Cables & Low Capacitance

[dlaloum]

Hi Folks

New here.... I registered specifically due to this cable problem I have which although not a Quad issue, is probably something that many of you have had to handle and overcome to get your Quad systems working properly.

I just got back into Vinyl, and am in the process of configuring and tuning my Turntables...

My interest is primarily (for the time being) in Moving Magnet and Moving Iron cartridges, of which I have snapped up a reasonable collection used..... and purchased top flight (shibata, line contact, micro-ridge/SAS) styli for. (where possible on a light weight high quality cantilever)

That was the "easy" part - starting setting things up and measuring things - pink noise track for Frequency response, wired up a series of loading plugs so I could vary the impedance loading to anything I want (I currently have made plugs from 10k to 100k ohm).

The next issue was cable capacitance .... and many of the best MM/MI cartridges come from the Quad period (and were designed for Quad!) - and they all require a total capacitance of 100pf.

This sounds simple - so I buy myself a new Multimeter with capacitance capability (only $30!) ...

I take the cartridge(s) off the tables and measure capacitance of arm to rear sockets...

JVC QL-Y5F 60pf
Revox B795 30pf

Then I measure the cables:

Amber IC 70cm - 420pf
Jaycar STP 50cm -170pf
BlueJeans LC-1 130cm - 90pf


I have some other cables - but this shows you my quandary...

How am I to achieve less than 100pf total capacitance when a low C cable + turntable internals already adds up to 120pf MINIMUM

To add to the difficulty - my loading is done manually using RCA double adapters (and rca plugs with resistors soldered into them)- which add around 15pf, and I am currently using a Mic Pre / ADC with digital RIAA - which requires 1/4" (TR) jack input - so those adapters are another 15pf... so another 30pf overhead!

Looking at the LC-1 low capacitance cable, it is specced at 12.2pf/ft for the cable alone - yet the completed cable measures at 87pf.... leaving 34pf to be accounted for by the RCA connectors.

If I was to make a 30cm LC-1 cable....

Cable - 18pf
Cable RCA's - 34pf
Turntable (best) - 30pf
plugs/adapters - 30pf

Total pf = 112pf

I'm struggling here!

What is the secret of achieving an effective 100pf setup with a turntable... (short of hardwiring everything end to end.... which makes it impossible to experiment to tune the system...)


The input of experienced MM/MI quaddies eagerly sought...

thanks

David

(The other distinct possibility, is that I am hitting a wall which contributed to CD4's demise, and the rise of the "Dominion of MC's"....)

 

[oaqm]

Some reading that may be relevant to your issues:
http://www.hagtech.com/loading.html

You may get some insightful help on the Audio Asylum, which is indeed aptly named, but there are some folks there who may well have gone down the road you are seeking to travel.
the vinyl forum:
http://www.audioasylum.com/cgi/etv.mpl?forum=vinyl
The cable forum:
http://www.audioasylum.com/cgi/etv.mpl?forum=cables

 

[dlaloum]

Thanks for the references...

I have used the Hagtech calculator as a starting point.... and it may be my end point too once I have finished measuring and testing...

The Hagtech suggested Impedance load tends to result in a high mid slump in response (not huge.... a couple of db)

Dropping the capacitance (147pf) and increasing the impedance to 62k + is showing flat F/R through to about 6khz then rising gently to a peak (+4 to +6db depending on impedance) by 14khz where it stays till 20khz and then plummets down...

Reducing capacitance seems to shift the start of that rise back, for 100k impedance and 500pf the rise begins at 5khz - for 147pf it begins at 6.5khz.

Clearly the end result will be a balancing act between mid slump and high peak - flat f/r seems unlikely - with this cartridge (Shure 1000e with SAS stylus).

Trouble is I am starting to hit the wall with regards to lowering C any lower!

I havn't looked on audio asylum ... will have a browse .... I am regularly on Audiokarma and VinylEngine.

bye for now

David

 

[JonUrban]

Would like to help but it's way out of my league. Good luck in your quest.

 

[oaqm]

The Hagtech suggested Impedance load tends to result in a high mid slump in response (not huge.... a couple of db)

That and a parametric EQ might be your answer. Narrowing the Q at the frequency of the slump and bumping accordingly just might yield sonic nirvana.

 

[dlaloum]

Yep PEQ is something I'm planning for, but the first step is to minimise all possible processing.... which means working through the measurements until I have the best possible base result...

Also if I can get it naturally flat across the entire midrange, then I will only need 1 filter for the LF "hump" (there's a reason Shure's have a reputation for warmth), and a second one for the HF peak.... (the HF peak may require more than 1 filter as it is asymetrical)

But if I choose the wrong setting, then as a minimum I would also need to boost the midrange....

I also need to consider whether I want to boost the lower bass region as there is a roll off of a couple of db below 50Hz - again some settings reduce the rolloff....

Each filter added is another bit of processing with associated phase/time issues, distortion etc...

I also noted that looking at 1980's and 1970's spec sheets, the frequency response graphs are often shown at such a low level of resolution that +/- 2db still looks flat.... and that many of the cartridges were specced to -3db points with "effective frequency" of 20-20k. (what is effective I wonder... -6db? -10db?)

This journey started when I recorded several different cartridges, and carefully level matched them to within 0.1 db - most of the differences between them disappeared.... well I initially thought so, careful listening showed the differences were still there but substantially reduced...

So
1) normal acoustic level matching will never cut the mustard as resolution of 0.1db or better is not achievable realistically - so it has to be done in a digital recording
2) - Frequency domain performance of cartridges varies.... So I started to work on measuring and adjusting the loading individually by cartridge.... my gut feeling is that once this is done, another layer of "difference" will disappear.
3) Use a touch of EQ to flatten each cartridge out individually (and balance left/right channels)

I'm willing to lay odds that at that stage all the cartridges will sound the same in terms of tone, timbre, and it will be down to the fine resolving capabilities of each cartridge.... similar to comparing digital gear. Air - HF boost - will be gone, warmth - LF boost - also history.... just getting closer to the original recording...

But to achieve that I really need to get my capacitance down to 100pf.... given that this was a spec commonly used for a lot of cartridges in the 70's (Quad era) and that a lot of these were then sold as stereo cartridges with little or no modification.

 

[oaqm]

"Back in the day", ±3dB was the operative definition of "flat".

"Back in the day" referring to mid to late 60s through.... whenever.

I think ±3dB is still pretty much used as the benchmark for "flat" speaker frequency response. This is from memory, but I seem to recall that a "normal" human can't really perceive a difference in volume until somewhere around ±2 or 3 dB, hence the usage of the term "flat".

 

[dlaloum]

It takes about 2-3 db to percieve an increase or decrease in volume....

But we can discern differences as small as 0.3db or less - we just discern them as "air" (HF), "warmth" low frequency, etc.... we don't perceive them as changes in volume.

This is exactly what started me on this exploration - when I level matched recordings from different cartridges, and set their Average level identical to within 0.01db (each channel seperately to compensate for imbalances in the cartridge and amplification chain.

For some of these recordings they were previously only 0.5 to 1 db apart in volume - yet they sounded different.... but differences were hard to quantify... it was more warmth, more detail in the lower registers, sweeter high end.... very intangible.

Once I level set them all - they became very very similar... fascinating how the brain/ear works isn't it!!!
Also to ensure a fair method of listening I loaded all the recordings into a multitrack software - and time aligned them precisely - I could then switch interactively between the tracks in real time - very very hard to differentiate them that way. - but they still have differences in frequency response that give them away.

Which effectively means that 75% of perceived differences in components (ANY components) are probably down to level setting.... probably the next 15% of the differences are variations in frequency response..... then 5% would be channel balance and seperation....

(I am assuming that the starting point already has reasonable performance and relatively low THD and IMD....)

bye for now

David

 

[RSilverman]

Been reading the threads about low-capacitance cables. I'm thinking about re-wiring my Philips 212 to improve CD4 playback. On some albums, I get perfect quad, but on most there's so much "sandpaper" distortion as to make the experience cringe inducing. I'm hoping that better cables might help. The RCA cables on my turntable are not original - so even if Philips used low-cap cables, the ones I have may not be.

I can't seem to find anyone who says, "buy this cable and it will work." But I saw that someone recommended Blue Jeans cables, and they're not too expensive for a 5ft run (36 bucks). Can someone verify that this is a good option?

http://www.bluejeanscable.com/store/audio/index.htm

The Cable: Blue Jeans Cable LC-1 Low Capacitance Audio Cable

The most important attributes of a line-level unbalanced audio cable are (1) shielding, and (2) capacitance. Heavy shielding protects audio signals from interference from outside sources. LC-1 Audio Cable uses a heavy double-braid shield, with one bare copper braid laid directly over another for extreme high coverage and high conductivity to ground; this is the identical shield configuration to Canare LV-77S, which tested best in our review of audio cable hum rejection characteristics (LC-1 hadn't been designed yet so wasn't tested at that time). By shrinking the center conductor to 25 AWG and foaming the polyethylene dielectric, we were able to get capacitance down to an extremely low 12.2 pF/ft, much better than LV-77S at 21 pF/ft. Capacitance can be important, particularly in long cable runs, because it contributes to rolloff of higher frequencies. The softer dielectric material and smaller center conductor, meanwhile, make the cable highly flexible and easy to route. LC-1 is built exclusively for Blue Jeans Cable by Belden, the leader in American communications cable, and is rated CM for in-wall installation in residential and commercial environments. For more information and specs on LC-1, read our "LC-1 Design Notes" article.

 

[RSilverman]

Hey, well whaddya know?!? I just took a flashlight to the cables from my turntable, and it says right on them, "Teac professional low capacitance cables," so my sandpaper issue isn't to do with the cables. Maybe I'd be better off buying a demodulator where I can adjust the carrier signal -- I can't do that on the built-in one on my receiver. or even trying an AT440mla instead of the AT12S I'm using now. All I know is, sandpaper quad sucks big time.

 

[scifi]

https://www.quadraphonicquad.com/fo...th-CD-4-set-up&p=158903&highlight=#post158903

 

[RSilverman]

Thanks. I always clean the record and stylus before trying CD4. Either these records are in poor shape - though some look "new" and still give me sandpaper sound - or my cart isn't tracking right for some reason. I have a new shibata stylus on an old AT12s cartridge body that looks kinda rusty. I see I can still buy an Ed Saunders cart for 140 bucks, but will it make a difference or am I throwing good money after bad?