Welcome, Guest. Please login or register.

August 06, 2020, 11:59:10 PM

Login with username, password and session length

Recent Posts

 

Author Topic: Gibson LabSeries L5 (pics)  (Read 15143 times)

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Gibson LabSeries L5 (pics)
« on: August 18, 2006, 12:58:48 PM »
Hello... I took some pics of my L5 chasis in the crappy head I built (it didn't come with the original cab--got it cheap!). First one is right before I took it apart:



This one is the guts:



You can see the power supply (the transformer is mounted on the bottom of the chasis at the bottom of the pic), the power amp (upper right of pic), and the preamp is all behind the pots... you can't see them in this pic. You can read a little about it and see the schemo here: http://www.netads.com/~meo/Guitar/Amps/Lab/schem.html

Now that I've got you looking at this post, here's my problem. I've been using this thing for about a year now with no problems what-so-ever... but, about two weeks ago I was recording and it was cutting in and out... not exactly cutting out, but getting much quieter then louder again... I thought it was a pedal problem, so I took all pedals out of the loop and it worked fine for the rest of the night. So, two days ago I'm recording again and it starts cutting out again... I try every combination of equipment in the practice space and it has to be the amp... that's why I took it apart, but I couldn't find a loose connection or bad solder joint. A friend suggested that it might be overheating or that the power in the room is dipping too much... both seem unlikely to me because I've been using is for so long with no problem... If the electrolytic caps were failing (this things pretty old) what would that sound like? Do these seem like symptoms of anything you're familiar with? Ideas on fixing? Thanks!
« Last Edit: August 18, 2006, 01:00:38 PM by Stompin_Tom »
myspace.com/bestinventions

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #1 on: August 18, 2006, 01:09:28 PM »
Ok.. pic of preamp, too:

myspace.com/bestinventions

teemuk

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 1074
  • Chip Points: 282
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #2 on: August 18, 2006, 02:10:41 PM »
Are you using the compressor? I rememebr reading that a broken compressor circuit was the most common fault in that particular amplifier. That could fit the description of the problem - after all, the comp circuit controls the volume.

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #3 on: August 18, 2006, 02:26:11 PM »
I don't actually use the compressor... it does seem to work fine when I've tried it, though.

I just read a review on Harmony Central where the guy says that some of the earlier models had to have their rectifier diodes replaced... He didn't say why and it's just a harmony central review, so I'm not sure what to make of this. Maybe if the power isn't rectified properly it could cause weird volume drops? Do diodes go bad with time and use?

Here's the power supply, by the way:

« Last Edit: August 18, 2006, 02:56:25 PM by Stompin_Tom »
myspace.com/bestinventions

teemuk

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 1074
  • Chip Points: 282
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #4 on: August 19, 2006, 03:07:03 AM »
Diodes, as any semiconductors go bad with heat. I don't actually know what a bad rectifier diode would do but I assume they either:
1) Short: Your supply voltage becomes AC instead of DC. You would hear this as a massive amout of hum.
2) Open: One or several of the diodes would not conduct: You either would not have supply at all or it would be half wave rectified. Symptoms; a hugely increased amount of hum or no sound at all.

Usual problem from old filter caps is - again - increased amount of hum and sudden crackles and loud pops in the signal. If the the cap would have become totally shorted or open due to aging (actually quite rare but could happen as well) the symptoms would vary according where the capacitor is used in the circuit. A dipping supply voltage you hear as earlier clipping (lower headroom). It will not affect the gain on any level.

You could look the thing on a scope. In some amplifiers, the bias diodes (that control the quiescent current) do - at some level - control the gain. If they'd overheat too much it might cause something like this. You would hear a "buzzing" crossover distortion everytime the volume goes down.

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #5 on: August 21, 2006, 11:00:25 AM »
So I recorded with it again, and then practiced afterwards... It was on for several hours without ever cutting out. Strange. But the heatsink was very hot afterwards. I'd never noticed how hot it got before. I think I'll drop a fan into the head somewhere. Below the heatsink? Old computer fans run at half voltage work ok, right? Have to get a little transformer of some sort, too...

An additional advantage is that I've heard you can drive a 4ohm load with one of these if it has a fan to cool it. In fact, I'm pretty sure the '200 watt' L11 head they offered is just the same design, plus a fan and a 4 ohm cab... thus getting twice the wattage out of the same design. Not that I need it any louder, but just in case I have to use a 4 ohm cab at a gig...

Thanks for the help teemuk.
myspace.com/bestinventions

teemuk

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 1074
  • Chip Points: 282
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #6 on: August 21, 2006, 11:26:36 AM »
The L11 is basically the same design allright, except.... It has a fan and it was also designed for 8 ohm loads as well ( 2 x 16 ohm cabs in parallel). It also had a different power transformer with a higher VA rating and raised supply voltage. Logically, current demand for L11 was higher as well. All parts were selected considering the fact that the amp had a higher power rating. Also, I guess the output devices were different as well. If you look at the schematic it will tell you all of this. Look for points marked with "8" inside a triangle.

The 8 ohm impedance is the safe rating for this amp. Don't go below it - whatever the designers tell you! Remember that they DID design this thing for 8 ohms! Most amps are designed to tolerate smaller than nominal load impedances for a short period of time but judging by your description this amp already has problems even with the nominal load. If you expect to double the output power by halving the load impedance I will assure you this hardly ever happens. Most likely your amp will not have the required current feed capacity and you get only 1/3 increase in output power. As the price your amp will probably overheat, then blow.

If your heatsink is already substantially hot I'd suggest you start looking for failed parts instead of adding a fan and going for even smaller output loads. You might have a failing bias circuit and in this case you are in a very near danger to blow your entire amp.

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #7 on: August 21, 2006, 11:39:57 AM »
I see. That makes sense... I'll take a closer look at the schematic.
myspace.com/bestinventions

swt

  • Regular
  • **
  • Posts: 11
  • Chip Points: 1
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #8 on: August 24, 2006, 12:34:47 PM »
hi. first time in here. usually the problems you're refering to can be, bad solder at big resistors, usually ceramic, as these are large, they usually crack the solder from vibration, and also, bad caps, or bad resistor. check the ps cables too, because sometimes turning on and off, can be caused by the voltage going out, in the output section of the amp.
80% of the times it's the resistors. check them and let me know.
if your amp is dissipating heat, then check the value of those resistors too, might be a shorted one, and also shorted caps...

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #9 on: August 24, 2006, 01:29:55 PM »
Thanks. It's at the space right now. I'll bring it home tonight and go over all the solder joints and check out the resistor values before installing the fan I just got. I hope nothing too serious has blown up in there. A bad joint or a slowly failing component would explain why it doesn't always cut out...
myspace.com/bestinventions

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #10 on: September 11, 2006, 12:47:14 PM »
Ok. I'm back again with this beast... I put in a fan and after looking at the power supply more closely it looked like someone replaced or changed the four smaller diodes on the power supply (CR205-CR208)... so I replaced those for good measure. I also went over all the solder joints on the power supply and power amp. Didn't look like anything was loose, but I resoldered everything that looked even slightly questionable. I took it to practice and, poof, same problem as before. I tried using a booster as a preamp and, yes, same problem as before. Must be in the power amp or power supply.

Teemu (or anyone else), can you help me with the bias circuit/diodes... I'm not sure what/where these are or how to adjust/fix them. I've stared at the schemo, but I don't understand enough of what's going on in their to figure anything out...

Thanks!!
myspace.com/bestinventions

teemuk

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 1074
  • Chip Points: 282
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #11 on: September 12, 2006, 07:13:36 AM »
Well, the power amp is quite basic: Q301 and Q302 form a differential circuit (also known as long tailed pair or LTP) that controls a simple one-transistor voltage amplifier stage (VAS). The VAS mainly consists of transistor Q304 in common emitter configuration and bootstrap current source load R316, R317, C310. The output of voltage amplifier stage is directly coupled to - and buffered (current amplified) by - the output "triplets" of Q307, Q309, Q311-Q3314 (NPN) and Q308, Q310, Q315-Q138 (Sziklai PNP). Each triplet correspondingly amplifies the current of it's own halfwave of the signal. It's pretty difficult to visualize at first but all the output transistors can be thought as only two powerful NPN and PNP transistors in emitter follower configuration. Current has to be amplified as well so that the output voltage swing from voltage amplifier stage can affect over very small loads. The VAS itself could not drive an 8 ohm load.

The network surrounding the two other transistors (Q306, Q307) is a VI-limiter circuit that senses the output current and shunts the drive signal to feedback path should the current rise too high. This protects the amplifier from too high currents produced by i.e. shorting the output. It's not full proof but will assure the output stage will not blow in nanoseconds in case of an error. The basic idea of this circuit is to be a lot faster than an average fuse.

What is located between the bases of the first output triplet transistors (Q307 & Q308) is a circuit that sets the quiscent current (bias). You can think of it as resistance forming a voltage drop between the bases. As you can see it consists of two parts: 1) An integrated diode string at the bottom and 2) VBE-multiplier circuit at the top . The VBE-mult transistor Q303 is mounted on heatsink to sense the temperature of output devices since the temperature affects the forward voltage (base emitter voltage) of transistors. Since voltage drop over R311 and R313 has to equal Vbe of Q303 (it is in parallel with it) the voltage drop over the VBE-mult circuit is controlled by the temperature of the transistor. It should be approximately: R310 divided with R311 + R313. Then just add 3 x 0,65V = 1,95 V from the diode string to know the figure you should have between the bases of the transistors.
Note that diodes are affected by temperature as well, however, I do think the effect is pretty minor since the integrated diodes are not mounted on a heatsink.

You should check that all these parts work as they are supposed to. The bias is adjusted with R311 and set by replacing the fuse F21 with an Ammeter and setting the reading to 200 mA with no input signal. (At room temperature, and after allowing the amplifier to stabilize for at least 3 minutes). After this R306 (DC offset adjustment) should be adjusted so that there is 0 volts (+- 10 mV) at the output. Note that the bias mainly controls the amount of crossover distortion in the signal, it should not affect the gain much - unless something is terribly wrong.

You want to read at least the link number 2:

http://www.netads.com/~meo/Guitar/Amps/Lab/Images/l5_notes-1.jpeg
http://www.netads.com/~meo/Guitar/Amps/Lab/Images/l5_notes-2.jpeg
http://www.netads.com/~meo/Guitar/Amps/Lab/Images/l5_notes-3.jpeg

And here is the board layout:
http://www.netads.com/~meo/Guitar/Amps/Lab/Images/Lab_Series_L5_L7_L9_L11_board_layout.jpeg

BTW, the service notes label CR310 and CR311 as "bias diodes". This is a little bit misleading since the basic idea of those diodes is to even out the differences between transfer curves of positive and negative half waves caused by using both Darlington configuration (NPN) and Sziklai configuration (PNP) at the output. These "Baxandall diodes" should make the transfer curves of both half waves more similar and therefore reduce the amount of distortion occuring at the point where the signal crosses over to another halfwave.
« Last Edit: September 12, 2006, 07:56:26 AM by teemuk »

Stompin_Tom

  • Master
  • ***
  • Posts: 94
  • Chip Points: 2
    • View Profile
Re: Gibson LabSeries L5 (pics)
« Reply #12 on: September 12, 2006, 11:21:16 AM »
wow, thanks. What a detailed discription. That's very impressive and actually very helpful. I'll check the trim pots as soon as I get the amp back home... Hopefully one of them just got knocked out of wack... Although I suppose most of those parts wouldn't be too expensive to replace if they have failed.
myspace.com/bestinventions