• The Gear Page Apparel & Merch Shop is Open!

    Based on member demand, The Gear Page is pleased to announce that our Apparel Merch Shop is now open. The shop’s link is in the blue Navigation bar (on the right side), “Shop,” with t-shirts, hats, neck buffs, and stickers to start. Here’s the direct link: www.thegearpageshop.com

    You’ll find exclusive high-quality apparel and merchandise; all items are ethical, sustainably produced, and we will be continuously sourcing and adding new choices. 

    We can ship internationally. All shipping is at cost.


JCM800 No-Sound

epluribus

Member
Messages
9,170
Long story how I got into this, but suffice to say I'm looking at a bud's '84 (IIRC. S/N xxxxxR) Marshall JCM800, 4210 flavor. The tubes and power jewel light up, but it puts out zero sound on either channel.

Simple things I checked so far:

The single input jack has no face nut on it, so the PCB takes all the strain when you plug in. That said, wiggling the jack slightly produces no sound or crackling like a loose connection. The PCB does not appear to have been broken.

Speaker works, incl. the lines and the 1/4" male plug.

Replaced all tubes with some known good ones I have around.

Replaced the fuses.

No sound at Line-Out

No sound at FX send

No sound running a line-level signal into the FX return. BTW, the FX return jack is also missing its face nut, and the jack was dirty and binding. Cleaned it.

No sound at the speaker-out

Speaker is a 15 Ohm Celestion G12M-70, the load selector is set at 16 Ohms. (It's also plugged in.)

The voltage selector is set for 120, US standard.

Good power cord, plugged in.

My test gear is limited to a digital multimeter, incidentally.

I pulled the chassis and drained the caps for a closer look. (I did not pull the PCB.) Doesn't appear to have had any of the several JCM800 kit mods done to it, judging by the color of the visible soldering. There are a couple of joints at the rectifiers and the power tubes with burned flux on 'em--those may not be factory. (Though I'm not sure how to test 'em.) Otherwise, I see no evidence of burned/overheated parts or obvious breakage. Any ideas before I send him to a tech? (He's on a tight budget.) See pix below, if they're of any help.





Anyhoo, thanks for looking. We'll just see where we end up.

--Ray
 

epluribus

Member
Messages
9,170
Do you have voltages througout the amp?
Hey Swarty, 'preciate the reply...

Without metering anything, I note that all the tubes warm up as you might expect. I imagine there are a few places on the chassis that amount to an end-point in the circuit that I could check...suggestions?

--Ray
 

Swarty

Member
Messages
1,131
Start with pins 3 and 4 on the power tubes and then pins 1 & 6 on the preamp tubes. I would expect 450V or so on the power tubes and 100-250V on the preamp tubes. If you've not done this before, be very careful you don't fry yourself or the amp.

You say no sound at fx send and DI. What did you use to test this?

Are you certain the speaker is OK and well connected?
 

epluribus

Member
Messages
9,170
Hi Swarty.

No sound at FX send and DI ... I ran these into a power amp, presuming them to be line-level each.

As for the speaker, I did check it out, but just to eliminate it I patched the speaker-out into an outboard cab.

Frying myself...yeah, I refused for a long time to even pull a chassis until I actually drained caps under the watchful eye of an experienced tech, just in case I was missing something obvious. As for a voltage test protocol, I'm using the technique outlined here:

https://www.thegearpage.net/board/index.php?threads/272560

Thanks again, back to the laboratory...

--Dr. Phibes
 

John Phillips

Member
Messages
13,038
If both the power amp and the preamp are dead when checked separately (as you've done), and the tubes light up, it sounds like an HT failure somewhere near the start of the chain. If the fuse is OK, do the power tubes glow blue - you may have to look inside the little slots in the plates to see it - when you flip the standby to on? If so it would indicate power getting to the plates and screens, and that the failure is further down, at the PI supply or below. If not, it's right at the start - PT, standby switch, rectifier or choke (first two unlikely since it would need a failure in both sides at once, and a failed rectifier usually goes short not open circuit, which will blow the fuse.) Difficult to make any better guess from here at this stage - I can't see anything obviously wrong in there.

BTW, the filter caps - certainly the left-hand one in the pic (it has a little leak) - probably need changing, but that won't be the cause of the problem... it would either hum very loudly or blow fuses if the cap actually failed.
 

phsyconoodler

Member
Messages
4,305
You need to be able to identify basic things inside the amp before proceeding.For example: you said you replaced the fuses.That means BOTH fuses?There is a mains fuse AND an HT fuse.
You also need to be able to safely check the voltages inside a working amplifier without hurting yourself.
If you worry about things like discharging caps before working on an amp,you will have to be doubly careful inside a working amp.
I would suggest getting some more safety and basic information before you proceed.Maybe you could get the same tech who showed you how to drain caps to show you how to check voltages?
 

epluribus

Member
Messages
9,170
You need to be able to identify basic things inside the amp before proceeding.For example: you said you replaced the fuses.That means BOTH fuses?There is a mains fuse AND an HT fuse.
Thanks for the words. I did indeed do both of them, but it's worth asking...always too easy to overlook the obvious. :)

You also need to be able to safely check the voltages inside a working amplifier without hurting yourself.
If you worry about things like discharging caps before working on an amp,you will have to be doubly careful inside a working amp.
I would suggest getting some more safety and basic information before you proceed.Maybe you could get the same tech who showed you how to drain caps to show you how to check voltages?
Did that when we were playing with the caps and many other things. The easiest way to describe the voltage testing technique I'm using was with the link. Agreed, though, live amps are best handled with significant respect. In such cases, I limit myself to doing things I've done under supervision.

BTW, the tech in question is an engineer who has worked on several tube-powered circuits, but never guitar amps. It's fun to work with him as he gets his arms around the whole idea of overdriving. Trouble is, it's exceedingly tough to get his schedule and mine to match, so I don't get to see him much.

Thanks again for the help, this is a learning experience.

--Ray
 

epluribus

Member
Messages
9,170
Hey John.

Figured I'd hear from you somewhere along the way--you're something of a fan of the early JCM800 as I recall. IIRC this is a bit further in the production run than the era I heard about, but it's an interesting amp nonetheless. (First time I've ever played the 4210 version. Unique beast.)

If both the power amp and the preamp are dead when checked separately (as you've done), and the tubes light up, it sounds like an HT failure somewhere near the start of the chain. If the fuse is OK, do the power tubes glow blue - you may have to look inside the little slots in the plates to see it - when you flip the standby to on?
New fuse, gave it the lights out check. Nope, no blue, nice orange glow in all tubes.

If so it would indicate power getting to the plates and screens, and that the failure is further down, at the PI supply or below. If not, it's right at the start - PT, standby switch, rectifier or choke (first two unlikely since it would need a failure in both sides at once, and a failed rectifier usually goes short not open circuit, which will blow the fuse.)
Plot gets thicker...

Swarty said:
Start with pins 3 and 4 on the power tubes and then pins 1 & 6 on the preamp tubes. I would expect 450V or so on the power tubes and 100-250V on the preamp tubes. If you've not done this before, be very careful you don't fry yourself or the amp.
...and thicker. I measured the voltages and got very strange (low-to-zero) results. So strange in fact that I re-checked the pinout diagrams and tested another (known) amp just to be sure I wasn't going bonkers. What gives?

John Phillips said:
Difficult to make any better guess from here at this stage - I can't see anything obviously wrong in there.

BTW, the filter caps - certainly the left-hand one in the pic (it has a little leak) - probably need changing, but that won't be the cause of the problem... it would either hum very loudly or blow fuses if the cap actually failed.
Power failure near the front, wierd voltmeter readings, new fuses, new tubes, hmp!... Light dawns. :jo

How did I miss it? Seems that, systematically trying one thing at a time, I replaced the fuses before the power tubes. But what if the old tubes weren't merely dead, but shorted? They'd eat my new HT fuse. Took 'em out to check 'em--the new HT fuse looked good in the glass but tested dead. Answer: replace the fuses and the tubes all at the same time. Bang!...sound! D'oh!

Hokay, cleaned the scratchy pots and the jacks. Yes indeed the caps could stand to be replaced--they all appear to be OEM. How you saw that leak in those fuzzy photos is beyond me... Rewired a frayed and nasty speaker line splice and it plays reasonably well. (BTW, the owner had the speaker load switch set to 8 ohms. I set it at 16.) Still needs biased but a Bias-Rite is in order first.

My thanks to all for the help and the concern. It's always fun learning about these things. My bud will be thrilled to have his amp back.

--Ray

Did I read somewhere I can set the bias with a voltmeter while I have the chassis out? Hm...
 

John Phillips

Member
Messages
13,038
Hey John.

Figured I'd hear from you somewhere along the way--you're something of a fan of the early JCM800 as I recall. IIRC this is a bit further in the production run than the era I heard about, but it's an interesting amp nonetheless.
No, this is the very first Split-Channel circuit. It was changed in '84 so it must have been one of the last ones of this version. This is actually the version I like, but popular opinion says is less good! It's easily identifiable by the dual-gang MV pot, which is a post-PI MV.

Power failure near the front, wierd voltmeter readings, new fuses, new tubes, hmp!... Light dawns. :jo

How did I miss it? Seems that, systematically trying one thing at a time, I replaced the fuses before the power tubes. But what if the old tubes weren't merely dead, but shorted? They'd eat my new HT fuse. Took 'em out to check 'em--the new HT fuse looked good in the glass but tested dead. Answer: replace the fuses and the tubes all at the same time. Bang!...sound! D'oh!
Ha! Typical JCM800. Apart from tube failures and the odd dead power-switch light :), they just almost never break... they're as close to bombproof as any production amp ever made.

The pots do often go scratchy though, worse than other amps (not sure why) but they almost always clean up without needing to be replaced.

Did I read somewhere I can set the bias with a voltmeter while I have the chassis out?
Yes, but it's a little dangerous.

You need to first measure the DC resistance of each half of the OT primary - from the HT fuseholder to pin 3 on each power tube socket - with the amp off and caps discharged. Write it down, and they are usually different by 1 or 2 ohms, that's normal. You should get around 35-40 ohms in a 50W Marshall of this age.

(To discharge the caps, simply leave the amp powered up but on standby for a few minutes, BTW - the standby switch is upstream of the rectifier on these amps, and the caps will discharge through the tubes.)

Then with the amp running, measure the voltage drop at the same points - you're looking in the 1V range, but be careful! That doesn't mean it's safe, because you're measuring the difference between two voltages which are both about 450V above ground :). It's best to clip one meter lead to the HT fuseholder before you start, and be very careful with the probe while working with one hand only. If you slip and the probe bridges pin 3 and pin 2, you'll get fireworks.

Now, divide the voltage measurement by the resistance you found earlier for each side, and you have the current.

This is the 'best' and most accurate method, and is a true measure of plate current, without including the screen current as the bias meter/1-ohm resistor methods do (this can cause quite a significant error with EL34s, which have quite high screen current, but at least it's in the safer direction).

Don't bias the amp up to 70%, keep it to under 60% (15W with EL34s) - it will sound better and stress the tubes much less hard when you turn it up. These amps put out way over 50W when pushed hard, and if you bias at 70% you'll probably get screen failures when you crank it.


FWIW, if you want to get the best out of this amp, replace the G12M-70. It's a poor speaker in this amp (although it is stock) - mushy bass, nasal mids, no top-end, and not very sensitive. See if you can find an old G12-80 (the speaker it was originally designed for) or if not, a Classic Lead 80. You'll get much more volume, about twice the bass response, clearer/more solid mids and more sparkle. If you don't want more volume, try a G12-65.

Another improvement you can make is to remove the Line Out level control (just reconnect the pink wire directly to the line out jack) and reuse the pot - it's the correct value - as a Presence control instead of the fixed value it has.
 

epluribus

Member
Messages
9,170
...

Yes, but it's a little dangerous... (Referring to using a multimeter to measure and set bias--ed.)
Well, that's me, Nick Danger, Third Eye... :BluesBros

You need to first measure the DC resistance of each half of the OT primary - from the HT fuseholder to pin 3 on each power tube socket - with the amp off and caps discharged. Write it down, and they are usually different by 1 or 2 ohms, that's normal. You should get around 35-40 ohms in a 50W Marshall of this age.
Whoa, way cool! Marshall has the pinouts on the bottom of the tube sockets numbered. Hokay, so far so good, they came in at 47.2 ohm on V7 and 45.1 ohm on V6.

(To discharge the caps, simply leave the amp powered up but on standby for a few minutes, BTW - the standby switch is upstream of the rectifier on these amps, and the caps will discharge through the tubes.)
Nice feature for us tinkerers. Gonna have to see how this was done on the schemo and do my other amps. Or maybe just bleed resistors...

Then with the amp running, measure the voltage drop at the same points - you're looking in the 1V range, but be careful!
Uh oh...strange readings. .6v at V7 (47.2 ohm on the OT primary) and 2.3v at V6 (45.1 ohm). Hm...

Re-checked the pin numbers, cleaned the pin sockets with DeOxit, retensioned 'em all, and flip-flopped the tubes just to recheck.

Did a voltage check on a 9v flashlight just to be sure I wasn't measuring wrong with the DMM. :rolleyes:

Cleaned small spots on the tube socket lugs to insure good clean contact, even though the resistance readings indicated I had good circuits.

Same same on both channels, BTW.

Re-checked all the resistance in the primary and voltage numbers after that and got the same results. Dang. Hafta look at the schemo and picture what might have drifted. ...Or what Yours Truly may have done wrong...

That doesn't mean it's safe, because you're measuring the difference between two voltages which are both about 450V above ground :). It's best to clip one meter lead to the HT fuseholder before you start, and be very careful with the probe while working with one hand only. If you slip and the probe bridges pin 3 and pin 2, you'll get fireworks.

Now, divide the voltage measurement by the resistance you found earlier for each side, and you have the current.

This is the 'best' and most accurate method, and is a true measure of plate current, without including the screen current as the bias meter/1-ohm resistor methods do (this can cause quite a significant error with EL34s, which have quite high screen current, but at least it's in the safer direction).
This is a cool technique for sure. Took my time, nice clean work area, clear reaches, good lighting, checked my clipped lead for solid contact before proceeding, took a practice stab or two so there'd be no surprises--and one hand in my pocket! Worked great.

BTW, why don't you have to measure screen current this way? Or am I reading you wrong?

Don't bias the amp up to 70%, keep it to under 60% (15W with EL34s) - it will sound better and stress the tubes much less hard when you turn it up. These amps put out way over 50W when pushed hard, and if you bias at 70% you'll probably get screen failures when you crank it.
Very much appreciated. Experience to temper the numbers--always a good thing.

FWIW, if you want to get the best out of this amp, replace the G12M-70. It's a poor speaker in this amp (although it is stock) - mushy bass, nasal mids, no top-end, and not very sensitive. See if you can find an old G12-80 (the speaker it was originally designed for) or if not, a Classic Lead 80. You'll get much more volume, about twice the bass response, clearer/more solid mids and more sparkle. If you don't want more volume, try a G12-65.
I hear ya, I'm not wild about this speaker either. Not many of my other amps like it, and yours is a pretty good description of how it sounds. Wonder what it was designed for? I'll mention it to the owner for sure.

Another improvement you can make is to remove the Line Out level control (just reconnect the pink wire directly to the line out jack) and reuse the pot - it's the correct value - as a Presence control instead of the fixed value it has.
Once I learned how to use them, I fell in love with presence controls. I'll pass this on as well. Frankly, I've found little use for most line-outs--most of 'em just bypass too much of the cool stuff the latter part of the circuit does.

And so the plot thickens further. Even with the puzzle it's terrific to make so much progress thus far. So it's off to see if I'm missing the obvious again...

Sure do 'preciate the help guys!

:BEER

--Ray
 

epluribus

Member
Messages
9,170
Start with pins 3 and 4 on the power tubes and then pins 1 & 6 on the preamp tubes. I would expect 450V or so on the power tubes and 100-250V on the preamp tubes...
Hey Swarty.

My apologies, I just noticed I hadn't replied to this part of your post. Your numbers gave me a good ballpark to work with, and they do indeed check out as you noted above. Er...I should say after I R&R'd both the tubes and the fuses together. :) (When I first checked this it was a big 'ol clue somethin' was very wrong...)

--Ray
 

John Phillips

Member
Messages
13,038
Whoa, way cool! Marshall has the pinouts on the bottom of the tube sockets numbered. Hokay, so far so good, they came in at 47.2 ohm on V7 and 45.1 ohm on V6.
A little higher than I said, but I may have accidentally told you the values for a JMP... getting old ;).

Nice feature for us tinkerers. Gonna have to see how this was done on the schemo and do my other amps.
Not all amps are like this, be careful. With some doing this leaves the first filter stage charged.

Uh oh...strange readings. .6v at V7 (47.2 ohm on the OT primary) and 2.3v at V6 (45.1 ohm). Hm...

Re-checked the pin numbers, cleaned the pin sockets with DeOxit, retensioned 'em all, and flip-flopped the tubes just to recheck.
It's most likely the screen resistor on the side with the low reading (measure the resistance, and the voltage on pin 4). Or, it could be a very slight leak on the PI coupling cap on the side with the high reading, but it would have to be very slight. Check the bias voltages on pin 5, although that small a difference might not show. (And metering this pin can sometimes make the amp unstable too.)

BTW, why don't you have to measure screen current this way? Or am I reading you wrong?
The screen current does not pass through the OT, so what you're measuring is the true plate current only. It's also more accurate (as well as a lot safer) than the transformer-shunt method - which allows you to read the current directly, but is crude and dangerous because it disturbs the circuit, relies on the meter having zero resistance (which it doesn't, quite), and if anything goes wrong will directly short the B+ through the meter, which will probably damage it and/or the amp - and maybe the person holding the meter probes... I would never recommend that method.

The normal bias probe or 1-ohm resistor in the cathode connection method includes the screen current automatically, because the screen current flows to the cathode too - the problem is that it isn't known, and can vary between about 1 and 5mA, even among the same tube type. Since this can be up to 20% of the plate current, it renders the method inherently inaccurate - there is no point in trying to aim for 'exactly' 70%, or 60%, or whatever. Most people setting current by this method don't understand this. The only good thing is that the true plate current is always less than the reading, so by setting to "70%", they're more likely setting to about 60% :).

I'm not wild about this speaker either. Not many of my other amps like it, and yours is a pretty good description of how it sounds. Wonder what it was designed for?
I don't know. It seemed to be a replacement for the far superior G12-65 (I have no idea why) but didn't last long and was very soon replaced by the G12T-75 (the early ones of which are much different from the current type and much better).

Much of Celestion's speaker development was driven by Marshall though - their largest customer - so maybe someone at Marshall actually liked it. They sometimes do strange things with speakers, even now... I've just pulled another stock G12E-50 out of a DSL201 combo and replaced it with a V30, because the G12E-50 is the absolute single worst 12" speaker I have ever heard... ever :). The amp sounds so much better with any other speaker, especially a G12M-25 or V30, that it's hard to believe it wasn't originally designed for one. The 4210 was definitely designed for the G12-80, it's shown on the original schematic, and very early examples have it.
 

epluribus

Member
Messages
9,170
A little higher than I said, but I may have accidentally told you the values for a JMP... getting old ;).
Seems to be going around... :)

It's most likely the screen resistor on the side with the low reading (measure the resistance, and the voltage on pin 4). Or, it could be a very slight leak on the PI coupling cap on the side with the high reading, but it would have to be very slight. Check the bias voltages on pin 5, although that small a difference might not show. (And metering this pin can sometimes make the amp unstable too.)
Fosse-natin'. The schemo shows the US-spec 1K 5W resistor connected to pin 4, but what Marshall did was connect to the unused lug at pin 6 and solder the resistor between there and the lug on pin 4, probably a somewhat neater, easier, and more durable arrangement. (BTW, just to be sure we're looking at the same resistor here, I measured the big black ones soldered to the bottom of the V6 and V7 tube sockets in the pictures, from pins 6 to 4. The specs are printed on the side as opposed to the stripe indicators. They look like power resistors.)

So I measured the resistance three ways: First, just as we did with pin 3, with one lead clipped to the HT fuse holder, and one at pin 4. The resistance at V6 was 1.085, but V7 shows no connection--as if the resistor isn't conducting at all. Hm...

So then I measured the resistance between pin 6 and the HT fuse holder, without either resistor in line, just to check the rest of the circuit. About 112 ohms each. Good, I think.

Finally, I measured across the resistor itself, from pin 4 to pin 6, just cuz I wasn't convinced V7 was for real. V6...965 ohms. V7...open circuit, no measurable conductance through the resistor, even on the leads. (And it doesn't appear any different than the other one.) Just to be sure, I measured the conductance of the various test points themselves to rule out contaminants or insulating coating. Good contacts, all. Hm...

The voltage on pin 4 was 7.9v at V6 (1.085 ohm) and 147.1 at V7 (the one with the non-conducting resistor.) So where's V7's screen getting all that voltage? Or have I simply missed something very basic here?

Anyhoo, on to pin 5. 482V at V6, 489 at V7. At least that was easy.

Curiouser and curiouser...

The screen current does not pass through the OT, so what you're measuring is the true plate current only. It's also more accurate (as well as a lot safer) than the transformer-shunt method - which allows you to read the current directly, but is crude and dangerous because it disturbs the circuit, relies on the meter having zero resistance (which it doesn't, quite), and if anything goes wrong will directly short the B+ through the meter, which will probably damage it and/or the amp - and maybe the person holding the meter probes... I would never recommend that method.
Thanks for all that, I've read about that technique and never gotten real sold on it. I'm liking the simplicity of this method, as it keeps the variables down to a dull roar and kind of appeals to common sense. And the added safety is always a good thing...

I'll be interested to see where this amp's owner goes as far as the speaker issue is concerned. (Along with a trip to a proper bench test.) I'm kinda hopin' he'll stick some dough into a tweak or two--I think it's basically a very nice amp that needs some TLC. I'm having fun with it at least. :)

Thanks again for the terrific help.

--Ray
 

macphly

Member
Messages
49
Nice thread guys...I'm just beginning to learn about this stuff and am happy to say I understand a lot of what you're talking about!

Just wanted to interject that you are SO spot on about speakers. I've got a couple of JCM800s and own pairs of those old speakers: G12-65, G12-80, and G12T75. I also have a pair of English-made Classic Lead 80s. The earlier speakers ROCK! The G12-80 in particular has the tone of a greenback to my ear, but tight bottom and higher power handling. IMO, the Classic Lead is very lifeless by comparison.

FWIW: Weber does a really nice version of the 65.
 

John Phillips

Member
Messages
13,038
The schemo shows the US-spec 1K 5W resistor connected to pin 4, but what Marshall did was connect to the unused lug at pin 6 and solder the resistor between there and the lug on pin 4, probably a somewhat neater, easier, and more durable arrangement. (BTW, just to be sure we're looking at the same resistor here, I measured the big black ones soldered to the bottom of the V6 and V7 tube sockets in the pictures, from pins 6 to 4. The specs are printed on the side as opposed to the stripe indicators. They look like power resistors.)
Yes. That's how it's done on most amps with individually-wired power tube sockets. Keeping the resistor that close to the tube is good for stability reasons too.

So I measured the resistance three ways: First, just as we did with pin 3, with one lead clipped to the HT fuse holder, and one at pin 4. The resistance at V6 was 1.085, but V7 shows no connection--as if the resistor isn't conducting at all.
As expected :).

So then I measured the resistance between pin 6 and the HT fuse holder, without either resistor in line, just to check the rest of the circuit. About 112 ohms each. Good, I think.
Yes, it means the choke is fine.

Finally, I measured across the resistor itself, from pin 4 to pin 6, just cuz I wasn't convinced V7 was for real. V6...965 ohms. V7...open circuit, no measurable conductance through the resistor, even on the leads.
Yes, blown screen resistor. This will have happened when the old tube shorted most likely - even though they're a fairly tough 5W wirewound type, the momentary power dissipation when the tube shorts can be up to a couple of hundred watts, and it's a race between the resistor and the HT fuse as to which blows first. I would replace both of them with even tougher hard-glazed ceramic-cased wirewounds - I have come across failed ones of the type in your amp several times, but never the glazed ones, their short-term overload rating is so high that the fuse should always go first. (The right ones look like a very slightly lens-shaped shiny dark green or grey cylinder, BTW.)

The voltage on pin 4 was 7.9v at V6 (1.085 ohm) and 147.1 at V7 (the one with the non-conducting resistor.) So where's V7's screen getting all that voltage? Or have I simply missed something very basic here?
You're measuring down voltages from the B+ supply. When you put the meter on the screen connection, a very small current flows via the meter itself, which allows a voltage drop - apparently quite large, because the resistance of the meter is. If you meter from ground up to these two points, you should find very nearly the full B+ on V6 and zero (or very close) on V7.

Anyhoo, on to pin 5. 482V at V6, 489 at V7.
Measuring from the HT fuseholder, yes :).

Sorry, I should have said that doing that was only for checking the bias. Normally, voltages are measured from ground (chassis). The voltages on pin 5 should be in the -30 to -45V range for EL34s. Check them again from ground, after you've replaced the screen resistor(s). They will not match while V7's resistor is missing because the tube is not conducting properly.

I'll be interested to see where this amp's owner goes as far as the speaker issue is concerned. (Along with a trip to a proper bench test.) I'm kinda hopin' he'll stick some dough into a tweak or two--I think it's basically a very nice amp that needs some TLC. I'm having fun with it at least. :)
I agree, it was always one of my favorite Marshalls. I had one about fifteen years ago, then sold it for some reason, and when I eventually got another it was the later version that popular (internet) wisdom says is better. I have no idea why - it never sounded even close to as good. It's the same amp in external appearance only, really.
 

epluribus

Member
Messages
9,170
Epluribus said:
So I measured the resistance three ways: First, just as we did with pin 3, with one lead clipped to the HT fuse holder, and one at pin 4. The resistance at V6 was 1.085, but V7 shows no connection--as if the resistor isn't conducting at all.
As expected :).
Thank you, you've validated my sanity. ...Well, not to jump to conclusions, but I had to look at that measurement about three times just to be sure I wasn't once again missing the obvious. :rolleyes:

I'll see what I can find online in the way of a replacement like you recommend--makes sense to me.

Yes, it means the choke is fine.
Good. Kinda what I gathered from the schemo.

You're measuring down voltages from the B+ supply. When you put the meter on the screen connection, a very small current flows via the meter itself, which allows a voltage drop - apparently quite large, because the resistance of the meter is. If you meter from ground up to these two points, you should find very nearly the full B+ on V6 and zero (or very close) on V7.
Tried that, you're right. The difference--when measured up or down--had me goin'.

Sorry, I should have said that doing that was only for checking the bias. Normally, voltages are measured from ground (chassis). The voltages on pin 5 should be in the -30 to -45V range for EL34s. Check them again from ground, after you've replaced the screen resistor(s). They will not match while V7's resistor is missing because the tube is not conducting properly.
Great, I'll do that. 'Preciate the numbers--nice to know my arithmetic isn't out to lunch either. :)

I agree, it was always one of my favorite Marshalls. I had one about fifteen years ago, then sold it for some reason, and when I eventually got another it was the later version that popular (internet) wisdom says is better. I have no idea why - it never sounded even close to as good. It's the same amp in external appearance only, really.
I'm finding that strange as I get to know the Marshall line better. The workings of their amps could change drastically even though they sported the same name and appearance. Gotta check the circuit sticker. Fender does that too, but it strikes me it's much more pronounced in the Marshall line.

Anyhoo, let's just say I'm anxious to play this thing once it's tuned up properly. I've played several sound-alikes, but I've always wanted to take the real thing for a spin sometime. The resistor should be here Tuesday or Wednesday, with a bit of luck. I'll post an update.

This has been a very steep and terrifically interesting learning curve. Thanks tons for all the help--it's been huge.

--Ray
 

epluribus

Member
Messages
9,170
FWIW: Weber does a really nice version of the 65.
Hey Macphly, thanks for the tip. I'm hopin' I can talk the owner into giving this amp the TLC it deserves, incl a better speaker. Hadn't checked the Weber boards yet, or e-mailed for a recommendation, but that was in the works. Kinda partial to Webers myself.

--Ray
 

macphly

Member
Messages
49
I have owned a 4210 as well...had one when they were new in the early '80s...while I do like the tone of the G12-65 a LOT, I found they were a little loose on the bottom in an open-back, 112 config. Given the speakers I have, and acknowledging that I like tight low end, I'd go with the G12-80 or similar (ceramic, high power handling, large dust cap, Kurt Mueller cone) in that amp.

The Weber 1265 I have is in a HK closed back, ported 112, and it sounds GREAT in that cabinet.
 

epluribus

Member
Messages
9,170
The Latest:

Got the parts in Thursday, stuck new resistors in today after a biz trip.

As for the blown screen resistor on the power tubes, the schemo calls for 1K 5W wirewounds. I ordered both the 5W flavor and some 10Ws, (Similar to the ones you mentioned, John.) so I stuck the 10 watt size on both tube sockets, soldered right at the pin lugs like the old ones, and replaced both just so they'd match. They bench well within 1%, while the old ones were zero and 965 ohms respectively.

However, after (natch! :)) I did that, I found an article at TubeFreak or Kcanos, IIRC, advising the use of the 5W size, soldered to the socket lugs as I did, but with the shortest possible leads. The 10W size, being a bit bigger, have somewhat longer leads just of physical necessity. I tucked the resistors in neatly, and they clear all the other components/connections very nicely, but perhaps the lead length or the higher fail point may still be an issue.

So, on to our elusive measurements...

Tube choice: The schemo calls for 6550s on the power tubes, and the amp had old US RCAs 6550s in it, so I stuck a pair of Sovtek 6550WEs in there. (Had 'em laying around. WE's are the cheap ones, IIRC, but they're cool tubes IMHO. The owner may want to roll his own.)

Resistance in the OT primaries is 47.2 ohm at V7 and 45.1 ohm at V6.

Voltage drop measured from the HT fuseholder to pin 3 is 2.11 at V7 and 2.02 at V6. Dividing E by R, the plate current works out to 44.7 ma at V7 and 45.1 ma at V6. Kinda cold?

Screen voltage, measured the same way at pin 4, is 5.8v at V7 and 6v at V6.

Grid voltage, measured from the chassis, is 49v for both tubes.

Bias: haven't tweaked it yet. I find varying stats for max plate current and dissipation for 6550s in an A/B push-pull config, and I'm not confident of the specs for this amp. Additionally, Aiken mentions the notion that such stats vary by the amp, owing to several variables, and that bias should be set based on the amp mfgr's service data. Soooo, Question...where do you go to get the best specs and info on setting up tubes in an amp? And that said, what would you recommend for the 4210 specifically?

As always, thanks once again for all your help and interest, everyone. Is it too nerdy to say that this has been a wonderfully intriguing project?

:BluesBros

--Ray
 




Trending Topics

Top