MELOS GK-1+1 Docs

[meles]

Inside or outside

Of the eight tubes in the Melos GK-1 and GK-1+1, four of them are 6GK5's. The original 6GK5's in my pre-amp don't look like other American or Japanese 6GK5's. They are coated on the inside. The yellow colour in the picture below is exactly how they look. Melos describes these tubes:

Uranium salts!?
...

You might try a cue tip with alcohol or water and rub on the outside to see if this material is present on the outside.

Again, no licking:
http://www.dartmouth.edu/~ehs/docs/uranyl-acetate.pdf

 

[meles]

PS upgrade to GK-1+1 status

Two smaller transistors that have some regulating function were also replaced — only because I suspected they had been stressed. The red one (ELM 198) and a Texas Intsruments TA 7805S. For these, I cut out some copper sheet and made heatsinks. The two devices were affixed with thermal grease.



You can see the D1092 power transistor on the big heatsink. It looks like I might have cut out a mica wafer to go behind it — with thermal grease of course.

Today, years later, all these parts look as fresh as daisies — no problem. The propeller seems more efficient. All the resisters were replaced with greater wattage ratings. One of the ceramic types in the lower right was made to stand on end. One leg was buttressed and then sleeved in fiberglass tubing. Being lifted off the board, it was more exposed to cooling by the fan and ambient air. Recently, I replaced my added-on cooling fan with a smaller one and it still runs very cool. If you add a fan, you can experiment with size and rpm to get the quietest fan that still does the job.

All diodes were replaced, but some types that are possibly tiny Shotkeys were unavailable and were reinstalled — very gently!

Lorne,
I've been following along the schematic as I look at your fine photos and description. Unfortunately, the schematic values are quite hard to make out. Here are my notes so far:
Heater Circuit:
1. Your picture identifies a bridge rectifier not on the board. The schematic identifies this as something like YELL and it would be awfully nice to have a part number just to confirm things. The wires go under the board to the rest of the heater circuit to my eyes.
2. 2.2ohm 25watt resistor and 3300uf 35 volt caps seem to match up well with schematic, but one of these appears to have a small bypass cap in the schematic. Not a big deal.
3. From the 2.2 ohm resistor I can see the trace going to the TA 7805s. This is a bit confusing but on the schematic we do see one end of the 2.2 ohm trace up and over to a blurry three legged box that must be 7805. The schematic shows a 1uf cap and I think your diagram shows three diodes(D) and a C? which I believe is the 1uf cap. We see the 470 ohm resistor in their too and IT WOULD BE AWFULLY NICE TO KNOW THE PART NUMBER FOR THE THREE DIODES. I have a feeling you might now have these hiding under the board.
4. The final transistor appears to be the original, unmolested LN3055 8352 on its original heat sink. Nice notes as this makes out and confirms the blurry identifier in the schematic.

The heater circuit looks good except for the three little diodes and I suppose one could just guess on those, but not knowing what I really am doing, part number or type would be nice.

HIGH VOLTAGE CIRCUIT (or fry and die):
1. Now we are really in trouble. The capacitors might match the schematic, but the transistors and resistors have very different values.
2. LORNE, WE'D NEED A SCHEMATIC OF YOUR UNITS HIGH VOLTAGE SECTION TO DO THIS UPGRADE.

At this point anyone who wants to do this upgrade as detailed previously needs to post on this page and Lorne might provide details. I'm still hunting for a unit and I'll ask myself if I get a stock GK-1. I can make out most of the schematic in this area except for one zener diode, so that might be a valid design that someone could use, but I'd rather have Lorne's circuit.

Lorne,
No need to do anything until someone asks.

 

[lorne]

I'm in a bit of a rush to get somewhere, but here is a question for you meles, and it may sound stupid: what schematic are you referring to? The one posted by capparelli in post #11? I may have to pull the head and fan off to answer your questions; however I may have a parts number page in my notes that I meant to post.

Indeed there is stuff under the board, and it was there before I rebuilt the board. As for the three little diodes: I remember that I identified two and replaced them using a Japanese parts reference book. In some cases it provided Japanese-made alternative part numbers. One diode— the smallest one — I could not identify. Neither could the clerks at the parts store nor their manager who is technically very qualified. For that reason, it was not replaced but simply reinstalled.

 

[meles]

I'm in a bit of a rush to get somewhere, but here is a question for you meles, and it may sound stupid: what schematic are you referring to? The one posted by capparelli in post #11? I may have to pull the head and fan off to answer your questions; however I may have a parts number page in my notes that I meant to post.

Indeed there is stuff under the board, and it was there before I rebuilt the board. As for the three little diodes: I remember that I identified two and replaced them using a Japanese parts reference book. In some cases it provided Japanese-made alternative part numbers. One diode— the smallest one — I not identify. Neither could the clerks at the parts store nor their manager who is technically very qualified. For that reason, it was not replaced but simply reinstalled.

That is basically the only schematic. No need to tear into anything yet until I have a GK-1 and hopefully I'll have a GK-1+1.

 

[jeb98]

I recently lucked into a melos gk1 preamp with the power supply. My preamp has 8 tubes sockets, and the power supply seems to have one transformer, not sure if its a GK1+1. The transformer does seem to have two separate windings in one housing if that makes any sense. Anyway, its missing all but 1 tube, and that tube doesn't appear to have the yellow coloring, so who knows if it's original, and who knows if this thing even works!

I wanted to see if anyone here would be able to take a look at this and fix it up. I would love to get a tube power amp and use this as my preamp (I currently have an eico ST-40 integrated). I would like to learn how to repair tubeequipment myself, but I don't think this is the place to stsrt. Please help!!

Best,

Jonas

 

[lorne]

Hi Jonas: AFAIK, the GK-1+1 had 4 only coloured tubes. Sometimes there are small red letters on the lower left of the facia that say GK-1+1. You probably have a GK-1 — not a bad thing at all and it would be great if you could get it working.

Read member meles' post # 55 and # 59. He did a lot of research in regards to identifying which is a GK-1 and a GK1+1. It would help us a lot know what you have. If you have a GK-1, from what we have found out in this thread, all 8 tubes would 6GK5's. If that is the case, then perhaps you can go on line and for about $50 order eight NOS 6GK5's, install them and see if the amp will power up and run a signal to some power amp.

QUESTION: What tube did you find? What is the alpha-numeric say on the tube. And also, in what tube socket did you find it? I can post a tube socket digram later, but for now maybe you can describe it — or better — could you post some photos of the whole amp?

I am pretty busy right now with work stuff, but i will try to help you. And I'm sure meles will get on board too. If you are not experienced with electronic repair, don't go doing anything with a live / powered-up circuit. I don't think that power supply would be very friendly if you were to ground yourself to current there.

 

[jeb98]

[/URL][/IMG]

[/URL][/IMG]

[/URL][/IMG]

I'll find out what type of tube is in there tomorrow. In the picture of the inside of the preamp you can see that there is one white wire in the lower right corner that looks like it is not attached to anything. I did get a variac along with this, so maybe it will come in handy of I decide to try the preamp out, but I will need a lot of guidance before I think about messing with it.

Best,

Jonas

 

[lorne]

Hi Jonas: firstly, as you will have seen in previous posts, my power supply is completely different and more complicated. Next, my chassis is different; apart from being enamelled black it is a different construction. Mine is a pressed out steel tub in which the single board is mounted on stand-offs. And to answer meles' question in a previous post, the top is steel as well.

Unless someone has changed it, the single remaining tube is almost certainly a 6GK5. If you find it is a 6FQ5, it probably means that someone replaced the original tube — BUT — the 6FQ5 is listed all over the place as being a direct equivalent for 6GK5. As to whether there are subtle sonic differences, I do not know.

The disconnected white wire: my guess is that this is one of the white wires that corresponds to left or right channel. It is connected to a pin on the switch that selects tape MONITOR position. It runs from that pin to one of the pins on either the L or R stepped-resistor gain switch. It exits on another pin on the same switch and runs to either the L or R sweeper pin on the volume control. Without that wire being connected — assuming I am right — on of the channels will be off-line. Melos used white wires for both channels.

I am in the middle of some boring paper chore right now, but later I will post pictures which should help you to identify which channel the wire beings to, and hopefully what pin it goes to. But, if you want to get ahead me, look for a pin on the switch that has solder on it but no wire or perhaps a bit of wire attached to it.

And while we are on the topic, permit me to enlarge a bit — something I did recently: see the next post. What you will see in the pics I will put up later is a record of the original wiring wiring that was intact before I installed a new volume pot and deleted the gain switch connections. I am not saying yous should do this, just be aware of what you will be seeing — a still connected original pot ranging and a new, upgraded pot installed in its place. The photos will also show the white wires marked for L and R channels and the MONITOR select switch.

It is great that you got a variac. I wish I had one. If you want to test this amp, power it up and diagnose stuff, I might suggest that you run over to AK's Tube Forum. The people there really know their stuff. For example, they will tell whether you should have the tube complement installed — or not — before you run the variac. I forget stuff like this and have to relearn what to do all the time. I am not the expert like some of these habitués of that forum. But, if you do run over there for technical help, PLEASE give us on this thread notice so we can follow your progress. And please do not touch a live circuit unless you are very sure that you know what you are doing. You can certainly test it with the variac, but power off and wait for caps to drain before you do anything on the board. There may not be anything lethal on the main board, but don't take my word for it. I DO know that my power supply could kill me!

It looks to me that you have a GK-1, and I for one would be happy to own it. Let's see if meles agrees.

My two cents here: get it running. Enjoy it ... and only then chip away at getting it up to snuff. For example, there my be some caps in the power supply that could be fresher because they will have been heated up over years of use. The main board runs very cool — even in the humidity of our Japanese summers, so most if not all the parts are probably just fine. Lastly, I for one would not rush to trying to make it into a GK-1+1 in any way. For one thing,, you have the wrong power supply for that — so I believe. I think you will like the amp a lot — just the way it is.

 

[lorne]

I was having trouble with my GK-1+1. For a long time, I had been suspecting something was not quite right, which was why I started running signal to the big Sansui from the TAPE OUT ports. I ran it this way for some times and then reverted to normal MAIN OUT mode. Then I lost a channel.

Jump to the next scene — I changed out the volume pot and deleted the gain control switch wiring. The gain assemblies remain in the amp, but they are wired out-of-circuit.

Melos installed an Alps 100K Ohm pot in this amp, but it was the cheap, unsealed variety — the part you can still buy any day for about $6. I replaced it with the sealed Alps pot that I believe is nicknamed "The Blue Silk".

The problem with not doing upgrades and mods in stages is that you don't know how much any single change made an improvement — or in some cases a disappointment. In this case, the sound improved very much. Moreover balance is perfect, whereas before it was just not right no matter where I set the gain.

Some people need balance because of their rooms. In the case of the GK, at least one owner has exchanged the stepped-R gain switches for quality, resistive pots. If you don't need balance, my suggestion is to just wire them out of circuit. I have no reason to believe the tubes that I am currently using were matched, but I have no complaints about the balance.

I find that moving the speakers around a little, or toeing one in or out can correct modest imbalances more effectively than balance controls. Moreover, I never was a fan of the Melos gain switches, and I am not alone on that score.

 

[meles]

Testing

Yes. If you inadvertently short out any of those high voltage high value capacitors with your hand, that could give you quite a jolt (I am no expert, but might wake up on other side of room after a few minutes!)

If it were mine I'd start the variac out around 90 volts and move up 2 volts or so every 15 minutes. I would run the unit with no tubes or maybe just tubes in one channel so you don't need a full complement of 8 6GK5s (your unit looks early and might even be a PGK-1). You can test voltages on the open tube sockets one at a time. Any multimeter will do. Just put ground probe on chassis and insert red probe on end of tube socket holes one at a time. Some will ready 6 volts DC. Some may read nothing. At least one on each tube should read high voltage (100 to 200). You also should have 6 volts on heater pins. If the tubes are not glowing after turn on that probably means you don't have heater voltage.

As the unit is warming up of course, smells or the lack of high voltage or heater voltage would be a sign to stop. If everything seems ok, then you might hook up a CD player and play a track on repeat. You can put the hot probe in the hole of the rca jack on CD player for fun to see what voltage it puts out. Then you can test at the output of the line stage for AC signal (which should be much higher.)

I think you could do this and fish around pretty safely. Of course if you stumble and stick your hand in the power supply or short things with your probes then the fireworks begin. The unit appears to have a metal base for the tube sockets. DO NOT SHORT THIS WITH YOUR PROBE WHEN TESTING AT THE TUBE PINS. Most tube units don't have shield bases so this makes this a more touchy operation.

 

[meles]

....

Jump to the next scene — I changed out the volume pot and deleted the gain control switch wiring. The gain assemblies remain in the amp, but they are wired out-of-circuit.

Melos installed an Alps 100K Ohm pot in this amp, but it was the cheap, unsealed variety — the part you can still buy any day for about $6. I replaced it with the sealed Alps pot that I believe is nicknamed "The Blue Silk".

The problem with not doing upgrades and mods in stages is that you don't know how much any single change made an improvement — or in some cases a disappointment. In this case, the sound improved very much. Moreover balance is perfect, whereas before it was just not right no matter where I set the gain.

Bypassing balance controls is a great upgrade especially if the Melos is messed up in this area. No audiophile worth their salt should have balance controls in their system.

The Melos and even the Alp blue silk are carbon wipers based on carbon resistor technology. These change values with age and often are not well balanced due to the lack of precision with carbon. Better volume controls are stepped with discrete resistors. There is quite the cottage industry in high quality volume controls.

The Alps is a good starting point, but if you are going to take the trouble to replace the volume control you may want to move up the food chain a bit.

Motorized Alps volume controls are quite common in high end equipment, so it is the most economic choice from a manufacturing point of view. But if I was paying someone $50 to upgrade the control, I'd look around. In my current preamp I upgraded to a Placette Remote Volume Control and it made a huge difference and so I am now a big believer in volume control upgrades.

Lorne has been using just his phono stage direct into an amp with volume controls at the input. That is a fine choice as the sonically compromising volume control is later in the signal path with almost no wiring after it. He also is not even sending signal through the line stage tubes so these two things look very good on paper.

Have a friend with monoblocks who likes to set his input volume controls lower so his preamp runs higher in the volume range. I was shocked at a what a difference this made on his system and really don't have a great explanation for this phenomena. Volume controls are evil.

 

[meles]

Virus

My PC has the virus from Hades. Don't click on any of those links above! Yikes.

 

[jeb98]

OK, thank you for all the replies Lorne and Meles. I will try and find some 6gk5 tubes. The one tube in mine is a 6gk5, branded Westinghouse Japan. Meles, you mentioned "do not short this with your probe when testing at the tube pins". Can you elaborate on what I shouldn't do, I just want to understand what you meant. I'm an amateur on tubes but I'm reading up and trying to learn more. I know how dangerous coupling caps can be, and I'm not just going to poke around. But while I'm pensive, I really want to get this thing working. Plus it would be great to be able to use this with a decent MC cartridge like the denon 103.

It looks like the detached white wire (now i noticed it isn't attached in two places) are possible meant to be that way, as it appears the same as in a pic of a GK1+1 I saw in a previous post from this thread. If anyone has a good pic of the inside of the GK1 so I can compare a bit with mine, that would be great.

Best,

Jonas

 

[meles]

OK, thank you for all the replies Lorne and Meles. I will try and find some 6gk5 tubes. The one tube in mine is a 6gk5, branded Westinghouse Japan. Meles, you mentioned "do not short this with your probe when testing at the tube pins". Can you elaborate on what I shouldn't do, I just want to understand what you meant. I'm an amateur on tubes but I'm reading up and trying to learn more. I know how dangerous coupling caps can be, and I'm not just going to poke around. But while I'm pensive, I really want to get this thing working. Plus it would be great to be able to use this with a decent MC cartridge like the denon 103.

It looks like the detached white wire (now i noticed it isn't attached in two places) are possible meant to be that way, as it appears the same as in a pic of a GK1+1 I saw in a previous post from this thread. If anyone has a good pic of the inside of the GK1 so I can compare a bit with mine, that would be great.

Best,

Jonas

I think all the best photos on the web are in this thread so you keep going back and download the images for your reference including the schematic.

The solid state power supply will quickly fail if you short a dc power supply voltage wire to ground (tube stuff won't like it, but you might get away with it with no lasting damage.) So, when you put a metal probe in a given socket hole generally it is very safe, but with those metal collars on the sockets there is a good chance they are ground. If the probe touches a 100 volt socket hole and then leans into the metal then you have created a short and this will likely damage the unit. The probe is insulated so you'd probably see a spark and jump out of your skin. If you make the short then with this unit or any high voltage high capacitance tube unit, then you get a free super defrib which is not something you want to do!

Your old Eico ST-40 might give you something to remember if you short something. The capacitance values in the Melos are ten times as high so that makes for ten times the fun. I've got a 1700uf photogenic capacitor in my tube phono stage which scares me pretty good just looking at it. I've heard of high power tube amps running at 2000 volts. Truly deadly no matter what your state of health.

 

[lorne]

Jonas wrote:

It looks like the detached white wire (now i noticed it isn't attached in two places) are possible meant to be that way, as it appears the same as in a pic of a GK1+1 I saw in a previous post from this thread.

Maybe you saw a pic of yet another GK amp with a DE-ttached wire. I do not mean to be flippant, but just to say that I for one would be surprised to finally see a piece of electronics that has a wire that ends up in "no-where" on a circuit — unless it came off from stress, fatigue, bad soldering or ??? ... or someone cut it.

Sorry, I have been up working all night at its past 8:00 A.M., so I will not post the pics I promised this morning. Please excuse me.

In the meantime, get a magnifying glass and look closely at the pins on the MONITOR switch. If you were here in my city, I'd bet you a decent lunch and a bottle of claret that you will find solder on a pin where a wire must have been. And, as you seem to suggest, it is also detached at the other end. Examine the gain switches. If there is a pin with solder in it, but no wire attached, then I'd suggest that this is the other terminus of the white wire. I do not see how you will get two two channel stereo unless someone either bypassed the gain switch with another wire. Or, was someone trying to make this a mono-amp?

Just my two cents, and I am not trying to contradict meles: he is instructing a diagnostic investigation. Sensible for someone used to working around some nasty current, but maybe a bit too adventurous for you. Most of the circuit is pretty mild — IMHO. But some of it could hurt you. So ....

I'd put the variac on the amp — no tubes — and see if the power supply survives. Start low and go work up to 110 volts over 30 minutes. If the amp has not been run for ages, you may get the slight smell of aluminum — some electrolytic capacitors heating up and reforming. It will gradually dissipate and go away — but of course it will leave you wondering about he state of old electrolytic caps. But let's not worry about that now!

If everything is OK, but for a bit of oder of Al and ozone — not strong, no smoke, nothing fiercely hot on the board — continue. Now stick the tubes in and run it. Everything OK? Then just run your interconnects to source and then to power, then onto the speakers and see what happens. (I always like to run a set of speakers that are expendable when testing because I KNOW that I can make mistakes. Some pro's here in Sendai who I know always use the same speakers and never blow them! They really know what they are doing.)

NOTE: This is NOT expert advice. If I were seeking experts, I'd ask the Tube Forum people, or take the amp to a pro'. I am lucky; there is a repair shop about 2 Km from my house. (They are older engineers who know both tube and solid state electronics. Their shop is fascinating!).

That being said, meles seems to know what he is doing — but do you trust yourself to be probing around?

If it were me, I would not be inclined to probe. If the there is no indication of heat, smoke and very bad smells, I'd mount some new tubes and and go for it. See if it plays. If it does, listen for how it plays. If you blow a couple of tubes ... well ... they are not $100 big fire bottles.

That loose wire is something we should look into very closely!

Please do not be discouraged. If you want to hear what a piano really sounds like in good audio, the GK will give you a head start in realism. If your power and speakers are up to snuff, you will fall in love with this amp — even if it is not the last word in regards to the GK pre-amps. I trust that they were all aiming in the same direction.

 

[lorne]

Hi Jonah: The following two post show the wiring for the front panel. I hope it helps with the loose wire in your amp. I think that the wiring should be the same as yours even though it is an earlier model. The pics are in no particular order. Here is the first set.

 

[lorne]

Some GK1+1's that were the result of factory upgrades apparently had the mute switch deleted. I use it, so I've kept mine in the circuit. If you ever decide to change out the pot, be sure to notice which pins are the output (on the ganged wipers), the ground and the signal output. The pot in the pic below is an Alps "blue" like my new pot, but the wiring on your stock pot should be the same. You can check to see if all your wires are still attached — particularly the white wire you referred to.

Meles mentioned that there is a long way to go in pots for the best upgrade — stepped resistor types being preferred by enthusiasts. I agree, but this pot works well enough for me, and it is a lot more affordable. Resistive pots have an optimally effective lifetime. They may work, but they may be past their prime. DeOxit will not cure the wear, but might be a band-aid.

 

[lorne]

As you can read in the annotations, I deleted the wiring for the gain switches. But, in the pics, all the wiring is still intact. To delete the gain switches is not difficult in theory, but I find that wiring to small pins on switches and pots one of the harder jobs in soldering — getting a good mechanical connection and using just enough heat so as to be healthy for the pot while at the same time avoiding a "cold joint".

To delete the gain switches, you just run wires from the monitor switch straight to the input pins on the pot after deleting the wire that runs to and from the switches. I used 24 gauge, Teflon coated wire (not easy to work with). I wired the output with the same wire to a terminal strip to join with some heavier gauge wire that runs under the board. Someone might disagree with this added connection, but I did not want to use the heat it would take to wire the thicker gauge to my new pot which had smaller, more delicate pins than the original pot.

I agree with meles. If you don't need balance, just delete the feature — even if they are stepped-resistor switches as in the stock GK1. The more usual 'balance pot' contains a resistive carbon track that varies in width.

Now, you may think this is jumping the gun at the stage you are in, but over the years people have reported that they had trouble with the gain switches. In fact, the previous owner of my amp sent it back to Melos to have new switches installed when it was still in its youth. So, as you try to get this amp working, you might just keep this in mind.

In the next picture set, note the diode (annotated) and the wires leading to it — especially the blue wire that is rather fragile. This diode is part of the relay circuit, and if any of the wires is making a bad connection or is broken off, the relay will not trigger power to the board. When you power on, assuming the mute switch is in the 'down' position, the diode will turn from red to green, at which time you will hear the relay click. This can take a seemingly long time. Mine takes even longer than usual cuz I have a thermister in the power supply circuit that slows the whole process down even more.

When you finally get to powering up on the variac, I do not know how much power it will take to trip the relay on, so you may have to be patient. More knowledgeable people on the Tube Forum would be more helpful than I on this subject.

Finally, take note that the diode assembly as shown has been stripped of its plastic, insulating sheath. The blue wire broke off, even though I was not horsing around in there. If you too have to reflow or solder any of those wires, I'd use a heat-clip before the diode. Heaven knows where you get a split coloured diode; I've not seen one in my city.

 

[jeb98]

Thank you so much Lorne. I really appreciate you and meles help. I will check out your pictures in relation to mine, and then I will take some more pictures of the wire i am talking about (I won't be able to take pictures until tonight). Mine does seem to have different wiring, but it should be similar enough.

Thanks again,

Jonas

 

[jeb98]

[/URL][/IMG]

[/URL][/IMG]

[/URL][/IMG]

The first two pics show the two white wires that look like they were snipped, as there is no lead showing that could have become detached. I'm not sure where these wires should have gone.
The third picture shows the volume pot, which is where these two white wires lead to. You can see the two thicker wires soldered to the volume pots terminals.
My question is, what do you think? Where should these wires coming from the volume pot lead to? Any reason they would be snipped?

Best,

Jonas