What a cool trick, after comparing USFP w/MC and FSP using the Shure V15 I settled on 62K (the actual sweet spot was 66.2K). Since I installed pins in place of R1, I was able to install the trimmer (62K & 68K) and then swap the 47K resistor, 62K sounds the best regarding overall response. The gain difference between the USFP with a MC and the FSP with the MM was negligible, so I will not be making any changes there.
It was very nice to be able to switch inputs while listening to the same recordings (double album), the MC/UFSP combination is brighter (a little sizzle on "S's" of voice compared to the MM/FSP combination. The MM/FSP combo mellows the sizzle of "S's". Perfect pairing since some recordings need the brightness and some do not.
Once again, thank you for all your help and wisdom.
Yes, Tea Bag sent three different sizes with pins, I ended up with the 220pF as it sounded the best. Although, it was hard to tell the difference between the three.
Yes that's the most popular value. If some MC brightness is due to overall response, less Ω input loading or more C2Y pF can help. If it's due to little mistaking, it's more down to revisiting overhang, VTA, VTF etc. But sometimes it's just the tip profile, cantilever, and suspension design. How much tip to vinyl HF resonance.
Will the FSP using the Shure benefit from adding 110pF to the input since Shure recommends 250pF and I have 140pF into the FSP?
I have a third turntable (Thorens TD 125 MK1) and preamp (Threshold FET 10Pe), I'm using an SME 2009 series 2 improved with a Sumiko Blu Point Special high output MC. The Sumiko recommends 47K and 250pF load, I would like to try it with the FSP, any recommendations for tuning?
I have a third turntable (Thorens TD 125 MK1) and preamp (Threshold FET 10Pe), I'm using an SME 2009 series 2 improved with a Sumiko Blu Point Special high output MC. The Sumiko recommends 47K and 250pF load, I would like to try it with the FSP, any recommendations for tuning?
It already has 105pF of MM mode capacitance (Ciss+2xCrss Miller as Cascode of single 2SK369). Approximate value because the Rs degeneration resistor eats away some of those pF but the build adds some. RCA connectors, internal signal cable, layout etc. have some pF.
The above does not exclude you may find some happy combination of even more pF and a custom kΩ input load value.
About the Sumiko, HMC carts although there are Cload recommendations some times, they have far less Zout than MM thus they are far less susceptible to Cload variations. They usually play more in line at the 43dB gain mode. But they still work with 40dB.
The above does not exclude you may find some happy combination of even more pF and a custom kΩ input load value.
About the Sumiko, HMC carts although there are Cload recommendations some times, they have far less Zout than MM thus they are far less susceptible to Cload variations. They usually play more in line at the 43dB gain mode. But they still work with 40dB.
Also see this load report for a V15 type III and the Valve Itch phono: https://www.diyaudio.com/community/threads/valve-itch-phono.140635/post-6917554
(That V15 has an old replacement tip)
(That V15 has an old replacement tip)
I should be running and testing UFSP soon, but a couple of questions please!
Question about LED strings was asked on #18,918, then answered on post 18,920 - From your answer I'm going with 4 leds, 1st channel VF = 7.12 and second channel VF=7.13.. These were sorted with 9v batt and 1.5k resistor and the 4 string measured with 9v batt and 1,5k series R.. This is good?
and the 3 string led in UBIB is not so important to match?
Question about LED strings was asked on #18,918, then answered on post 18,920 - From your answer I'm going with 4 leds, 1st channel VF = 7.12 and second channel VF=7.13.. These were sorted with 9v batt and 1.5k resistor and the 4 string measured with 9v batt and 1,5k series R.. This is good?
and the 3 string led in UBIB is not so important to match?
Last edited:
Hello,
so i have made three recordings. One will be using old russian military caps, one using Mundorf supreme, but too big of a capacity value, and third using required 0,1uF Mundorf supreme caps in interstage C3. All three for blind A/B testing. Also a link for recording of third configuration separately. Any comments highly appreciated. Thanks.
3 recordings for A/B: https://we.tl/t-F5D2CVrGJ7
Final configuration: https://we.tl/t-pUApWjOWcB
I think this is highly system dependent. I've found that playing with treble trim and loading with Ultra had a bigger influence in my setup. I also have FSP running config for MM, but with Cinemag SUTs for MC on another arm/cart combo in addition to Ultra. What cart seems to be the biggest influence in SQ to me. I move between HanaML, OC9XML, and a Soundsmith Zephyr MIMC star, on different arms—very easy to A/B—and they are all glorious with subtle micro differences depending—really can't decide what's "better"... One could say the SUT route is more laid back, and the Ultra as compared is more precise and up front. Just words though... and thoroughly meaningless? Music style is a thing too... We listen to a lot of Jazz, and not much therein past 1968 or so... but also a lot of pop and rock and blues... soooo.....
Anyway given all that... try something from Clarity, I had Mundorf Supreme in there for a while (in UFSP) and much preferred the change when I went to Clarity Caps. One might use a word like "neutral" for the Clarity caps... but that's just my cup of tea.
Wav 1 and 2 are very similar, not sure I have a preference. I'm hearing what I could call distortion (not in a good way) in Wav 3.
I love this stuff.
Anyway given all that... try something from Clarity, I had Mundorf Supreme in there for a while (in UFSP) and much preferred the change when I went to Clarity Caps. One might use a word like "neutral" for the Clarity caps... but that's just my cup of tea.
Wav 1 and 2 are very similar, not sure I have a preference. I'm hearing what I could call distortion (not in a good way) in Wav 3.
I love this stuff.
Last edited:
In UFSP BOM - R8 and R11 "depends on idss of Q4 and Q5" see sheet. Not sure what sheet?
In FSP build guide - Q4 idss 8.2 - 8.5, Q5 idss 7.5 - 8.5, Q6 idss 9 - 10: I can easily find these values from my LSK170B supply. If I use these values(tested with DCA75) do I stick with R8=43 ohms and R11=1.5kOhms?
In FSP build guide - Q4 idss 8.2 - 8.5, Q5 idss 7.5 - 8.5, Q6 idss 9 - 10: I can easily find these values from my LSK170B supply. If I use these values(tested with DCA75) do I stick with R8=43 ohms and R11=1.5kOhms?
About R8 R11 mentioned "variable" in the BOM and to look them up, possibly means in a 2SK170 samples availability spreadsheet of what Tea sorts with what resistors. I instructed him a jig at a point on how to measure K170s R8 R11 combos for equivalent in circuit Id/gain to the prototype's listed values.
So not to stop the project due to lack of typical Idss values until all useful JFETs are depleted.
DCA75 tests at low VDS, it will show correct matching but not full Idss, still good for pre-sorting. Also to match trace curves. But perform a final check with 9-10V and a DMM in known simple test circuits for JFETs. There you will see their full Idss. It's important for best channels matching in this phono because of high gain and no loop feedback network to fix it in dB. Like vacuum tubes sorting for mutual conductance on a TV-7 or similar tester. No NFB circuits accuracy depends on their intrinsic gain.
Since you have samples falling in the prototype's values listing, use those. Q6 can have even nearer Idss to Q5 than mentioned, but not lower.
So not to stop the project due to lack of typical Idss values until all useful JFETs are depleted.
DCA75 tests at low VDS, it will show correct matching but not full Idss, still good for pre-sorting. Also to match trace curves. But perform a final check with 9-10V and a DMM in known simple test circuits for JFETs. There you will see their full Idss. It's important for best channels matching in this phono because of high gain and no loop feedback network to fix it in dB. Like vacuum tubes sorting for mutual conductance on a TV-7 or similar tester. No NFB circuits accuracy depends on their intrinsic gain.
Since you have samples falling in the prototype's values listing, use those. Q6 can have even nearer Idss to Q5 than mentioned, but not lower.