• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Help with Keith Kirby's DIY Tube Preamp with dual 6H9C/6SL7 and one 53YGT/5AR4.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
A good Hello to all:

Note: Don't visit/participate on this site (aside from maybe/seldom Acoustat Answer Man's forum/topic). :ashamed:

Have recently acquired a custom DIY tube preamp on which the cover page of the manual indicates the the builder as the late Keith Kirby. Here's the link to the ad.

Tube Stereo Line Stage Pre-amplifier (Custom Built)

Am reaching out to all, and may be more-so to those who have contacted worked with Keith on/with custom builds. (From what I've been able gather from the Internet prior to his passing (in 2013), he made wonderful/beautiful DIY cases which he sold via his eBay site). I have no clue/idea what the circuit or what circuit this premap was based upon. It utilizes one 53YGT (or 5AR4) and two 6H9C (of 6SL7) tubes. Have not opened it up to check out the internals and what is possible in terms of upgrades/mods. Only have the details from the manual but it doesn't provide a visual cue/clue to the internals and layout. :scratch2:

Also do not know if Keith went by a moniker on this site if he communicated on/within this site.

Given that the preamp appears to be designed for 6H9C/6SL7 tubes (and equivalents), I was wondering if 6SN7s could be utilized and subsequently tried - and yes, it does work and as such is an improvement upon the 6H9C tubes. However, the specs for the the 6H9C/6SL7 are different from that of the 6SN7 tubes and that they are not usually interchangeable (at least not from 6HC9/6SL7 to 6SN7 but possible the other way around). Would/does using the 6SN7 have any electrical impacts, such as, resistive and amperage issues? Would it cause the transformer or any other parts to work harder/overheat? :yikes:

Trying to gather as much detail/info as I can prior to opening it up to take a look further. (Seems too nice a casework to be tampering with it! :cool:)

Thank you all for any assistance you can provide/illuminate on this preamp. :)

Sincerely,
Kingsley.
 
Last edited:
Without a schematic, it's a total crap shoot. However, there is a huge difference between a 5Y3 and a 5AR4. The 5Y3 is good for 125 mA., while the 5AR4 is good for 250 mA. Forward drop in a 5Y3 is large and forward drop in 5AR4 is small. Alone, those facts have me shaking my head. Differences in B+ rail voltage have a substantial effect on circuit behavior.

Are you confident enough in your skills to trace the schematic out?
 
Good Evening, Eli:

Thank you very much for responding! Was hoping that you might be willing to provide some input/help out. :D

I get your point about the tube differences spec-wise. Going from the short manual provided with this preamp, it can take one 5Y3GT/6087 or 5AR4/GZ43 (as, I'm suspecting this being the tube rectification) and two 6H9C/6SL7/VT-229 (as the driver/gain tubes), one for each channel. So, given this, I would presume (maybe correctly or incorrectly/over-presumptuous) the circuit was designed/built with some flexibility in regards to tube types.

Haven't opened it it yet as I was really hoping someone else might have had the same/similar custom DIY preamp built by Kieth or at least commissioned him to do it. Else it could be a one off and the one and only one built for Jim Parsons (as per the manual cover and of whom I haven't the faintest/foggiest idea of who he is).

As for tracing out a schematic - I'm not that good at electronics but when I get around to opening it up (as I am curious and also want to see if there is room for a stepped attenuator such as Kozmo or DACT and maybe replace the selector switch with an Elma), maybe I can start with taking some pics of the internals and go from there.

Thanks, Eli.

Sincerely,
Kingsley.
 
Kingsley,

You have a very nice mechanical engineering package and what seems to be a decent power transformer. That's stuff to work with. Frankly, the 6SL7 is a miserable choice for line stage service. It exhibits high gain and very poor drive capability. Many power amps are sensitive enough to not need line stage gain, when fed from a "standard" 2 VRMS CDP.

BTW, I would not be in a hurry to install a costly, and likely "cold" sounding, stepped attenuator, even if it fits. First, see what's inside the "box". FWIW, my hunch is that a conductive plastic control is present. You can realize a sonic improvement over conductive plastic by installing a Canadian product. Look here.
 
Hi Eli:

Yes, you have a good eye in noticing that the mechanical engineering package, including the vintage transformer, is nice and I would concur. :up:

That's interesting info about the 6SL7. I wonder if that might be the reason the preamp was exhibiting some sound distortion with the 6H9C tubes. After swapping them with 6SN7s, it was across the board difference and only a very slight hint of sound distortion that is barely discernible and not always (could be the recording also). Could be that the 6H9C tubes are/were on their way out too. Also not sure if that barely discernible distortion after swapping to the 6SN7s is a result of the 5Y3GT tube life expiring (no idea how much use/time this tube experienced as it and the 6H9C tubes came with the preamp).

Too bad I didn't see your reply/remarks regarding the 6SL7 tubes as I bought some 6SL7WGT NOS Philips and a couple of VT229s a day or two after I had bought some Tungsol 6SN7GTB black plate chrome tops! :blush:

Also interesting is your remark regarding the cold (and sterile?) artifact as per the noted stepped attenuators. From reading, most make mention of it's superiority to wiper pots in tracking. precision, and sound. (And supposedly LDRs are even better than stepped.) Thanks for the link to PEC pots (Good Tip!), I had searched for them some time ago but wasn't successful and thought they were no longer made/produced - maybe didn't look hard enough or searched incorrectly (for sealed PEC/Pansonic(?) if such exists). Yes, your 'hunch is that a conductive plastic control is present.' is borne out by the fact that it's an Alps Blue Velvet. Don't carbon wear out faster than plastic conductive (and have lesser tracking precision)?

Thanks. :wave2:

Sincerely,
Kingsley.
 
The PEC product is milspec tough and will last a long time. The wiper is a spring loaded, machined, block of carbon. Lesser quality controls use a formed piece of sheet metal that gouges the resistance track.

Vacuum rectifiers have LONG service lives, unless subjected to abuse. Of all the tubes, the 5Y3 is least likely to have failed.

6SL7s are useful, but not in a line stage. Phono preamp and mic preamp are possible uses for the tubes.
 
Here are some pics. (Haven't open it up yet to take some pic of the internals.)

Again, hoping someone may have come across this build (aside from the person who commissioned it as noted on the manual cover page).

Thanks.
Front_view.jpg Back_view.jpg Manual_cover.jpg Manual_page_1.jpg Manual_page_2.jpg Manual_page_3.jpg

Sincerely,
Kingsley.
 
Two 6SL7's are likely to consume 2-5 mA HT total.
Both types of rectifiers are way overkill. At this low current, the voltage drop for both will be low, so there won't be much difference.
The document says that the 5y3 is used (measuring strong), so my suspicion is that it was chosen because he had it at hand.
If the capacitance used is within datasheet limits (or choke input) the rectifier will outlast you.
One thing to check when switching to 6SN7, is the heater current: 6SN7's draw twice as much current.
The circuit for the DC heater supply needs to be able to deliver that without sagging or overheating.
Just open it up (leave it unplugged for a while yet assume there still is voltage inside, don't touch) and post pictures of the inside.
 
Kingsley,

How adventurous are you? There is a 100% safe way to not overtax the heater supply, however it's put together. Use a 6J5 (same triode as in 6SN7s) in each channel and buffer the line O/Ps with a MOSFET, say the IRFBC20. I am assuming a fair amount of B+ current is available, which is not unreasonable, given the 125 mA. capability of the 5Y3.
 
Thank you Parafeed for some additional insight on the characteristics/parameters of the 6SN7 current requirements (which I've briefly read about) and the longevity of the 5Y3GT of the tubes.

Thank you Eli for offering an alternative to reduce harm to the heater supply when using 6SN7s. (Might be a bit over my pay grade as I'm not an electronic tech. I can solder and replace parts (with the sane or slightly higher voltage spec ones) and mostly read a basic schematic circuit diagram but to implement circuit mod without some example/diagram, I'm afraid I'd be like shooting in the dark.

Anyway, opened it up and took some pics (hope they are not too small but still with enough clarity/details). Maybe it will shed some insight for you two and others who may be interested. The work certainly looks quite good and I'm thankful that it's all hard wired and not having circuit boards but it might be a more difficult job for me to solder/desolder without damaging the part/component as the ends are wound/wrapped. :redhot:

IMG_0001_1.jpg IMG_0001.jpg IMG_0002.jpg IMG_0003.jpg

IMG_0004.jpg IMG_0006.jpg IMG_0007.jpg IMG_0009.jpg

IMG_0010.jpg IMG_0015.jpg

Haven't traced out a schematic yet. :cool:

Sincerely,
Kingsley.
 
6sl7 has a lot of gain for a line stage. The designer likely traded in quite a bit of it for lower distortion and reduced output impedance, by applying global feeback from the plate of the second tube to the cathode resistor network of the first. Have you plugged it in and tried it out? It looks well built, and stands a chance of sounding really good.
 
Good afternoon, Midnightmahem:

Wouldn't know about the circuit design and that is one of the reasons for reaching out to this community. (Although you are probably correct in the gain factor as I understand the the 6SL7 has higher gain versus the 6SN7.) Aside from the other factors of the tubes' electrical differences, am still wondering why there was audio distortion using the 6H9Cs (6SL7 equivalent) that came with the preamp versus barely any with the 6SN7s. Unless the 6H9Cs were at the end of life, could it be that the lower gain of the 6SN7s mask/not exhibit the audible distortion? Pretty new at this stuff. :confused:

Am awaiting the arrival of some 5Y3GT and 6SL7/VT-229 (and 6SN7) tubes to try/test further.

Yes, it is well built and does sound good with the 6SN7s (but not so much the 6H9C). (Just concerned if any damage/harm will be done given the difference in current draw as noted Parafeed and subtly by Eli.) ;)

Thanks.

Sincerely,
Kingsley.
 
Last edited:
Kingsley,

Previously, you stated that a 6J5 switch needed schematic help. I've provided the beginnings. Values for the triode's plate and cathode resistors plus the FET's cathode resistor are not specified, as they depend on the B+ rail voltage. You must measure the B+ rail voltage and report it.

Make absolutely certain that you read the safety "sticky", before putting probes into tubed circuitry. A mistake can, literally, get you killed.
 

Attachments

  • 6J5 Hybrid Line Stage.jpg
    6J5 Hybrid Line Stage.jpg
    286.4 KB · Views: 264
Good Afternoon, Eli:

Thanks to the schematic of the 6J5 mod. Not sure if I really want to explore this option.

Have finally traced out a schematic of one of the channels. (Hope it's correct.) :scratch2:

Not sure if the schematic provides enough info/details to discern whether swapping the 6H9Cs with 6SN7s .

One_Channel_DIY_Custom_5Y3GT_6H9C_Line_Stage.jpg

Thanks, all. :wave:

Sincerely,
Kingsley.
 

PRR

Member
Joined 2003
Paid Member
> traced out a schematic of one of the channels. (Hope it's correct.)

Plausible. (It will pass audio and make it bigger.)

It has a LOT of gain, and the turn-down is *after* the gain. If you feed a 2V CD or DAC in, it wants to make a 66Vrms 92V peak signal, which it can't do (on 260V supply), and is FAR more than you could ever need.

I also see the NFB loop has a different bass rolloff than the output, so when lightly loaded it is liable to bump-up slightly at 20Hz (not a huge fault).

Different tubes won't really change this.

The gain control normally goes *in front* (right off the Selector switch).

The "330" resistor could perhaps be 2k for lower gain. (We can't reduce the odd 4.7k+5.6k string because it is already a *heavy* load for any small triode.)
 

Attachments

  • audiobasic20001.gif
    audiobasic20001.gif
    8.2 KB · Views: 241
Good Afternoon, PRR:

Thanks for checking out the schematic/circuit and pointing out the good/bad and making suggestions. :cool: Something else to work with/on if not making the mods as Eli suggested.

That 330 ohm change to 2k appears to be a good and easy one and probably will try it out.

Something that's been on thought for along time - if one swaps out a lower value volume/attenuator pot, e.g., 25k (vesus the existing 50k one) - what effect does it have? Lower the gain/sound output level or increase the gain/sound output level?


What about the other way around, e.g., from the existing 5ok volume/attenuator pot to 100k (or even 200k) - is the gain/sound output level lowered or gain/sound output level higher

Also in both instances of the attenuator swap, does it affect the circuit parameters and loading or just a slightly different means of lowering/increasing the output gain to the RCA/preamp outs? (Think I have searched but didn't really get a clear understanding/picture of the differences and thier results.)

Thanks all and please, by all means, keep the discussions/suggestions coming. It's all helpful!

Sincerely,
Kingsley.
 
Good Evening, PRRm et al:

Apologies. Made a visual band reading and a typo on the schematic... the 330 ohms should be 3.3k :ashamed:

According to PRR's suggestion of swapping the 330 ohm with a 2k ohm resistor to reduce the gain seems to have been incorporated even if the resistance noted is larger in value/ohms.


Also have read a little more into the 50k attenuator/pot versus a 100k one and there does not seems to be much of a gain/output level difference (if any) but rather more of a when/where the wiper would be positioned, how sooner/quicker the gain/out level rises, and the loading of the triode (too high a load apparently is not good/recommended). Hope I'm understanding it correctly. :scratch2:



Thanks.


Sincerely,
Kingsley.
 
Too difficult a load is to be avoided. Low impedance loads are more difficult than high impedance loads. The triode in the 6SL7 is particularly poor at driving loads. Using it in a line stage is forcing a square peg into a round hole. :mad:

I've provided an example (original RCA Phono) that uses the 12AX7/ECC83, which is similar in character to the 6SL7. Notice the warning about a minimum 220 Kohm load impedance. BTW, the 7025 shown in the schematic is a variant of the 12AX7 than contains hum bucking, spiral wound, heaters. That feature can make AC heating at low signal levels tolerable.
 

Attachments

  • Original RCA Phono Preamp Schematic.gif
    Original RCA Phono Preamp Schematic.gif
    116.5 KB · Views: 208
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.