Cary Cad-300b Se

[kmtang]

Hi,

I got a pair of CARY CAD-300B SE monos from my friend. I tried it out in my home. I noticed that the mid and high are pretty good. Unfortunately it does not have sufficient bass.

I checked the circuitry. It employs 6SL7 SRPP as driver. The B+ for 6SL7 is 450Vdc. The cathode resistors are 4.75K. The bottom cathode resistor is bypassed with a 500uF 25V electrolytic capacitor and a 2.2uF Solen film cap.

The coupling capacitor between the 6SL7 and 300B is a 0.22uF vitamin Q like capacitor. The 300B is with fixed biasing and the B+ is 440Vdc. It is set to 60mA bias current.

I am wondering if any one have worked on this Cary 300B and any experience in improving the Bass of it.


JOhnny

 

[commstech]

60mA sounds like a very conservative bias current. i shoot for at least 70-75mA.

 

[kmtang]

The B+ is about 440Vdc with fixed biasing. Therefore, I am afraid to run the 300B to its maximum power limit.

Anyway, I placed a 0.22uF Jensen copper-in-oil in parallel with the Vitamin Q like coupling capacitor in the amplifier this morning. Both dynamic and bass were improved.

One other thing I found is that the B+ filtering is:-

5U4GB - 1000uF - 20-ohm - 1000uF.

The first filter capacitor is way way higher than the specification of 5U4GB. I just put a diry chip 5U4GB that I got from swap meet at $2 each. I saw sparks came off when first power on the HT.

I have also measured the primary inductance of the OPT which is 24H according to my DMM. May be this is one of the reasons why the bass is not powerful enough.


Johnny

 

[commstech]

5U4GB - 1000uF - 20-ohm - 1000uF.

The first filter capacitor is way way higher than the specification of 5U4GB. I just put a diry chip 5U4GB that I got from swap meet at $2 each. I saw sparks came off when first power on the HT.

1000uf?

its crazy....
the 20ohm is the DCR of the supply choke? or a power resistor?

 

[Sherman]

As Commstech says 1000uF after a 5U4G is way too much! That rectifier is rated for 40uF when used with a cap input filter. I find it hard to believe that it isn't arcing and self-destructing with 1000uF.

 

[sgerus]

It employs 6SL7 SRPP as driver. The B+ for 6SL7 is 450Vdc. The cathode resistors are 4.75K

Hi,

Are you shue that's 4.75K? 475R would sound more like it.
What is the voltage at the cathode of the bottom 6sl7

-Scott

 

[kevinkr]

440V fixed bias with any 300B is pushing your luck, the JJ and WE 300B can handle these voltages, but just a few years back none of the Chinese made ones could.

Unless you can get these for a very good price you can build a better amplifier, and probably for less money.

Cary has a long history of to put it nicely "questionable" design practices.

I have used 6SL7 SRPP to drive 45 and 2A3 since the 1990's - it does a great job with these tubes, but the 200Vpp swings required to drive a 300B at these plate voltages seems imo to be asking a bit much.

1000uF input capacitance is unconscionable with a 5U4, some of them will arc with this value of capacitance, and all of them will have considerably shortened service lives due to the very high repetitive charging currents, not to mention the inrush current as the rectifier warms up.

24H is not an unreasonable primary inductance for a 300B provided that value is maintained at the design plate current. This should give you flat response down to somewhere around 30Hz ime.. (Tamura F-5002 has similar primary Z.) Note also that your meter probably is not that accurate, and the inductance varies as a function of standing current and the ac signal current superimposed on it - in operation the inductance may be higher or lower than you measured.

Given the way Cary designs things I would not assume that the OPT has a lot of current headroom built into the design so I would not change the op point. (Verify that the output transformer is not saturating at full power when you do. If overall power increases and THD at the previous highest output power decreases them you have a little margin to play.)

Not my place to say, but were it my amplifier I would reduce the input capacitance to 50uF or less, (check winding dcr if less than 220 ohm end to end use <40uF cap) switch to a 5AR4 rectifier, install a nice hefty 10H choke with a dcr suitable to drop the plate voltage to 400V with a 60 - 70mA Q point, change the second filter cap to a 100uF PPIO motor run cap and I would boost the current in the 6SL7 SRPP to 2mA, and install a better and larger coupling cap.

 

[kevinkr]

Hi,

Are you shue that's 4.75K? 475R would sound more like it.
What is the voltage at the cathode of the bottom 6sl7

-Scott

Actually I use 6SL7 in SRPP at 330V (I'm pushing it) - with 1K cathode resistors the q point is 2mA. I suspect the idle current here is under 1mA, and frankly the tube is running way beyond its safe voltage ratings for both max plate voltage, and cathode/filament insulation breakdown rating. Another fine example of the Cary approach - no respect for design maximums.

6SL7 specs here: www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6SL7GT.pdf

 

[sgerus]

Kevin,

I was thinking that each tube would have its grids, plates and cathodes connected in parallel, so you could get more like 4mA through the srpp. And, the upper valve could have its own filament supply if needed, based on the voltage of the upper cathode.

In other words this is a 2 tube SRPP with each tube connect in parallel?

 

[kevinkr]

Kevin,

I was thinking that each tube would have its grids, plates and cathodes connected in parallel, so you could get more like 4mA through the srpp. And, the upper valve could have its own filament supply if needed, based on the voltage of the upper cathode.

In other words this is a 2 tube SRPP with each tube connect in parallel?

Hi Sqerus,
Been there, done that - it does address the fundamental issues you raise, but frankly a 6SL7 or 6J5 driving a 6SN7 or 6J5 based SRPP will almost certainly sound better IME.. Run the 6SN7 at 10mA or so.. Because the 6SN7 is a medium mu tube another gain stage is required as noted above. Another option D3A or 5842 driving a transformer.

 

[sgerus]

So if we run the 6SN7 at 10mA we no longer have an SRPP. Are you talking about using the 6SN7 as a cathode follower?

 

[kevinkr]

No, I'm talking about 6SN7 in SRPP at 10mA. (400Vp at your risk, 820 ohm cathode resistors.)

 

[kmtang]

Thank you guys for your valuable responses.

The Cary CAD-300B SE employ C-R-C filtering for the B+. It is a 20-ohm power resistor.

The cathode voltage of the 6SL7k is approx 2.9Vdc.



Johnny

 

[kevinkr]

Thank you guys for your valuable responses.

The Cary CAD-300B SE employ C-R-C filtering for the B+. It is a 20-ohm power resistor.

The cathode voltage of the 6SL7k is approx 2.9Vdc.



Johnny

Hi Johnny,
Based on your measurements this is about 0.6mA - far from what I would consider optimum even for the far less demanding task of driving a 2A3 or 45. (Try 1.5K resistors instead and see what that gets you.) Eli and others have raised valid objections to the SRPP topology, but I will admit to using it a lot and finding it to have more than adequate performance in the applications that I use it in. I have used it for about 20yrs now, far before it became popular, and I find I like its distortion signature in some applications. (Guess that makes me sick. ) Its relative simplicity is also appealing to those with component count phobia like me. I have done extensive measurements and it works pretty well in most cases, but there are many instances where a CF loaded into a CCS will provide much better performance. Incidentally in all of my SE amplifier designs the output stage is the dominant distortion source, regardless of driver stage topology.

The power supply is just a crummy brute force implementation, the changes I mentioned above will do the job nicely. The idea here I think was to provide good dynamic regulation, but this is a class A amplifier so beyond a certain point it is doubtful that there is any real benefit as long as the supply source impedance is a very small fraction of the reflected load impedance, and ripple is low enough..

YMMV

 

[kmtang]

Hi Kevin,

Thank you for your most valuable inputs. I think it is very easy for me to add two resistors in the 6SL7 driver circuit and see how much improvements it gonna to make.

By the way, my friends are very impressed with my 45 tube amp using Lundahl 1620 SE 30mA OPT. Excellent strong bass and very fine mids and highs.


Johnny

 

[kmtang]

Cary CAD-300B SE Modification

Hi Kevin,

I placed a 1.5K resistor in parallel with the 4.75K cathode resistor of the 6SL7 driver. The cathode voltage is now at 1.8Vdc. I tried it out and found the sound gets more powerful and solid.

Then, I replace the 6SL7 with 6SN7. The cathode voltage reads 6.0Vdc which is reasonably good. It sounds even better than 6SL7 after breaking in for one hour.

With this SRPP configuration, I can use either the 6SL7 and 6SN7. That's really cool.

Unfortunately, the bass is still very weak from this Cary amp. I will try to adjust the 300B bias current later to see if there's any improvement with the bass later.

By the way, I replaced the 5U4GB with 5R4 and the B+ now dropped to 440Vdc for the 300B.


I read the brochure of Cary CAD-300B SE Signature. It employs choke filter for both 300B and 6SL7 driver stages. May be this would help in sound quality.


Johnny

 

[kevinkr]

Hi Johnny,
Can you measure the frequency response of the amplifier into a resistive load down to 20Hz or so?

Other possible causes for weak bass besides insufficient primary inductance could include speaker/amplifier interactions. I don't know if your speakers were designed for amplifiers with a specific source impedance other than 0 (i.e. perfect voltage source approach to design.) I designed my own speakers (Onkens) and they definitely were designed for a specific source impedance - lower source impedances than I deliberately designed for result in an overdamped low end response with much less bass - I'm wondering if this is the case here..

Tell us about your speaker system. If adding say 1 ohm of resistance in series with the amplifier output results in more (not flabbier) bass then it may be the result of the speaker design.

 

[kmtang]

Yes, I did.

It can go down to 20Hz with very little decrease in amplitude. The 20Hz square wave has sharper fallen slope than the Hammond and Lundahl OPT's that I am using.

I am using Tannoy floor stand speaker with one 8" mid and tweeter in same, one 8" woofer, and one passive woofer. This is not a common speaker which is triangular shape looking from top and about 4 ft tall.


Johnny

 

[kevinkr]

Doesn't sound to me like the amplifier is the issue, try adding some series resistance (say 0.5 to 1 ohm) to the output and see what happens. What is the impedance and what is the model? May just be the case that this particular speaker is not the best match for this amplifier.

 

[Alastair E]

Kevinkr I believe has hit on the main issue here....

Its one I have seen a couple of times, That of insufficient Current through the valve.

I have used the 6SL7 many times in different ways and projects, and it always seems to 'Like' around 3-4mA current flow....

To achieve that value, a cathode-resistor of around 220-470 ohms is needed, according to the plate-volts...

Again Kevinkr has spotted huge flaws in this 'Cary' amp, as it is clearly exceeding the Heater-Cathode insulation value specifed for the 6SL7

The lack of current flow through the gain-stage will increase its o/p impedance and give a weak, reedy sound lacking in bass. Some like this type of sound, Not me!

I guess if you sort this issue then great improvements will be had to the Bass and general sound of the amplifier...

Cant quite believe they have a 1000uF cap fed from a 5U4G...!
You gotta do summit about That!

(Although when I was young, we had a Pilot radio. It had a type 80 rectifier, I guess about half the ratings of a 5U4G. Its mains transformer failed, so I replaced it with one from a guitar amp. And a large 800uF cap from an old TV set was added, as the set always suffered from mains-hum-It was used every day for breakfast time...

It still works some 30 years later..The type 80 rectifier glows frighteningly 'blue' inside for a second or two after which it settles down to normal....)-It hasnt been used for the last couple of years though-Guess I'll dig it out one of these days and see how it is