One member responded to this and offered to sell me his completed unit. So now I have a working version, but have only had a few minutes to listen so far.
Since I showed up so late in the thread, let me briefly explain my interest. A friend has a Soundsmith SG and it has some undeniable qualities. But I'm unconvinced about some other things. In particular I was unimpressed with the idea that we should just ignore the frequency response deviations and not try to optimize that part of the performance. Being an incorrigible modifier/improver, I want to try a few things, but I can't really do a lot of experimenting on a multi kilobuck cartridge. The Panasonics are available enough and cheap enough that I can try some things out. I have been listening to one with the Panasonic preamp and dreaming about how much better the cartridge could sound with a nice tube gain stage.
I had been intending to build a tube bias/gain/EQ stage and see about getting a smooth and widerange FR performance. I had in mind a somewhat different circuit than Kevin's, which I may yet build up. But the EQ cards in Kevin's circuit make it easy (comparatively) to jump right in and get started.
The previous posts on damping and bodybuilding also point to possibilities for big improvements, as do the suggestions about different cantilevers and styli. I will explore those avenues too, but first, the EQ.
I would also be interested if you ever do decide to make boards available again. I have two EPC450CII cartridges and no preamp for them
I have a new experimental EQ I will be implementing in the next few days, it requires some modifications to the amplifier cards, and more or less a hack job on the EQ module. It is much simpler but some account of miller capacitance in the second stage may be necessary. ( Possibly may be necessary to trim resistor value for best results.)
I will report back once I have some results. In theory this should result in considerably better HF response, the Achilles heal of my current design.
If it works as hoped and there is sufficient interest I will make new blank EQ PCB available to current owners at cost.
Details attached, note I have not tested, experiment at your own risk!
I will report back once I have some results. In theory this should result in considerably better HF response, the Achilles heal of my current design.
If it works as hoped and there is sufficient interest I will make new blank EQ PCB available to current owners at cost.
Details attached, note I have not tested, experiment at your own risk!
REV.8 Strain Gauge EQ
I do not recommend implementing the EQ in the previous post, that design has an unintended effect on the low frequency response - very bass shy.
For those interested I have a new EQ that you can experiment with and am in process of getting some real boards made to test in my pre-amp. I am going to release the Gerber files for these boards and you can go to Oshpark and get your own made for roughly $4.00
The new filter requires some minor rework to the pre-amplifier card which consists of removing R8, R9, R10, R11, and C3. R10 is jumpered, and there is a jumper from R9 and R11.The EQ will now all reside on the EQ card, except for the final LPF which relies on the rather excessive miller capacitance of the succeeding triode amplifier stage to do its job. (Fortunately that value is not extremely critical.)
I'm a lot happier with this EQ than I was with the prior. I will probably continue to improve the EQ where possible.
Basically low pass step filter between 0 - 200Hz, a high pass step filter between 500Hz - 1.8kHz, and an LPF at 5kHz. Coupling capacitor choice provides about -3dB at 10Hz high pass to partially compensate the cartridge's rising bass response below 50Hz.
Note1: I don't recommend reworking the EQ card to implement this EQ as it is a kluge at best. New boards from Oshpark are nominal cost.
Note2: Connector shown on bottom view of EQ board is wrong gender due to a library issue, but is correct footprint for the required part.
Note3: OSH Park ~ Just drop the zipped gerbers on this point indicated and they will do the rest. (Total will come to $4.10 if you just need one set and a spare..)
BOM added
Please provide feedback if you do the upgrade described in this post. Thanks!
I do not recommend implementing the EQ in the previous post, that design has an unintended effect on the low frequency response - very bass shy.
For those interested I have a new EQ that you can experiment with and am in process of getting some real boards made to test in my pre-amp. I am going to release the Gerber files for these boards and you can go to Oshpark and get your own made for roughly $4.00
The new filter requires some minor rework to the pre-amplifier card which consists of removing R8, R9, R10, R11, and C3. R10 is jumpered, and there is a jumper from R9 and R11.The EQ will now all reside on the EQ card, except for the final LPF which relies on the rather excessive miller capacitance of the succeeding triode amplifier stage to do its job. (Fortunately that value is not extremely critical.)
I'm a lot happier with this EQ than I was with the prior. I will probably continue to improve the EQ where possible.
Basically low pass step filter between 0 - 200Hz, a high pass step filter between 500Hz - 1.8kHz, and an LPF at 5kHz. Coupling capacitor choice provides about -3dB at 10Hz high pass to partially compensate the cartridge's rising bass response below 50Hz.
Note1: I don't recommend reworking the EQ card to implement this EQ as it is a kluge at best. New boards from Oshpark are nominal cost.
Note2: Connector shown on bottom view of EQ board is wrong gender due to a library issue, but is correct footprint for the required part.
Note3: OSH Park ~ Just drop the zipped gerbers on this point indicated and they will do the rest. (Total will come to $4.10 if you just need one set and a spare..)
BOM added
Please provide feedback if you do the upgrade described in this post. Thanks!
Attachments
EQ Rev8
Kevin, this is good news.
I had to set this project aside for a while 'cause things got...a little complicated. But I am getting back to it now. The unit I bought had Rev5 EQ, and I was gearing up to bring it up to date to Rev7. Now I'll skip that and go right to 8. Have ordered the boards, will start to gather the other parts.
Once it's completed I will be able to do listening and some rudimentary measurements. I don't have all my test gear up here at Hole In The Wall, my hideout in the Sierras. I'll be able to measure deeper once things settle down. Mostly, though, I look forward to the listening.
Kevin, this is good news.
I had to set this project aside for a while 'cause things got...a little complicated. But I am getting back to it now. The unit I bought had Rev5 EQ, and I was gearing up to bring it up to date to Rev7. Now I'll skip that and go right to 8. Have ordered the boards, will start to gather the other parts.
Once it's completed I will be able to do listening and some rudimentary measurements. I don't have all my test gear up here at Hole In The Wall, my hideout in the Sierras. I'll be able to measure deeper once things settle down. Mostly, though, I look forward to the listening.
I'm looking forward to your feedback, I have struggled to EQ this cartridge, but I am happier with this than I was with the previous iterations. The simulations align fairly well with the known anomalies and at least partially correct them without screwing things up elsewhere.
I expect I will find the need to further fine tune over time. I just switched from an ET-2 with a 451CR on it to a prototype Sorane ZA-12 with a 465CR on it. Probably needs a bit of work there too.
The 465 has sat for 2 years since it was last used and they get a bit weird when the elastomers are not exercised for an extended period of time. I also suspect that they may have some sort of warm up time, I again suspect it is mostly the few mW of heat in the confined space just heating the elastomers a bit.
It's not quite a match for a Mutech Hayabusa on a TK-850L into one of my Muscovite Mini IV pre-amps, but it is very pleasant and relatively resolving. I can probably do something to better damp the cartridge.
I expect I will find the need to further fine tune over time. I just switched from an ET-2 with a 451CR on it to a prototype Sorane ZA-12 with a 465CR on it. Probably needs a bit of work there too.
The 465 has sat for 2 years since it was last used and they get a bit weird when the elastomers are not exercised for an extended period of time. I also suspect that they may have some sort of warm up time, I again suspect it is mostly the few mW of heat in the confined space just heating the elastomers a bit.
It's not quite a match for a Mutech Hayabusa on a TK-850L into one of my Muscovite Mini IV pre-amps, but it is very pleasant and relatively resolving. I can probably do something to better damp the cartridge.
I was previously using C2 = 0.1uF on the pre-amp board and simulations previously indicated that 0.068uF was a better choice to deal with the rising response below 50Hz. See the pdfs for the details, the first one contains graphs aligned showing the cartridge response (from long ago review) and the current EQ. The other is an updated schematic.
I think I slightly prefer 0.068uF based on a short listening test.
I think I slightly prefer 0.068uF based on a short listening test.
This is a test jig you can use for pre-amplifier measurements, while it is intended for use with the strain gauge pre-amp discussed in this thread it is suitable for use with others, and offers a reasonably accurate attenuation of -40dB if fitted with 0.1% resistors.
The Gerber Files are included and these boards because they have a fair number of vias will run $7.50 for 3 at Oshpark. They are intended for personal use. (zip file)
Values, tolerance and recommended manufacturer for parts is included in attached schematic. (pdf file)
1 100 mil 2 pin header and jumper is also required. Gold pins recommended.
10 ohm resistor (R4) not critical as to value or tolerance. (1206 package)
Use BNC, RCA or any other means of connection preferred.
Solid ground plane on bottom, top ground plane is tied to bottom at multiple locations using vias.
The Gerber Files are included and these boards because they have a fair number of vias will run $7.50 for 3 at Oshpark. They are intended for personal use. (zip file)
Values, tolerance and recommended manufacturer for parts is included in attached schematic. (pdf file)
1 100 mil 2 pin header and jumper is also required. Gold pins recommended.
10 ohm resistor (R4) not critical as to value or tolerance. (1206 package)
Use BNC, RCA or any other means of connection preferred.
Solid ground plane on bottom, top ground plane is tied to bottom at multiple locations using vias.
Attachments
Regarding post #654, changing C2 to 0.068uF seems to be subjectively right based on several nights of listening. Overall this EQ sounds much closer tonally to my latest LOMC which is a Hayabusa, and a pretty good reference as these things go.
Please let me know if you are upgrading to the new EQ.
I have received all of the parts except the boards which went to fab a couple of days ago, and ahead of schedule, so I am hoping to have them by late next week.
A few days ago I posted an offer to provide boards or Gerbers for another GB. This release would include all of the updates specific to the new EQ. Over the past few years I have received a lot of queries and requests to offer the boards again. I suppose the current timing is not the best, but so far no interest has been expressed. I will update the board design, and if the interest is insufficient for a GB I will provide the Gerbers by email only to anyone who agrees not to share them with others. This will of course include the Gerbers for the PSU as well. No power transformers, and no hand holding outside of this thread.
JLCPCB would be the recommendation for getting boards made - they are 2 layer.
Please let me know if you are upgrading to the new EQ.
I have received all of the parts except the boards which went to fab a couple of days ago, and ahead of schedule, so I am hoping to have them by late next week.
A few days ago I posted an offer to provide boards or Gerbers for another GB. This release would include all of the updates specific to the new EQ. Over the past few years I have received a lot of queries and requests to offer the boards again. I suppose the current timing is not the best, but so far no interest has been expressed. I will update the board design, and if the interest is insufficient for a GB I will provide the Gerbers by email only to anyone who agrees not to share them with others. This will of course include the Gerbers for the PSU as well. No power transformers, and no hand holding outside of this thread.
JLCPCB would be the recommendation for getting boards made - they are 2 layer.
I've got everything but the boards - including the parts for the test jigs.. LOL
I have been listening a lot over the past few days using a pair of the old card design that I bodged to implement the new EQ.
I am pretty happy, and am using one of Dave Slagle's boron cantilever MR stylus equipped replacement needles - it is a lot better than even the Panasonic OEM shibata.
Replacement styli are highly variable in quality, but there are some good ones, hit or miss pretty much is all I have been able to figure out since the OEM styli became a bit rare. (I have saved mine, one or two of them will be suitable for a future retip.)
I have been listening a lot over the past few days using a pair of the old card design that I bodged to implement the new EQ.
I am pretty happy, and am using one of Dave Slagle's boron cantilever MR stylus equipped replacement needles - it is a lot better than even the Panasonic OEM shibata.
Replacement styli are highly variable in quality, but there are some good ones, hit or miss pretty much is all I have been able to figure out since the OEM styli became a bit rare. (I have saved mine, one or two of them will be suitable for a future retip.)
My boards are in the mail and I hope to have them by this week-end. I'll report back on the outcome once I get them built. I went with standard commercial grade electronic components on these boards because I noticed I could the required capacitor precision fairly easily. (And they are a lot more compact - we'll see how they compare to the bodged boards with standard audiophile fare..)