"Too much gain"? (Cary SLP05 question)

Hello all,

I thought it would be worthwhile to add to this thread as the unit does have a ton of gain that will severely limit use of the main volume pot. After reading through this thread and others, there has been some good information and some incorrect conclusions. I thought I would relay my experiences here to help future owners resolve the question of what to do with the excessive gain in a pre that they love.

First, let’s be clear what the circuitry is. ALL inputs, xlr (via input tube buffer) or rca get routed to the right and left input trim pots. The pots are a cheap Honeywell plastic potentiometer, 100kOhms. Recommended use for this pot from the Honeywell sheet is in ATVs, joystick controls, and more. Ruggedized for military standards. These Honeywell pots are directly wired to the two plastic potentiometer layers of the Alps 100 kOhm pots, which has a motor control. So you can use either the trim pots or the volume knob to control volume, or both. It doesn’t matter electrically, you just have 2 pots in series.

I have played around with the pots on the Cary, as well as digital attenuation. Re digital, some software driven attenuators are definitely better than others, but it seems like -10- to -15db is the limit where I don’t hear issues in volume matched back to back comparisons. Yeah, I know it shouldn’t be the case mathematically, but something is off. Besides, even -30 or -40 db digital attenuation isn’t enough to give you full use of the volume knob. So that isn’t the answer.

Playing with the pots is interesting, because you will quickly notice with volume matched testing that the balance trim pots are clearly filtering the signals. This shows up with low level information at all frequencies, but it is most easy to see with higher frequencies. Maybe a system runs a bit bright? Well ok, but the full spectrum loss of clarity and detail is not a favorable trade. Buy tubes that are rolled off on top, *cough* I meant warm sounding.

Playing with the Alps volume pot gives similar results, but with less severity. The channel imbalance near max attenuation is a problem though, as is the high frequency filtering which makes the bass seem exaggerated.

So what are the solutions? I guess you have to decide what you want out of this pre. Do you use the headphones a lot? The input trim will likely be important unless your tubes are perfectly balanced. Input trims are useful for troubleshooting, but that is limited case scenario when it is also easy to just pull a cable. Only use the volume knob and you just want it to work better?

If you want it all I would look at replacing the volume and the trim pots with high quality stepped attenuators. Goldpoint,Khozmo, or others are good examples. Yes they are expensive. But your $6-8k amp deserves better than 3$ and 12$ pots. I put in a 24 step mono 100k pots in place of the trim as I don’t use headphones. If you do, maybe see if 47 step will fit.. First off, night and day improvement in clarity at all frequencies, you will sit up in your chair. Second, with high quality pots you will notice that the Alps isn’t exactly bad. If you do most of the attenuation with your stepped pots, then you can run the Alps from 12o’clock to full on with good results. In messing with this configuration over the last month, I would say that the less the Alps is in the chain, the clarity improves but with diminishing returns. My next step will be to replace the Alps with a 47 step attenuator and call it good. The upper range of the 24 stepped input trim is just a teensy bit limiting in my system.

Motorized volume knob mandatory? Well there are solutions depending on the attenuator you choose, but they are expensive. I would go this route if you plan to keep the pre for as long as you can. I haven’t decided what to do yet, I might see if I can use the existing RC design and fabricate an adaptor for the new attenuator. We’ll see.

Hope this helps. Yes, get a professional to do the work and the usual disclaimers. I have no knowledge of EE or circuit designs, these conclusions are my own experience and you may get different results. Make modifications at your own risk.

I haven’t tried the Rothwells, so I cannot comment on those directly.

In regards to attenuators, there are some things that should definitely be pointed out.  First off, any attenuator that isn’t matched to output and impedance runs the risk of increasing roll off in the upper or lower frequencies.  Because attenuation works by dumping voltage to ground, and allowing the remaining voltage to pass to the input circuitry, current is also passing to ground.  I tried some basic 600ohm T style first because the case was reusable and I could always try different topologies, buy as many resistors as I wanted to find the sweet spot.  The 600 ohms dumped so much current the tubes were unable to drive any low frequencies.  Keep in mind that the 400 ohm output impedance of the Cary rises to 4000+ as the signal drops to ~20 Hz, as measured by JA in the Stereophile review.  This unit runs out of juice quickly unless you have a high input impedance amp.  50k+ would be my recommendation.   Note that OTHER highly regarded tube preamp makers state they can increase the cost with beefier output capacitors to match equipment.

I was able to find a resistor match that did measure tonally the same. However, and these are my listening impressions only and should not be regarded as a claim to anything but, I found after considerable testing that the XLR inputs and outputs of the Cary sound different (edit: compared to RCA).  With 6 identical tubes, the XLR pathway loses warmth, bloom, depth, microdynamics and more.  Some say adjust with the input trim.  That is BS, they are directly wired to the Alps pot.  Putting 2 cheap plastic trim pots in series serves no purpose except to allow balancing for mismatched tubes in the headphone stage where it will be most noticeable.   Regardless, both the input and output XLR circuitry both have the same character (edit, compared to RCA).

I did ask Cary about this issue and if they would recommend a local service place to diagnose the issue, and they didn’t reply.  Commonly held belief is that properly made RCA and XLR circuits should sound identical.  Indeed, in my other gear they do.  

Having said all that, running RCA with only 4 tubes (leaving 4 empty), replacing the Alps pot with Kozmo, Goldpoint, or other high quality attentuators and removing the input trim pots will leave you with the potential for glorious sound, as long as you buy exceptional tubes.  I did the work myself, but it will cost you around $1k if you get a tech to do it.  

The gain issue should not be ignored.  Running my XLR 4V outputs through the Cary into a Pass amp (+26 db) required the attenuators to dump 99% of the voltage to ground after the input tubes going by resistance level and L circuit configuration. Many users note minimal volume control issues.    Note also the alps is also the most susceptible to internal RMF signal interference at the bottom and top 1/4 of its range.  Note that with output relays turned off you can clearly hear an input signal at low and high volume, but not at 12o’clock.  So you have to be able to run the unit at 11-2 ish to get best sound.  Gain matching and impedance matching are critical to getting great sound with this unit.  YMMV, and others have reported great satisfaction.  

Cheers

Edited to remove ambiguity, and noted as edits

@immatthewj   Hi there!  Good point for the discussion, no mine does not.  I bought it 2014 as well, it probably has around 7000 hours on it.  

The upgrade package reportedly has a complete capacitor overhaul, which could affect the SQ in a number of ways (especially driving 20k input impedance).  However, I have not read anyone doing a direct comparison between the two, which is not surprising.

Also, and this is just a giant guess as I haven't created a circuit design and asked an EE about it, but if the ultimate upgrade improves the capacitors, AND the circuitry that converts the balanced signals to unbalanced (and vice versa for output) then it absolutely should be better in technical performance and adding input tubes should add tube character. 

One of the reasons why I replaced the input trim pots with Goldpoint attenuators is so that I could figure out some issues I discovered with tonal shifts during digital vs analog attenuation.  Long story short, both the cheap honeywell pots and the Alps act as low pass filters when attenuation is nearing the max.  IE, they remove high frequencies.  Thus sound is perceived to be rich in mids with fat bass.  Using XLR and low volume controls increases this effect.  It can be very pleasing and rich!  However, it comes at the expense of clarity and dynamics.  Listening to a bunch of systems over the last couple years with tubes left me unsatisfied with mine, so I tinkered.  The Goldpoints allow tons of clarity, richness, detail, space and everything else through.  I keep the Alps at max so it is out of the circuit for music, it only gets used now for TV volume.

Having said all that, nothing needs to be done when you love the sound, XLR or RCA.  New purchasers should definitely try the unit before buying or with a full return option.   It is an interesting box that needs the right situation to sound at its best.  

@immatthewj   It sounds like we have essentially the exact same unit (depressions and all), excepting the ultimate upgrade.  I am curious, but the cost is definitely a factor.

The mix and matching is truly the appeal, with the variety 6sn7s out there.  If I dream and ignore consequences, it would be fun to remove the headphone stage, drill another pair of holes, then be able to choose from 3 input tube stages!  Crazy talk... 

Yes you have it, you could replace the pots with quality stepped attenuators.  I had to enlarge the holes a bit to fit the goldpoint 24 steps.  I thought about doing 47 but size was scary close, and digital attenuation below 20dB is totally fine if you want to go that way.  The main benefit was the improvement in clarity.  Running the alps at 12-2 o'clock wasn't bad, and if I did volume matched comparisons no Alps was better, but I will admit it was getting close enough that blind testing might prove me wrong.   But yes, I run digital -6db to -25db for music, the girls do the remote control Alps volume for TV roughly at 12 o'clock.  Input pots never change

I'm currently smitten with the Linlai e6SN7 tubes as a quad. 

@avanti1960 Now that you mention it, I should go back and see if the rectifier tube actually does anything for me.   I swapped in a mullard copper edge from Vintage Tube services, but didn't notice anything different at the time.  The time to swap just makes anything short of night and day really difficult to see.  Sounds like fun!

Glad everyone is enjoying happy listening though.

@immatthewj  Good advice on the 5AR4s.  As long as the power rectifier circuit was well designed any change of tube should have a minimal effect.  At that point it should be noted that variations in tube production over the years might also have just as significant an effect, positive or negative.  My Sovtek had ~6000+ hours on it, figured getting a Mullard to try couldn't hurt. I will likely buy a backup 5AR4 from new production, unless the Mullard clearly shows it is better.

Yeah, the holes are a bit of an issue.  If I did it again I would get a step bit or reamer and practice on some scrap first to make sure it remains perfectly circular and centered.  I suppose I could have made a jig to get it on the drill press without total disassembly too.  Meh.  I had to do some hand filing in the end to get it where I wanted.  The soldering is super easy, and if you remove the mounts temporarily on the headphone stage there is enough room to do all the work.

@testpilot I totally agree.  The honeywell trim pots are a couple dollars each and are made for ATV controls, among other gross implementations.  The Alps you could buy for under $25 with the remote motor.  There is a ton of room inside the case for upgrades.  It should be an option to go stepped or switched relay.

@jaybe  they might have changed the LR balance pots, mine clearly were labelled Honeywell and were designed for ATVs and crude motorized controls.  Re the Alps, I doubt it looking at the Alps component catalog.  If they still had the premium $100 pot from 20 years ago I would have bought that in a flash.   Cary's sales volume, and anybody else for that matter, is way too small to justify custom design and production.  

Aric can also fine tune the output capacitor to match amps.  Carys ~4400 ohm output presents rolloff in low frequecies regardless.  

Check remote function.  In my unit, it controls volume and nothing else.  The buttons are for the cinema unit I think.

Def have a return option, buying new unheard is a big risk.  Used market ok but reselling these units can take over a year unless you want to burn some cash.  Buying hifi is getting harder as dealers close shop.

Dang, typed a bunch of stuff and the site freaked out!  Not going to spend more time trying to deal with the picture upload (archaic), but basically I get ~2/16th of a rotation for XLR from 0 to 68 dB at 15', and a a bit more than 3/16th for RCA.

If I want to use volume on the Cary with XLR I set -46dB on the goldpoint pots (dump 99.5% of signal to ground - linear??) and it sits around 2pm, so 1/2 a rotation.  I need a scope to see what is really going on, gut says that can't be linear but only measurements can be trusted here.