Schematic analysis of the Marshall Valvestate 8008

[J M Fahey]

Thank you for the tip Enzo,
I've seen the compatibility while searching for substitutes, but then I've found some cheaper BDV pairs on ebay and bought them.

Mmmmmmmhhhhh, I'd trust MUCH more a couple TIP142/147 bought from a reputable supplier than any "EBAY" part, doubly so if they are hard to find/obsolete/unavailable, and being cheap.
Even Marshall themselves switched to TIP142/147 in later production.

I trust them as much as these Dollar bills:

[Roberto]

Thanks again,

What do you suggest me to check when I'll receive them?

[Enzo]

Just try them out, they either work or they blow up.

I'm with Juan, I never buy parts on ebay.

[J M Fahey]

Agree, meters can test the *gross* problems as in shorted transistors but they apply very low voltage (a couple Volts) or current (say, 1 to 5 or 10 mA) to the transistor to really stress it so unless shorted or open fakes will measure good.

The real test lies in the power amp itself, where you apply 40 or 80V to them, load them with a speaker, and pass 5 or 10 Amperes ... that separate the boys from the men .

As Enzo says, now you solder them in the board, mount them in the heatsinks with mica, grease, the real job, first try the amp without load (neither speaker nor resistor) and using the lamp bulb limiter, just in case.

If you drive it and can get output signal to sweep from rail to rail, even clipping, and you see nothing weird , at least bthey stand the voltage (compare that to the couple Volts applied by a meter) .
Then connect a load, still with bulb or using the USA way: using a Variac, which lets you start with much reduced mains voltage and gradually rise it, and make it work a little.

Even with the bulb in place (you may use a larger bulb, say 100W) you can get some 10/15W into the load, if fine plug amp straight into the wall and drive it more and more, up to clipping.
If fine, fine

With variac you rise mains applied slowly until full wall voltage is applied, same thing.

Common problem with fakes is that they work at reduced power, in fact for people repairing home Hi Fi (say, 15 to 40 W/channel) they may even work forever, but Music Instrument amplification *must* match a drummer which means using amps between 40 and 100 W RMS (if SS) , a heavy job.

Fake transistor is such a problem that large factories (Peavey/Fender/Marshall/Laney/Crate/H&K) buy *only* straight from manufacturers; small fish like us must rely on smaller distributors or suppliers.

*Maybe* your BDV are actually relabelled TIP142/147 ... which is actually good.

But old standards, such as MJ15003/15004, unobtainable old 2SK/2SJ Hitachis, and the like, which are both hard to find or plain unobtainable are good business for fakers.

Counterfeit Transistors

this in Spanish, because first were detected in Mexico, unfortunately worldwide today:
Transistores falsificados

a guy got SO fed up with high power amplifier repairs coming back to his shop that he invented this "swim-or-die" circuit: he uses suspect transistors to direct switch on/off a 100W`120V bulb connected directly to 120V mains !!!!!!!!!!!!!!!!!!!!!!!!!
His logic: "I lose less money nuking (literally) 50 or 60 transistors out of a box of 100 in my shop, and picking the good ones, than have them die on a delivered and paid for Customer's amplifier, maybe in the middle of an important show and worse, maybe burning his speakers"
This is the "brutal test" :


Useless for your BDV or TIP where not even the real ones will handle it, but acceptable (sort of) for BIG transistors in 100's of Watts power amps.

EDIT: Another Enzo truism: if possible, buy straight from Manufactures (say, Peavey), you pay a little more but *they* have the good ones

[ReadyTeddy]

There's been lots of good information posted on this thread. I thought I'd add a few things. C2 keeps DC out of the amp, and C1 keeps RF out of the amp. The network comprising C19, R57, and R58 reduces the amp's DC gain to 1. C15 and C17 are so-called pole-splitting capacitors and add dominant-pole compensation to the amp--without them, the amp would almost surely oscillate. When SW2 is in the voltage position, the amp's circuit is that of a normal SS amp. When switched into the current position, it's output impedance is raised a bit so as to imitate the output impedance of a typical tube amp.

When analyzing amp circuits like this, it's important to note that the various r/c networks within the amp's forward-gain circuitry are not there to modify the amp's closed-loop frequency response, but rather to modify it's open-loop phase, frequency, and gain response so as to ensure a stable closed-loop response under the various speaker impedances it may be connected to.

[Roberto]

*Maybe* your BDV are actually relabelled TIP142/147 ... which is actually good.

They actually are just TIP142/147, not relabelled. I've installed them yesterday evening, but no way to try them... next weekend!

Thank you very much for the tips to start-up the new transistors, really helpful!
I will search a pair of photos of the rack and will update it soon.
Hopefully next weekend I'll start the 8008 up.

[Roberto]

Thanks to you all for the help, the power amp is running perfectly.

After some recent readings, I would like to ask you what do you think about using positive current feedback and negative voltage feedback on that amp, instead of the switch for current/voltage negative feedback.
The idea comes after some readings of Nelson Pass documents on the topic, and some tests done on diyaudio forum.

What do you think?

[Enzo]

Just my opinion, but if

the power amp is running perfectly.

Then I don;t see the need to change it.

[nickb]

Thanks to you all for the help, the power amp is running perfectly.

After some recent readings, I would like to ask you what do you think about using positive current feedback and negative voltage feedback on that amp, instead of the switch for current/voltage negative feedback.
The idea comes after some readings of Nelson Pass documents on the topic, and some tests done on diyaudio forum.

What do you think?

My suspicion is that if you do get it to work it will be boomy and not suitable for the intended purpose of making music with a guitar. I wonder if it might damage the speakers which are not designed as bass drivers too.

[J M Fahey]

Mr Nelson Pass caters to the Hi Fi Audio crowd, which is fine, but we are into rough brutal sweaty Rock sound, different needs.
His ideas tend to reduce distortion, while ours ....

[Leo_Gnardo]

What do you think?

What Juan said ^^^. If you apply some other Nelson Pass ideas, you'll have a rack of filter caps and about 48 output tubes. Great for hi fi, but for us lot not so much.

Funny, it would be a treat to listen on a Pass amplified stereo rig, to recordings made with any and every rock guitar amp. All that bandwidth, WOW!

[Roberto]

Hi Enzo, and thank you for your feedback.
I would totally agree with you if there would't the possibility to get a different sound from it.
So my question is: it works, but does it work like I would like?

[Roberto]

Hi Juan, and than you for your advise and for all the help you already gave me on this thread. I'll be in Bueno Aires for job next week, if I'll have time I would like to visit your shop if possible. The results I've read on other forums about the positive current feedback talks about a different character in distortion, not less distortion in general. More low order harmonics, and this could not be so bad.

[Roberto]

Can I ask you to explain me why it should do and behave like that?
Thanks in advance

[Roberto]

Thanks for your opinion Leo.
Yes, I would like to stop a bit before going OTL.