R24 for being in the signal/current path.

The whole point of this modification is to match the channels as closely as possible to each other, accomplished by using resistors sold in tight tolerances. PRP resistors are also audiophile brand resistors that claim to impart "less sound", so if there's anything to that, then there's that benefit too.

The other modification I performed was by replacing the coupling caps that go with the 12AX7 (C2/C13). The stock capacitors are bulky ugly yellow ceramic (?) bad boys that one would not usually see in audio applications. I happened to have some extra metallized polypropylene film capacitors of the Sonicap brand. Again, these are audiophile caps, but honestly, any polypro capacitor would be an upgrade from stock. The main problem with these type capacitors is their size - the Sonicaps are about .5" dia x 1" long, so some components had to be bent out of the way. Matching cap values is ideal, but not as critical as matching resistors, so no need to sweat this swap. In fact, the stock caps in this position differed by a factor of ten (!) so it's not hard to improve things here.

There's also a series of large value electrolytic capacitors near (or in) the signal path that could be upgraded. Namely, C36/C37, C11/C22 (the RCA output caps) and/or C9/C20 (the headphone output caps). This is where audiophiles should chime in. I used ELNA Silmic IIs, but others might prefer "audio-quality" Nichicons, high-grade Panasonics, or the expensive Blackgates.

In the power supply there's a bank of seven 47uF capacitors. If one was in the mood for overkill, these could be replaced with low-ESR type caps of the Panasonic brand.

Further modifications that I have considered but not attempted would include swapping out the output buffer op amps with other types to compliment overall system synergy, as well as upgrading the stock Rolls wall wart with a better-regulated quieter model. I also suppose the RCA jacks could be replaced with higher-quality brands, but I couldn't really be bothered as the stock ones are still in fine shape. If I do any of these, I'll post back.

IMPRESSIONS

I was looking for the most open, dynamic, low-noise, balanced and neutral phono-pre amplifier I could get out of the Bellari. By far the largest improvement was had by finding a more suitable tube for it, which is fortunate because it's also the easiest modification! However, the reason I'm writing all this up is because the other procedures were not without benefit: I was rewarded with a quieter background, a less grainy presentation, a slightly improved stereo image and the same tube "warmth" without sluggishness.

Have fun. Thanks for reading. I hope it helps.

[Dave Cawley]

Well, thank you so much for all of that, totally fascinating!

I really can't believe there are two op-amps in circuit all the time, what a waste! And R5 will significantly load the anode resistor R3, it's almost as if they are deliberately introducing extra, extra valve distortion to make it sound sort of musical?

Regards

Dave
_________________
/ Technics SL-1200MKII / Jelco SA-250ST arm / Shelter 501 II cartridge / Neglex arm interconnect / A.N.T. Kora 3T MC phono stage / Marantz SC-7S2 preamp / 2 x MA-9S2 Marantz power amps / Focal Electra 1037Be loudspeakers /

[bauzace50]

To: Dave Cawley
there is a Private Message for you.
Thanks,
bauzace50
_________________
Benign is what you will be after you be eight.

[Brinkman]

EDIT: The following post refers to the VP-129, which is made to be used with a wall adapter that puts out AC. The VP-130 uses a switch mode wall adapter that puts out DC.

I'd been reading about Bellari owners who had upgraded the stock wall wart (12VAC 800ma) with better quality ones in an effort to reduce the noise floor and/or improve the performance of the VP129, so I thought I'd shop around for one myself.

Sadly, I have found there's not much one can do to improve upon the stock wall wart. The problem is that the Bellari requires an AC power source , not a DC power source, and the selection of wall adapters that put out AC is very limited. So as a result, finding an adapter that puts out more VAC than the stock adapter has not been working out.

The 12VAC is needed to charge the power supply capacitor bank that builds the +125V on the plates of the 12AX7LPS. If one were to use 12VDC, the charge would not build up enough and there would be a loss in gain. 15VDC would make a loud humming noise and possibly damage the device.

Additionally, the stock adapter is rated at up to 800mA and the Bellari isn't going to use any more current than it needs, so using an adapter with the same voltage output but a higher current rating wouldn't make a noticeable difference.

According to the datasheet for the op amps in the Bellari, the power supply maximum of the BA4560 is +18V, six volts more than the 12 the schematic specifies. Which means a 15VAC wall wart could be used and possibly make an audible difference, however it may also raise the noise floor by pushing the op amps closer to their max limits.

So I guess what I'm saying is there's no real need to go out and replace the stock wall wart.

[MJM]

Thanks for telling about your efforts. I've heard many needledrops done with a Bellari with a Sylvania tube, and they all sounded good. But I have noticed a bit of distortion, for example on cymbal crashes. I wouldn't want to hear a classical record through this, but for blues and rock it does have a nice sound. I've seen them for $200 new, so I may pick one up someday just for kicks.

[Brinkman]

My positive experiences have led me to think the Sovtek 12AX7LPS is one of the best current-production miniature tubes out there. Optimizing a circuit for use with this tube was a very wise and pragmatic effort on Bellari's behalf.

Personally, I would love to roll one of these bad boys in there. Might be very advantageous to use in a starved-plate design...

[Brinkman]

So yesterday I received the Techtube E813CC (12AX7 replacement) and inserted it into the Bellari VP-129. Unfortunately, because I was barbecuing ribs for the dinner engagement later in the evening I was only able to listen for about two hours or so.

My impressions?

Upon initial power-up I heard an audible (but not loud) ocean-like static sound coming from the left channel of the Bellari and was worried there was somehow some incompatibility with the Techtube. The good news is the noise was gone within the the duration of one side of an LP, so whatever problem there was resolved itself during use.

The change in presentation is rather remarkable. Individual elements of songs come to the fore in a very detailed manner and the tempo seems to take over and make the room breathe with it. In fact, considering that I was a bit hung over, this seemed a bit disconcerting at first so I felt compelled to change listening material to determine if this was just a characteristic of the tube itself.

For instance, reviewers will state like to use the cliche that such-and-such a change led them to hear things in familiar tracks they'd never noticed before, but in this case familiar songs became less recognizable. I really hate to use this analogy so forgive me, but it was like listening to a familiar track for the first time while stoned; percussive elements took on an infectious vibe, backing embellishments became less inconspicuous and predominant features were magnified a bit. Toe-tapping was at a maximum.

I have no complaints with frequency extension as bass was strong and treble was clear, and the tube was neither veiled nor was it bright. In fact, a sibilance issue I've been encountering for a while has seemed to have been remedied a bit with this tube, a benefit that has me rather ecstatic. I'd conjecture that the input capacitance of this tube is different that the 12AX7LPS, so perhaps this tonal improvement is the result of capacitive loading differences when used with my ortofon OM20 cartridge. For that reason alone, the E813CC may be a permanent part of my Bellari VP-129.

So all I was able to listen to was a bit of my 180mg Silver Apples S/T reissue, a side of a Seeds reissue and a side of my "African Scream Contest" compilation (amazing comp BTW). Later in the evening one of my guests wanted to hear my system but the rectifier in my amp blew during power-up, so this is unfortunately all I have to base my impressions off of until my replacement arrives. When it does, the E813CC will definitely be put to more use, and I'd encourage anyone who uses a 12AX7 to seriously consider the Techtube E813CC.

[ehoove]

This is an assesment of the TJ Full Music 12AX7

As a background I offer this post on my Bellari VP129:

The Mullard 12AX7 Re-issue is the overall winner. Silky smooth midrange/fast resolving top end, and tight solid bass with tons of upper-bass definition. It has a wide deep sound stage with precise player placement. Sounded a tad too warm until after around 15 hours when the pace picked up and took off.

2nd place (Tie) The JJ Telsa ECC83S showed more air in the top end, slightly dryer in the mid range, with good solid bass. not quite as rich as the Mullard, and a shallower sound stage but quite good. Great match with the Sumiko BluePoint #2

2nd place (tie) The Sovtek 12AX7LPS had a overall very nice presentation with just a hint of dryness on female vocals, but very solid on Rock and Roll pieces tighter sound stage but deeper than the JJ Telsa. Was my favorite tube with the Denon DL160.

3rd Place TungSol 12AX7 This tube just did not get it done for me, but didn't do anything wrong. It just did not make me want to put it back in after 3 turns. Wide Un-defined sound stage with medium depth. I may still try another one as it may just be that sample, because I know some really like this tube

4th Place goes to the Groove Tube 7025. This was the brightest of the group, with a real solid overall presentation, that at first was almost hypnotic in the detail area, although it was very fatiguing after a short time. I like this tube and thought it rated much higher over Christmas when I used it with my Allison Two's. It is just too bright a tube to use with the MGIIIa Ribbon Tweeters IMHO

While these are my ratings please note the equipment listed as I really feel system synergy has a lot to do with the final result. I will have some time this weekend, and will redo the test using my AKG K701 headphones straight into the headphone output of the Bellari VP129, which is how I listen to Vinyl with the Cans. I will then compare the results.
Regards,
Jim
__________________
Main>Acurus L10/Bellari VP129,Acurus A200-A200X3,Beringer CX2310 X-over,Pro-Ject RM5SE/SpeedboxII/Sumiko Blackbird,Marantz SA8001 SACD,Magnepan MGIIIa,NHT W2 Sub,Cans:AKG701


So I was offered an in system demo of these high end tubes from Ian at http://grantfidelity.com/site/ , and here are the findings after burn in and listening.

Wide open sound stage rather deep with very good player placement. Mid range was silky and a tad dark. The bass was solid tight and fast, with upper bass to die for. The top end ..... Silky smooth and very delicate with good attack and pace. The stick hitting the cymbal was there, and the ride was right on. Vocals were very lifelike and believable. Overall a solid if not amazing presentation. At 3 times the price of the Mullard re-issue, it is still a value in my book worth a solid look. This tube will be my new Bellari VP129 Tube. For Now :>)
Regards,
Jim
_________________
Let's drop the stylus in the groove and listen to some tunes!

[Brinkman]

Because I was so impressed by the Bellari with the Techtube E813CC installed, I decided to measure the input capacitance of the Bellari with and without each of the four of the 12AX7s I have at my disposal. I was curious to see if differing input capacitances were providing capacitive loading in regards to the cartridge on my tonearm, which would in turn affect the the frequency response of the Bellari.

So below is a post of the Left and Right channel input capacitances (in pF) of the Bellari with a given tube. Click on the tube to read it's provided datasheet.

Tube Left Right

Sovtek 12AX7LPS 201.5 200.4
Ei ECC-83 196.4 197.7
Electro-Harmonix 12AX7EH 193.6 194.2
TechTube E813CC 191.3 194.0
none 168.8 173.8

If you peruse the datasheets, you'll see all but one specify a grid-to-cathode capacitance of 1.6pF, with the exception being the Techtube at a low 0.4pF.

Out of curiosity, I did a cold measurement of each tube between it's grid and cathode (pins 2 and 3, respectively). All the tubes had a grid-cathode capacitance of 2.5 - 3.0pF, with the lowest being the Techtube. I'd imagine operational voltages and the heating of the cathode increases conduction through the tube which lowers this capacitance to the measurements seen on the datasheets. If this is true, one of the advantages of the planar design of the Techtube is this lower operational inter-electrode capacitance.

Unfortunately, I don't see a strong link between circuit measurements I took and the info on the datasheets. So whether or not the low capacitance of the TechTube is behind my preference for it in the Bellari is still up in the air.

I'd still like to try rolling a New Sensor Mullard or the TJ Fullmusic 12AX7 in there...

[W8AAZ]

I once had a Dynaco PAS-3 that I upgraded with new 1% resistors and better caps. I had access to old Telefunken and Mullard 12AX7's to compare in the unit. Musically, I preferred the sound of the old Mullards as they were noticeably better sounding to me. Cannot think of any esoteric superlatives to describe it now, this was over 20 years ago, but it is certainly fun to experiment with different types. And easy.

[Brinkman]

So today I went through with swapping out the U3 and U4 positions of the PCB with op amp sockets. If you refer to the schematic, you will notice U3 handles the left channels for both the line and headphone outputs, whereas U4 handles the right channel duties. You will also notice the schematic erroneously designates the op amps as BA4580s, as opposed to BA4560s. Not a big deal.

What is unfortunate, however, is that if you want to roll op amps, you have to not only do it in pairs, but dual pairs. Which means I need to order mates for the seven or eight duals I have floating around if I want to audition them. Which leaves me with the LM4562s and OPA2134s of which I have several.

Also, I should note swapping the BA4560s with op amp sockets was a total PIA. I destroyed not only both BA4560s, but I melted an op amp socket and pulled a trace and a couple solder pads in the process. No wonder I was too exhausted to install bypass caps. I'll save that for next week.

The first op amp up was the LM4562. The tube hadn't really warmed up for more than a few minutes before I started my evaluation, but I really can't say I was that impressed. I felt like I was missing some low-end. A few episodes of Lost later (I'm in the middle of catching up to the current season), I rolled in the OPA2134. Boy, was I impressed. Even though the OPA2134 is kinda an old man in op amp terms, I immediately noted the bass was back, without seeming to sacrifice anything else. In fact, the Bellari itself sounded more resolving and fluid than I could recall it ever being before.

I'm not really trying to go overboard with descriptors because I'm not trying to assert a "best" op amp for the Bellari. In fact, the warm-up time could have a lot to do with the differences, as could the installation of some bypass caps down the road. What I'm trying to say is that op amp rolling, like tube rolling, is a worthwhile consideration for this model Bellari and a stock VP-129 can definitely be improved upon in this manner.

Have fun.

[BartJY]

Brinkman,

Does any of the above information/mods apply to the V130?

Thanks, your posts have been very interesting.

[Brinkman]

From the Needle Doctor website: "The VP130 is exactly the same as the VP129, except that the 130 uses a 15V, switch-mode energy-saving power supply."

So yeah, it would apply. I'll change the thread title to reflect this.

[BartJY]

Very good.

Thanks.

[fjdaigle]