DIY JTM 45/100 Style Amp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Joined
Aug 7, 2008
Messages
22
Location
Lower Mainland, BC, Canada
I just finished a DIY kind of job on a JTM 45/100. I put a build page up at:

http://www3.telus.net/public/wbrooke/45-100.htm

Lots of pictures, so beware if you're on dialup. The website isn't finished. I hope to add more pictures, more data, and sound clips.


A few pics:

IMG_3352.jpg


IMG_3356.jpg


IMG_3357.jpg


IMG_3358.jpg


IMG_3359.jpg


Really cool amp. I need to start thinking more about compressors, eqs, pres, and other such, though!
 
While we're here, I'm wondering what the advantages/disadvantages of putting the two center taps of the power transformer (both HT and heaters) on the same fuse to ground. In the last photo, on the transformer, you can see I've used a green wire with a yellow stripe on it to connect two of the PT lugs together, and then to ground via the flush mount fuse mounted just below the PT in the picture. The top side is the 375-CT-375 side (blue with white stripe is the common lug), the bottom side is 3.15-CT-3.15 (heaters), and the two CT's are green w/yellow stripe. It works beautifully as is, and works beautifully with the two separated (and running the heater center tap directly to ground without fuse, but fusing the HT CT). Having the heaters fused is nice. Maybe I'd use more fuses over the years, by a small amount? I know my ass is covered so far as protecting the transformers are concerned. I can deal with the odd tube replacement. But technically speaking -- is there a best way to do this? Put that fuse someplace else entirely (there is a separate Mains fuse, the layout is near the bottom of my website)? Hmm. Thanks in advance.
 
> I know my ass is covered so far as protecting the transformers are concerned.

No, it's not.

There are MANY failure modes. As your private message implied with all the "What IF?"s.

The original plan will blow a fuse if the OT primary shorts to ground.

The original plan will probably blow a fuse when a power tube sucks big-time; but not certain. Zero-bias on these tubes, depending on circumstance, could end up near 0.9A RMS in the AC leg.

Which opens another question. Why fuse in the AC side? The very spiky waveform makes fuse rating a very uncertain thing.

A minor point: those are "250V" fuses, you have 500V when the fuse lets go. I'm not concerned, because they are rated for maybe 1,000 Ampere fault current and to fail "safely" with 250V 1,000A of available energy. On the PT secondary you are unlikely to have over 10 Ampere fault current. 200% overvoltage and 99% undercurrent, it will probably fail safely, or safer than an alternative (smoky fire in a room full of drunks with only one exit).

That's all-original in the plan you used, a well-known amp which has served well for decades. So empirically it isn't "wrong".

But other "What IF?"s:

A VERY likely failure mode is rectifier death, especially as B+ approaches 500V because this plan needs 1,000V diodes plus ample safety margin, and >1K diodes are hard to find. Sure "1KV" diodes will stand 1,100V or more. But common wall-outlets get spikes from elevator motors or even bass-amp switch-off. You play the lottery that you do NOT hit a 1,300V jackpot tonight. So an amp may work for months, years, even like my microwave oven for two decades, before a spike comes along which exceeds rating plus safety margin.

Rectifier failure always happens as dead-short. (Actually my oven blew open, but that exception proves the rule.)

If you fuse the CT, and short one diode, the other one flows infinite current half the time. Yes the fuse will blow but the abuse continues. The PT will burn up.

Fuse each rectifier individually. Fuse the B+ before the OT CT. Fuse each tube cathode. Fuse the heater hot leads. And the bias tap... it may be rated 50mA but it will really deliver more current than the HV legs. And of course fuse your 120V primary.

But that path introduces many-many contacts, and contacts are potent failure points. Fuses themselves do fail "for no reason". There are interactions: if one rectifier fuse fails, the other rect and PT carry the load, limping, and could overheat without further fuse-blow. And it looks more like a fusebox than an amplifier.

You want to live in perfect safety, don't use tube amps, don't play in dives, develop a love for accountancy and work from home.

Otherwise, a well-chosen primary fuse, and common sense, is the best path.
 
> both HT and heaters on the same {wire} to ground.

There is no "ground"!!!!!!

There are various things which are connected together for simplicity. This simplicity is so compelling that we never think any other way (except balanced lines). But the connections must be made with thought.

Just because wires go the same place, does not mean they should be the same wire.

I can walk to the mall on this nifty walking path, or zoom in my car. Why have both? Make walkers use the same lane as the cars! Saves pavement!

The PT HV CT should go to the main filter cap "-" lead. There are BIG 120V spikes in that line, and NO compelling reason to flow that crap through chassis. In fact there should be an absolute minimum of conductor in-common between the rectifer side and the load side.

Taking PT HV CT to chassis does give a short to-chassis fault-path for some unlikely PT faults; however good audio "ground" wiring will carry tube PT HV fault current fine, even if it takes the long way.

My re-thought Champ has all the HV stuff "floating". Rect/winding to main caps. A lead comes from the other side of the main-cap "-" node to power-stage common, to driver common, to preamp common (and OT secondary common/NFB return), and only hits chassis at the input jack.

The heater on the other hand... the amplifier will work without heater CT. If the PT winding is done right (dubious) and you route to the socket pins optimally (sadly not possible on the stupid 12AX7 6V pinout), you may not need any grounding on heaters. Arguably an un-grounded heater circuit will keep-working throuh heater-cathode shorts which would burn-up a hard-grounded CT heater circuit. Best practice is probably to provide a low but non-zero ground path: 100R from CT to preamp common, or the same with two 100R resistors when no CT is supplied. "Normal" 50K heater-cathode leakage will be drained-off by 100R; abnormal dead-short from one heater hot lead to ground will make 0.36W of resistor heat and high hum level, but you finish the gig and get paid.

The wall-cord 3rd pin MUST go to chassis as QUICK as possible, and many codes require it to have a dedicated screw. When you get 13 electrical experts on a committee debating "What if?", it is easy to concoct scenarios where a shared screw loosens and HV CT gets injected to wall not chassis, and chassis is levered to HV level. Not real likely, but do be sure you trust that shared screw.

I'm not saying you should rip-up this amp. It is far too nice, and works too good, and there is no compelling "wrongness". Also far "wronger" things have served us very well. (Ever see a 1951 Fender? Or any Kay?)
 
> this plan needs 1,000V diodes plus ample safety margin

OK, I see you've stacked three of the 1N5... 800V parts. Rectifier failure is never impossible, but you do have plenty of margin, far more than most gear.
 
PRR: Thanks for the time and consideration. I think I had better either donate to the board, or send you something by PayPal!

Yes, the six larger diodes are 3A / 1000V. Or at least they were advertised as 1000V, I could believe that 800V is the most a person would want to deal with. The one smaller diode on the preamp board is 1A / 1000V (or it was advertised as such).

B+ in the amp is 495V.

It does work great as is! Unless this changes I don't intend to mess with a working thing. Your points on an amp full enough of fuses to be simply called a fuse box and two routes to the mall are well taken, and pointing out that these old Fenders (or a Kay, and these old Marshalls, too) are nowhere near the apex of engineering (yet they work well for what we ask of them) is not contestable. However, for my own personal education, and in the instance of serious problems with my amp:

-The CT of the heaters, I now am understanding that the CT is not a part of the heater loop that carries current, but rather just a branch. Checking into this, plenty of older amps have heaters with no CT, but the complaint is some hum -- so as a person could take two 100 ohm resistors at some point along the heater chain, and make an artificial center tap to ground. So it is not part of the heater loop that sees a steady current. Or so I think...

-The PT HV CT going to the negative lead of the mains filters -- in an effort to have the rectifier side and the load (signal?) side of the amps share less of a common conductor (chassis?). Hmm. Never even crossed my mind. And I thought I was being a bit clever by separating the mains filter ground from the screens filter ground! :oops:

-As for the heater CT, I could run it, then, through a 100 ohm (1w?) resistor, and head it over to ground at the same point as the rest of the preamp grounds (seeing if I have got all this right)?

-I can put in a ground lug in right beneath the 3 prong socket for a dedicated chassis to IEC connection (as short as possible), if I feel I don't trust a shared grounding screw.

-Although I'm not ever really safe so far as transformer protection goes, I can probably do a bit better than I am now, although it is not necessary.

Your Champ, with minimal ground points, sounds extremely well thought out.

Your help is much appreciated, sir!

Edit: Donation has been sent to Prodigy Pro forums.
 
thousandshirts,
i'd say you are worrying too much, but it is your amp, so you can worry all you want, i suppose.

you seem hell bent on protecting the transformer.
Which one do you hold most dear?
power or output?

i've owned tons of tube amps over the years and never blown either, in any of them, and i'm a "jam the front end with a hot signal and crank the volume to get powertube distortion" kinda guy.
(NON master volume for me)

if you use good components, then a bad power tube will be your culprit more than anything. And there's a mod using a diode(i forget it offhand)
off each power tube to protect the output transformer.

now, no one WANTS to blow transformers, but unless you are using vintage ones, they can all be replaced.

Also, i suppose i have done the 100ohm mod in amps with no center tapped
heaters, but i can't say as i always hear a difference.
When we get our hands on a 20-30 year old amp and there's some hum,
that doesn't mean it was there when it was new. Could be lots of other things. I also used to be the "tight wound heater guy",
then i bought a boutique Vox clone and the guy did just a single
crossing loop between tubes with the heaters, and it was the quietest, hum free AC30 type i'd ever played.......who knew?

either way, that's nice work. I'm building an 18watt tmb clone now using a weber chassis and board, with an ac30 clone next.
 
[quote author="okgb"]that's nuts [ in a good way ] all the attention to detail
tight twisted wires and heat shrink on the pots

rock on ![/quote]

[quote author="justanalogue"]Old skool craftsmanship.

Very impressive!

Analogue.[/quote]

[quote author="MasonAtom"]Beautiful work.

Love the tightly twisted heater wires and the heat shrink on virtually every bit of what would have been exposed wire.[/quote]

Thanks very much for having a look, folks! I appreciate the kind words.
 
[quote author="Gwaggin390"]thousandshirts,
i'd say you are worrying too much, but it is your amp, so you can worry all you want, i suppose.[/quote]

Perhaps I'm being overly inquisitive. But I am certainly not worried. I know that the original JTM 45/100's and JTM 45s up to and around 1967 were wired this way, and I know that the subsequent changes in fuse placement by the "Brain Trust" at Marshall amps ( :razz: ) may be better, but still, they aren't perfect either. Despite having been recorded onto countless good rock albums, and around the world many times on grueling tours of rock duty :shock: I have another amp, actually, wired this way, that has been operating nicely for years. But--I just wanted to know more.

if you use good components, then a bad power tube will be your culprit more than anything. And there's a mod using a diode(i forget it offhand)
off each power tube to protect the output transformer.

Right. Daisy chain three 1N4007's between pins 3 and 8 of your (octal) output tube sockets, banded end of the diodes towards pin 3. I think this idea found its source (at least, in the guitar amp world) in the Musicman amps.

Also, i suppose i have done the 100ohm mod in amps with no center tapped heaters, but i can't say as i always hear a difference.
When we get our hands on a 20-30 year old amp and there's some hum,
that doesn't mean it was there when it was new. Could be lots of other things.

I was really wondering mainly about the center taps, and grounding. Heater hum was just an example. No heater hum issues here. Silent service with this amp.

I also used to be the "tight wound heater guy",
then i bought a boutique Vox clone and the guy did just a single
crossing loop between tubes with the heaters, and it was the quietest, hum free AC30 type i'd ever played.......who knew?

I have done other amps with some looser twists before. As you point out it is not just the heaters or how they are twisted that keeps an amp silent as a sub. But this one was a personal keeper from the start. You know how it is. Sometimes it's easy to get all carried away. Especially if it's for yourself.

either way, that's nice work. I'm building an 18watt tmb clone now using a weber chassis and board, with an ac30 clone next.

Thanks very much sir. Sounds like you might be a fan of the 6BQ5/EL84? If nothing else, I could say that they are a lot more reasonable in price than these KT66. :cry:
 

Latest posts

Back
Top