Carlsbro TC60 Repair and Design Study – Part 1

Print Friendly, PDF & Email

Having a repair job on another one of these models gives me the chance to learn about the one I already have, which I did not have to look when I bought it, as it was fully working already. A comparison of two similar amps is always interesting too – any major component changes, power output differences etc.
I can start on this in the same way I did with the Klipp, by studying the circuit before the amp arrives, as Chambonino has a schematic an pic as usual for this model:
http://www.chambonino.com/carlsbro/carl60tc.html
This may be what the “PCB” looks like – as usual for Carslbro, as tidy and robust with good quality components, as the Marshall JMP head:

It’s still point to point enough to work on without hassle – the “no PCB chips yet” era – good!
Here are two possible schematics:


I’ll go with the more complex one – with “Drive”. There are 4 x ECC83 preamps – one of which (V3B) seems a dedicated OD triode for clipping – looks similar to the Klipp channel cathode follower:

Note the NFB link to the 100k at the top of the 100k presence. You can see that the variable “load” alters the gain from 47k to 147k, of the triode2 stage, similarly to the load Line graphs below for different common value anode load resistors – 47k, 100k and 220k.
I’ll start with the Mains PT logically and see what needs doing there – straight away, the fuse should be moved from the ground to the PT secondary so the rectifier diodes up to the standby switch doesn’t stay live if the fuse blows, a la Blencowe:

This “grounded fuse” was common in some old amp circuits I’ve seen – e.g. the Tremolux, and JMP. It should to go in series in either green PT wire, but I’ll see what’s involved as to whether it’s worth the trouble or not, as it’s not a “safety” issue unless you are working on the amp plugged in, in which case you should know what you are doing anyway and not be touching anything when plugged in…

All else seems fine, and I like the arc discharge cap across the standby switch so the stored choke energy can dissipate and not burn out the switch.
There is a variable 10k slider for the power tubes cathode bias resistor too, like in the JMP, which is good if way out like the Tremolux and Klipp were, running too hot/red plating, instead of (hopefully) changing a resistor value completely. It looks like it sets the power cathodes to -37V. This is the same as the Tremolux values of -36V so should be about right for 6L6s/EL34s at high 500V HT levels.
I’ll start the signal path with the input channels and draw a load line for them as they are identical – standard 100k load resistor but a low value 470R cathode resistor which is shared by both inputs, instead of the usual 1.5k. Also the voltages are a bit lower than 300V, at 210V or so.

Doing the calculation for those values:
210V/100k = 2mA so the line can be drawn on the ECC 83 grid curve, roughly seeing -2.5V to 0Vpp grid input signal unclipped headroom, for a cathode bias point at about 1.2V, and an anode swing of about 200V-70V = 130Vpp:

Be interesting to see what the actual values are (after the fault is fixed!).
For 1V change in grid voltage (-1 to 0V), a corresponding change of about 125V-65V occurs, so a gain of about 60, as expected from a clean first stage triode.
Both channels have independent bass, treble and volume controls but Ch2 has a Response knob, It looks like channel 2 is the brighter channel as more bass can ground via the 500pf cap than the Ch1 100pF.
This would fit with my twin cab model – Bass and Treble channels, 1 + 2.


When considering the CH2 Response control – which maybe the same as my cab “Top Cut” control or the Presence knob maybe – I found an interesting comparison of Marshall and Fender tone stack response differences here:
http://www.gmarts.org/index.php?go=212

“Here is a simple comparison of Marshall and Fender response with what might loosely be called ‘typical settings’ of Bass on 3, Middle on 4, and Treble on 6. The most obvious difference is that the Marshall lets more level through, and their tone controls have less range of adjustment. The higher level means that by using the same number of preamp valve stages, a Marshall can overdrive the output stage more.”


I can’t read tone stacks well by looking at them, so I can’t compare the TC60 stack with these to know which it is more like, if either.
I did some more recordings of clean tone comparisons of my TC60 and the Tremolux using the BR80 cap mics, and there is little between them, so I ‘d say these amps are more Fender than Marshall for clean – but as most valve amps use EEC83s in the first stage there won’t be much between any of them tonally until you get to the tone stack (and depending on the bypass cap value). This is where a manufacturer’s trademarks tones are crafted and so recognized mainly.
http://stevepedwards.com/ElectronicsStuff/?page_id=5436
Search (F3) “TC60 Bass Channel and have a listen – the TC60 has wee bit more clarity than the Fender for me – but so little in it.
Overdrive is a slightly different issue of course – things get complex moving toward square waves – but the tonal essence – the frequency content – fed to the OD stage is still created primarily at the tone stack, as stated above – not just in terms of volume fed from there, that may drive later stages a bit harder, maybe 12dB more as the Marshall does in that graph.
I haven’t compared the Fender/TC60 OD tones head to head yet (too loud!) so I’ll do that in DBS next week maybe.
After the EQ stage the two channel signals can be mixed via each volume knob and sent to the grid of the Drive cathode follower input, below. Note each channel output triodes have a shared 1k cathode resistor, same as the input triodes are also shared by the low 470R.
The output triodes should have a slightly higher gain with a 220k load and a 1k cathode resistor, if you remember Blencowe’s curve with the three common EC83 load values – 220k, 100k or 47k. The larger the anode load, the larger the anode swing for the same grid input, so the higher the gain. A 0V-1V grid change gives 50V for a 47k, 65V for a 100k , and 75V for a 220k (for a 300V HT).

For the Ch 1+2 output triodes, I’d expect a gain of less than 75 each here though because the HT is lower at 210V.

Looking at V3A+B, the Gain (Drive for the LTP stage next), the grid curve can be looked at to try and see what happens for the drive side of the cathode follower to try to understand how the distortion happens here, but may need actual readings to avoid the maths:

Normally, a symmetrical cathode follower gives unity gain, but these two triodes are quite unequal, one with a 220k load and the other with a 100k+47k cathode load, so 70k difference so should be a stage gain of less than 1. This is about tone retention from the prior stages, variable gain, and to impedance match to the next LTP stage.
V3B has a 100k+47k load resistance on the output side at 245V.
This gives variable gain between:
245/147k = 1.6mA. and
245/47k = 5mA
The Load Line would be between:
DriveLoad.jpg
This is more difficult to work out where the bias points is for these extremes as such high voltages exist at the grid of triode2 of this DC coupled cathode follower, and it is variable anyway. I may as well wait for real measurements and draw the Load Line in that Post. The 147k is a similar slope to a normal 220k anode load with maybe 190V cathode output swing here, so maybe a gain of up to 95 so high gain, and maybe very asymmetrical/clipped depending where the bias point sits, to feed the LTP stage.
V4 is an LTP with three immediate points of interest – Limiter, NFB and Asymmetry.
Global NFB from the 16 ohm OT winding is applied via the 68k resistor to the non signal grid (blue).
The term “limiter” means…what?
It seems it affects the fixed amount of NFB applied at the non signal grid (pin7) by tapping an amount of main signal (orange) to add to the NFB (blue) via the 100k limiter pot:

The LTP is also slightly asymmetrical with 82K and 100k anode loads at the same HT. This causes distortion, so 2nd order harmonic distortion too, and probable hard clipping depending how much gain from the prior stage is sent, as usual with an LTP stage. The waveform from this LTP stage is usually what is reproduced reasonably faithfully in power terms at the power stage.
The scope will show me what does what exactly when the amp arrives.
I’ll do the part 2 Post when I know what I’m up against – hopefully not blown transformers as worst case, as I don’t know where I’d find replacements…
31/5/14
Amp arrived and its a new 2008 model not the old one, so that schematic study does not apply very well!
See next part 2 Post