My experience with long ICs


I posted some questions a while ago about needing to move to a long IC between my preamp and amp (a Bedini 866 pre and Bedini 803 amp). I wanted something sonically along the lines of my current (short) Kimber KCAG, but was not willing to spend a bazillion dollars for it. The new IC needs to be 25 feet long.

After reading responses to my other posts, and checking here and the Asylum, I decided upon the following cables:

Signal Cable Silver Resolution
Guerilla Audio Silver
Oritek X-1

I picked these three because they all offered a return policy on long cable lengths (something to watch out for - not all manufacturers who offer a return policy will do so for non-standard lengths), they were all reasonably priced for the long lengths, and they all indicated that they do offer long length ICs and have sold them in long lengths. Some manufacturers indicated that their cables are not meant to be built to long lengths.

Now, all of these cables were unbalanced RCA, The Oritek and Guerilla are unshielded while the Signal Cable is shielded.

My preamp uses an emitter-follower output stage that is capable of driving up to 2uF. The 803 amp is a sliding bias Class A design, with a pure Class A input stage. The circuit design of the amp has no built in bandwidth limiting.

Some of you may be able to guess what happened...

I burned in all the cables (except the Oritek, they are burned in by Ori) using a varying FM noise source into switched 20KOhm and 1KOhm resistive loads for at least 80 hours (I built a small test fixture for this). The output of the noise source did not exceed 2 VPP.

I tried the Signal Cable first after listening to some music with my existing setup. I immediately was impressed with the cable. Not as brittle as my KCAG in the upper vocal area, and more power in the bass. Maybe not as quick in the bass, but I wasn't sure because after about 5 minutes of listening the internal fuses in my amp blew. I replaced the fuses and tried again. This time I was not quick enough; the fuses blew after about 5 minutes of rather nice sounding music, but I did not turn off the amp in time. One of the channels blew out.

Upset, I called John Bedini (he still repairs these old beauties) and sent the amp back for repair. He was kind enough to turn it around in no time; three output transistors and a pull up resistor in fact did fry.

I got my amp back in four days. This time I decided to try the Guerilla Silvers. I first checked the amp with the old KCAGs; sounded just like before except in need of some break in time. I hooked up the Guerilla cables, my finger on the power amp switch. In this case the fuses blew immediately. I turned off the amp about 2 milliseconds later. Replaced the fuses, plugged my old KCAGs back in, everything worked like a charm.

I called John Bedini and spoke with him about this. He had previously told me that long ICs could be used with the 803. When I called him back and happened to mention that some of the cables I was testing were unshielded he stopped me dead in my tracks. He said the 803 would in fact react exactly as I experienced with unshielded cables. He figured I knew enough not to try something like that. Something about noise pickup, RF, and the input stage going into oscillation. I thought oscillation was a byproduct only of high reactance. He indicated that the equivalent of a Zobel network at the input of the amp would not prevent this problem, as it was not being caused by an impedance type situation.

We're not quite sure why the Signal Cable Silvers - a shielded design - had the same effect, although after some minutes of playing time. Maybe over that length the shield does not provide a level of rejection that the 803 can deal with. That's just speculation.

The bottom line is the I am advised to use an "internal conductors" style cable (i.e. twinaxial) with a high quality, high coverage overall shield. Note that I am not in any way indicating that the cables I tried have something wrong with them; the Signals, in fact, were beginning to sound to me like something I would like very much. I wish I would have had a chance to listen to the Guerilla. Right now I'm afraid to try the Oritek; Ori assures me the design of the cable - while unshielded - provides superlative RF rejection and that the cable has about the lowest capacitance per foot (~10 pF).

So what do I do? I'm still a believer in wanting to use short speaker cables in the tradeoff, but that means that I now have to look at these long ICs. Are there value priced ICs that are comparable to the ones I've already looked at that offer a high quality shielded design and still sound open, quick, and extended?
tonyptony
Tonyptony
Did you contact Frank at Signal, and tell him about your experience? Maybe he can do something custom for you. Just a thought.
If you have a ground loop issue, I would add a by-pass capacitor from the female RCA ground on the preamp to the chasis of the preamp amp. This can help isolate the ground potential difference between your preamp and amp. Before you blew your amp, you should have heard a lot of distortion picked up by the cable.
In any case, I would only use balanced cable for such a long run.
Too long for an RCA, IMO. Most audiophile shielded single-ended cables have a floating shield and with this length you can see hf noise. Balanced XLRs is the best solution in this scenario or use a closed shield RCA (which may sound bright).
This sounds odd, I have 21' kimber hero innerconnects, rca going froma ar pre amp to my quicksilvers v4. Had ground loop issues with the long innerconnects so I put cheeter plugs on the amps and it was solved. Since then I have gone balanced with same amps and hero innerconnects and a clyspo pre amp with no problems. I often questioned if that was to long for unbalanced and most people said no, but balanced was better. I have my wires run behind my crt 49" TV in a mess of wires and no noise. Maybe try the hero if you like the kcag innerconnects. I tend to think your issue is a ground issue, maybe try taking the grounds off everything. It's odd a amp will blow because of distortion espesally at a low level.
Oldpet, Frank and I emailed quite a bit about this situation. At that point we were still scratching our heads about why this was happening with a shielded design, but he did not make any recommendations for any custom solution. Maybe I'll drop him another line.
S23, this sounds like an interesting idea. Any thoughts on the value for the cap? And I assume the RCA jack on the preamp would have to be isolated from the chassis (if it isn't already) in order for this to work.
rch10, I know this is long but it's been done. In fact, back in the early 90s when I was experimenting with building interconnects and speaker cable, I built an 18 foot pair of ICs using a Belden twinaxial cable in a single ended shield design. Woked fine with the same amp, but I was not crazy about the sonic qualities. Others here and on the Asylum have used long unbalanced ICs successfully. Yes, balanced would be better but I don't have that option.

By closed shield do you mean terminated at both ends? If so then, yes, I was wondering if I had no choice but to go that route.
Hmm, that's another note about a potential ground issue. I am taking this seriously, programmergeek, but my quetion is then why does this not happen with the shorter ICs?
Hi Tonyptony,
50 uf /10v+ axial load type should do the trick.
You can get them from digikey. They're quite small like a resistor size.
Used to have this issue with Counterpoint Preamp. This fixed was suggested by Mike Elliot and it did the trick.
I've applied the same cap in other noisy interference preamp and it did worked well.

Good luck
"Used to have this issue with Counterpoint Preamp"

S23, are you saying that you had your amplifier blow up like mine did?
I bought long IC's from Oritek. It was 4 or 5 meters, I forget. Ori modified the design of the cable, so that it would sound and function the same as the one meter. IMHO, it did. I used it for quite a while on a Cary SLI-80 and was very happy with it.
Art