[ncaudio]

Bob,

Good stuff, I have 2 Toshiba 3960's (one $65 new from best buy, one $16 from ebay), a panasonic S47 and a Toshiba 4960 (all bone stock). All impressions are using units as transport only playing redbook CD's  feeding 2 pretty good systems.  Sound quality ranking (worst to best) Panny S47, Tosh 3960, Tosh 4960.

I think if the 4960 had come out before the 3950/60 they would have grabbed everyone's attention (although maybe not quite as easily modified with single board) and people would have jumped all over them. I think the cheapskate in a lot us still would have liked the idea of a cheap(er) player like the 3950/60, but I think the 4960 sounds better to start.

The last thing you asked is what other people suggest for future mods, I would suggest mechanical mods in the form of damping the chassis, maybe try some ERS cloth, better power cord.  For damping, try floor tiles from Home Depot, duct seal (clay like stuff) and (for transport use) a high quality digital transformer (Lundahl).

As an aside, I really wanted the Panasonic S47 to best the 3950/60, nicer display that tells something, cool looking blue led lit clear plexigas drawer and with a buffer so big that the player continues a full 15 seconds after the disc is stopped (mechanically) while playing.

Roger

[nathanm]

If either bobm or Rob Babcock wishes to edit bobm's posts you would just go in and remove the following string from his\your picture links and they will all work.

Code: [Select]

?action=view¤t=
Of course they'll all still be on the huge side, but hey better than broken!

[Rob Babcock]

I went in and tweaked them per Nathan's advice and it looks like the pics are there now.

[BobM]

I tweaked the image url's and I can now see them on my browser (when Photobucket.com is up). Oh well, just use the links themselves in that case - sorry for the inconvenience.

I've got some Shottky diodes coming in (4A 600V) and will definitely have to do some creative component placement to get them installed. The big power supply cap sits right on top of at least 2 of the existing diodes. The problem is that the new Shottky's at TO-220 type and stand up off the board, while the original's look like typical round diode component's that lay flat to the board. I expect that this change will speed things up a bit - it did exactly that when I modded my DI/O. We'll see.

One other observation this weekend, on SACD I noticed a dramatic difference between a cold player and one that has had at least 45 minutes to warm up. Don't judge this player in its cold state. I believe Swenson noted this with his 3950 - that it sounds much better after the chips are hot. But you have to have a disk spinning or it will shut itself off (anyone know how to diable this "feature"?)

As for damping, I've got a reasonable amount of Moretite all over this unit - chassis, transport, cover and such. I've probably doubled the weight of this unit (it's now probably a heafty 10 lb beastie   ).

Enjoy,
Bob

[BobM]

Got the Shottky diodes in on Friday and put them into the 4960a nd let it burn in (Cree CSD04060A 4A 600V zero recovery rectifiers in a TO-22002 style).  I went with a 4A diode because some others had reported problems with 1A replacements blowing up in their 3960 mods. It's overkill but that's often what mods are all about.

The original part is a 1A 600V standard style (looks like a resistor) - the new ones are TO-220-2 style. Because of the difference in style between the small round original diode and the new TO-220 I had to deal with placement problems. These diodes are right next to (and under) the big cap. I managed to put 3 of the diodes in a row right un the top of the board. The 4th one had to go on the bottom of the board.

BE CAREFUL: these TO-220 stlye diodes have a conductive metal backing. I wrapped all 4 in electrical tape so they wouldn't short against anything. I also put some tape on the chassis under the board where that 4th one sits, just in case. If you have heatshrink to use around them it might even be better. Make sure you double check the orientation of each diode separately, they do not all go in the same direction on the board.

Since I had everything opened up I also applied some Mapleshade Silclear on the fuse ends and where the power cord snaps into place. I have a Belden 83802 cable coming in the mail to replace the stock power cord. In the meantime I treated the the stock cord's plug tongs with more Silclear, and plugged it into my Jon Risch digital isolation transformer which is fed from a Jon Risch choked power conditioner. I think this is as clean a power supply as I'm going to get without replacing the transformers inside the unit.

Enough for prep. I definitely got some additional transient speed out of the unit with the new diodes, as I had hoped. I think I also got some additional top end extension and bottom end tightness - that could be the Silclear. Spacially, things opened up into a more 3 dimensional way - that is probably the power conditioning and digital isolation transformer.

I think I've gone as far as I am capable of going on my own with these  internal mods. I think perhaps I'll try a piece of metal (aluminum) epoxied to the bottom of the unit and some DIY rollerblocks for it all to sit on.

I would love to hear how some of you others are progressing and anything new that you might be doing to your players. I'm very happy with the sound now, and will probably use this unit to replace my old and increasingly finicky Cal Labs DX2 transport (intermittant digital hash is coming out of this unit now, so it's time to retire it).

The DI/O still has a slightly better 3 dimensional presentation (very slight). The modded Toshiba has a more organic presentation which is probably due to the capacitor analogue outs rather than the op-amps. I would say that the Toshiba's sound is very close to the DI/O's sound using a transformer output instead of the op-amps (I built one of those too about a year ago). Laid back but detailed, sweet and enjoyable. A nice combination to be sure. I will probably go back and forth between the two for some time and never reach a definite conclusion about which is better. A nice problem to have, to be sure.

Enjoy,
Bob

[dnlyko]

Can you choose a DVD-A group without using a TV?  Does it have a group select button on the face or on the remote or do yo have to have it connected to a TV to navagate the DVD-A groups?

[BobM]

The biggest flaw in this player, aside from the cheaply made transport and casing, is the user interface. Very little information on they player's display. Just track number and time pretty much. That's OK for CD's but not so good for SACD and especially DVD-A (where there is a menu structure involved and it can be different for each disk).

You can pretty much play anything easily. But if you want to use the repeat fuinction, or program the play order, or anything more advanced like that, then you're guessing without a monitor attached.

I have a portable DVD player that we use in the car on long trips. I have the red/white/yellow cable hooked up to the back of the Toshiba and coiled on the side. When I need it I plug it in and a few seconds later I'm set and running.

On another topic: Just tried some Vibrapods under the unit and it's all to the good. Isolation pays off. I've got som Mi-Rollers coming in the mail. If you haven't seen it yet look at the review on http://www.EnjoyTheMusic.com. Then go to http://www.MiHorn.com and see the special pricing pop-up. They've lowered the price of these down to a "what the heck" level to coincide with the positive review.

Enjoy,
Bob

[Sei-young Jang]

Hello Bob!

I read your post in digitaldrive with lots of fun and read through all your posting in this thread. Impressed! Nice job! I'm determined to buy this player and do the mod.

Concerning the X-cap issue, I just wanted to give my 2 pennies, from my 10+ years experience in SMPS design. Across the AC line you normally see 115Vac, which could be about 200V at peak, whic is still far below 600V auri cap is rated. But practically across the AC lines, there are spikes and transients which can be as high as 3000V. They come from the inductive loads connected to your AC mains, like motors etc. One of the charateristic requred to be approved by UL or IEC is "self-healing" characteristic after puncture has occured. There are cases where actually fire has occurred because of the breakdown of insulation between primary and secondary. I have no idea this auri cap is safe from flammability as well. As this component is inside the metal housing, it may not cause fire but i thought why take such risk, while you can simply by X-cap. X-caps are usually PE and Auri is PP, so should have better freq response, but... Good X-cap are encapsulated in epoxy molding annd there are reasons. So choice is yours.

When I buy this player I think I will try using high-mu output transformer at the output of DAC IC which has balanced output. I did it for my P3A and was shocked by the improvement.


Anyway nice job and thank you for all the info you shared!
Sei

[BobM]

I think a transformer output would be a great idea, unfortunately you may have to place them outside of the unit. I built a transformer output on another DI/O that I had (called it my "FrankenDI/O") using Edcor transformers. It had a very nice sound. I went back and forth between the op amp versioin and this transformer version for some time - not ever really sure which I liked best.

I stuck with the op-amp mainly because it was a bit more dynamic with ever so slightly better transient response. Very much the same issue I have with the Toshiba's coupling cap output vs the DI/O. Enjoyable both ways.

Good luck with your project. Gl;ad I could be of assistance, and please post your impressions after the mods are done.

Bob

[BobM]

Here's a bit of news that won't surprise anyone. I experimented with isolation under the 4960, using Vibrapods and Mi-Rollers (Jr's). This unit DEFINITELY benefits from vibration control. Given the flimsy cabinet and plasticky nature it shouldn't be a surprise to anyone.

The Vibrapods gave a nice rounded but detailed picture. The Mi-Rollers were a bit cooler but seemed to bring more air and precision to the sound. Now this was not overly analytical or etched by any means, smoothness guaranteed by the coupling cap Swenson mod. Pick your poison and go with it.

Here's a link to the Mi-Rollers (at www.mihorn.com) - now on sale to coincide with a review of the product at EnjoytheMusic.com. Definitely a nice inexpensive alternative that works, just like Vibrapods.

http://mihorn.com/miroller.htm

Enjoy,
Bob

[adamofwales]

************NEUB ALERT***************


I take it that you could perform these same mods on a Samsung HD-841 just as well, since it's basically the same unit.  Right?

[_scotty_]

A question for modifiers of the Toshiba 4960 and Samsung 841. Do these players have a buffered crystal or clock in them. Thanks, Scotty

[BobM]

Yes, the Toshiba SD4960 is the same machine as the Samsung HD841. In fact, I obtained the Samsung's  maintenance manual to get the schematics. Here's where you can obtain them directly, if you're so inclined:

J&J Int'l at 1-800-627-4368

I believe it's just a buffered crystal, at least it looks that way to me, but I may be incorrect here. All I did was damp down the crystal with some Moretite and didn't pay any attention to the circuit around it.

Yes, I'm sure things would improve (especially the PRAT and probably jitter reduction) if you upgraded to a Guido Tent clock or something similar. But that kind of upgrade alone would cost more then the unit and the mods that I've specified together. Then there's finding room inside the case for it.

Enjoy,
Bob

[BobM]

Put a new power cord on it yesterday - Belden 83802 soldiered directly to the pins on the main board. This cable is rather stiff and poses a slight problem with the underlying rollerbearings under the unit. You see, I've put the Toshiba up on 3 MiRollers (http://www.mihorn.com). These are a slightly different implementation of the popular roller bearing style of isolation. But the stiff cable restricts the movement somewhat. Don't know that it make an audible difference, but thought I should mention it. The MiRollers or any other kind of rollerbearing isolation are highly recommended.

These last 2 changes have caused me to take my DI/O out of the system and shelve it. The 2 units are so close now that I doubt I could tell the difference in a sighted test, never mind a double blind. There's just no need to go through the hassle of using the external DAC now.

I'll keep the DI/O around though, since it definitely sounds audibly better with a better transport, so just in case I get a gifted Meridian from someone in the future ...

One question: I now have a single small Bybee slipstream that I was using on the digital input to the DI/O. Any ideas where to use it now?

I guess I could put it on the + power line coming into the Toshiba, but there's a lot of filtering there already (Jon Risch choked power conditioner AND digital isolation transformers) so I'm not sure it would add much. Thoughts?

Enjoy,
Bob

[_scotty_]

Bob, if there is a hex-inverter as part of the circuit containing the crystal this player would have a low impedance clock built into it.  You would have to look at the parts around the crystal and look up the numbers online of any likely looking suspects. Scotty

[BobM]

Scotty - not sure what you're getting at or suggesting here. Can you please explain? Are you hinting at a better quality clock replacement? Or a replacement of the stock clock parts and crystal with ones of better quality?

If that's the case then I'm all ears, but I doubt that I could justify a $300 or so clock replacement using some of the better regarded units (i.e. Tent clock and such).

Thanks,
Bob

[_scotty_]

What I am saying is that if you can identify the parts around the crystal you may find that there is already a low impedance,high speed, high accuracy clock circuit
already in place. This is becoming increasingly common in all multiformat  players. The parts cost to include such a clock as part of the player is around two dollars.
No replacement may be necessary at all.  The presence of such a clock may in part contribute to the observed performance gains that come from modding the player.  Scotty

[_scotty_]

What I am saying is that if you can identify the parts around the crystal you may find that there is already a low impedance,high speed, high accuracy clock circuit
already in place. This is becoming increasingly common in all multiformat  players. The parts cost to include such a clock as part of the player is around two dollars.
No replacement may be necessary at all.  The presence of such a clock may in part contribute to the observed performance gains that come from modding the player.  Scotty

[BobM]

I added .1uF Relcap RT's as a bypass to the 3uF Sonicaps that I'm using in the Swenson type analogue outputs. Now normally, one wouldn't use 3uF caps for this function and stick more to the 1uF or .47uF caps that John Swenson recommends. However, my preamp has a high input impediance so I thiought a larger value would deliver better bass.

It did, but I believe the top end could use a little improvement - therefore the .1uF Relcap's as a bypass (I had them lying around from another project anyway).

As you can guess they did make an improvement. Not in top end extension, but in better soundstaging and presence. More air and definition to things.

I compared this coupling cap output vs. the Toshiba's regular multichannel op-amp output (remember to set the menu-audio multi-channel controls so that there is no rear or bass or center speaker output, only the left-righ speakers full range). Initially the op-amp outs seem to be brighter and more forward. The Swenson mod is smoother and more enjoyable with lots of presence and foot tapping enjoyment. The op-amp output is clear and extended, but seems to have a bit of that digital edginess that goes up my spine. It also doesn't have that ease of listenability or the soundstage depth that the cap output produces.

I have no doubt that replacing the op-amps here would help things, but there's no need unless you're looking for multichannel outs. Interesting comparison though, and it easily shows the improvements from the Swenson mod over the regular outputs after the power supply mods.

Enjoy,
Bob

[BobM]

Another satisfied modder revealed. Read his review at Audio Asylum's digital forum:

http://www.audioasylum.com/audio/digital/messages/95353.html

Thanks Les for confirming my results.

Bob