Speaker winding must be grounded to chassis. This is an important safety issue.
All good fortune,
Chris
All good fortune,
Chris
says who?
Its something other people came up with.
Besides that the only excuse someone made up for it, it still didn't work when a stage got flooded.
Secondary windings are only separated from primaries, which sit at B+, by some enamel insulation, which has some capacitance. Best case, the secondary is at several hundred pF from B+, and worst case, a failure of insulation, is at B+.
This is ridiculously dangerous. Just don't do it.
Chris
This is ridiculously dangerous. Just don't do it.
Chris
🤔 ok,
Well, I guess lets look..
You see, when I wind a transformer I do a style called interleave. Let me show you a picture:
The top photo is an output transformer with the layers separated with Mylar laminated paper (PMP) for approx 2Kv insulation (4layers). 800V Hi pot tested
But I'm thinking of trying this new stuff that higher temp, approx10Kv/ layer (DMD) and go down to 2-3 layers.
Bottom Left shows the winding layout for a typical power supply, This is a power transformer.
Bottom Left show the typical interleave wound output transformer. This is a signal transformer.
I guess if they build with a power transformer style, I could see it having issues later that way.
Different standards have different requirements for safety. Formal compliance requires the standard to be defined, and then a whole raft of supporting docs and tests. For example, although the number of layers and the type of insulation tape may be identified, it is also worth identifying the winding margins and margin tape (if used, or otherwise how a margin is maintained), and how pollution is defined or managed, and the minimum creepage and clearance distances used. Also output transformers may have well in excess of an 800V working voltage, and require something like 2-3kV hipot testing.
Another safety aspect is that output transformer secondary side ac voltages can exceed various safety standard levels for public access/touching - certainly an issue for PA amps, or say 16 ohm outputs of higher wattage capability.
Another safety aspect is that output transformer secondary side ac voltages can exceed various safety standard levels for public access/touching - certainly an issue for PA amps, or say 16 ohm outputs of higher wattage capability.
I was just stating that I hi pot test my output transformers at 800V that I make. Most only test theirs to 500V.
.
So does speaker outputs in those 5KW amps. What about 100V speaker line?
I think binding post are a safety hazard. At least Nuetrik came up with the speak-on to replace it.
Threshold of skin is like 35V. How many audio amps produce more voltage than that?
DIY users implement their own tests - and hopefully tag their equipment as home made so that any new user (in years or decades to come when the maker has no control over it) doesn't naively consider the equipment as safe.
My concern would be about posts that portray a level of safety, or portray a level of naivety about safety matters, such as 'says who?'.
My concern would be about posts that portray a level of safety, or portray a level of naivety about safety matters, such as 'says who?'.
I think someone's safety grounding on the output transformer is made up, and it sounds like it could be possibly you.
Because you don't have to ground the speaker winding, until its is used in an unbalanced feedback circuit.
Actually grounding the transformer's secondary is bad because you are giving the winding a voltage reference.
BTW, I'm not a DIY-er when it comes to any of this, and I don't care what you think.
thisusername, if you are keen to get some insight into what present day manufacturers of audio equipment have to comply with to sell product to the public then there are some overviews of related standards on-line (given that actual standards cost money and are quite onerous to wade through). One link that seems appropriate is: https://www.seaward.com/gb/news/vie...ation-technology-equipment-part-1-iec62368-1/.
You may also find it insightful to appreciate how floating voltages are considered in safety standards when it comes to defining hazardous voltage levels.
You may also find it insightful to appreciate how floating voltages are considered in safety standards when it comes to defining hazardous voltage levels.
You'll find that many members of diyAudio are Safety Sallys, because we take a lot of responsibilty for the safety of children, pets, other loved ones, strangers who might encounter stuff we've made or repaired, or even just discussed here in public. Nobody gives RatsAssOne about my skunky butt or yours, but electrical safety is for lots of other people. There's no need to get defensive about a technical discussion, and there's lots of reason to counter unsafe practices. Especially if they have zero reason to be published. We're not talking about test pilots needing to risk their lives; it's just audio.
All good fortune,
Chris
All good fortune,
Chris
I'm pointing out its not really a 'saftey' or 'Protection' circuit to effectively do that function. It would need some sort of off current fault sense circuit.
Never had one yet get rejected from UL.
Btw, you cant do that on balanced line level tube gear. That would defeat fully balanced operation.
Failure of B+ to the low resistance secondary, if it's grounded, should be plenty to open a mains fuse. If not, there's a big fire hazard. Balanced output secondaries can be, and routinely are, grounded at a center tap. Again, there is no reason to float secondaries in a valve amplifier, and an obvious danger. Why argue about this?
All good fortune,
Chris
All good fortune,
Chris
Board mostly, but noticed this somewhat undocumented change that isn't really that good to use ever. I would come up with some other way. Never had transformer failures. Never heard of center tapped balanced used for grounding, just phantom power and that is always been a non referenced to ground.
Last edited:
A lot of old radios from the early days just left the heaters floating and added a capacitor (like 0.1uF) to ground on the last heater. So the heater reference finds its own level. Is this such a bad way to do it if there are no complexities where one tube has a much higher cathode voltage than another? (Which I suppose covers a lot of tube amp cases with concertina phase splitters, cascade, etc).
Vintage gear has a wide range of interesting aspects. Dynaco ST70 and Mk VI amps also had a common floating heater winding (capacitor bypassed to gnd) with it effectively sitting at an elevated and bypassed DC level due to the multiple high resistance connections through each valve’s Rh-k connection. In those amps, some valve cathodes are near to zero volts, whereas others are elevated, and so the heater supply will sit somewhere between the range of cathode voltages.
There is an emerging use of output transformer secondaries as primary-side cathode-coupled windings - to enhance local feedback in the output stage. That can work quite well, but requires the relevant secondary windings to be well balanced at HF. (https://audiokarma.org/forums/index.php?threads/untapped-potential-heaths-w-5m.1042474/page-4)
There is an emerging use of output transformer secondaries as primary-side cathode-coupled windings - to enhance local feedback in the output stage. That can work quite well, but requires the relevant secondary windings to be well balanced at HF. (https://audiokarma.org/forums/index.php?threads/untapped-potential-heaths-w-5m.1042474/page-4)
Those kit amps from them and Knight I don't consider commercial products. Looking at the Dynaco original schematic, that is crazy. I've seen that before, and I end up ripping out the power transformer because someone did that because the secondary had a short in it and it was someone's fix at the moment. Most designs, I elevate the heaters about 1/3 to 1/5 of B+ with a voltage divider, sometimes bypassing one or the other resistor. Because grounding the center tap usually adds it into the signal.
Last edited:
I gave been building and repairing guitar and HiFi amps for a long time. Much of that time was spent in the heat and humidity of South Florida. Most of the power and output transformers made in the tube era used paper for the insulating layers. Some are still built this way today due to its low dielectric constant which reduces the parasitic capacitance. Paper will eventually absorb some moisture. It is not uncommon to find an old OPT with a measurable leakage between the primary and secondary. In my past work I have run across two OPT's with a direct short between the primary and secondary windings. Both were in commercially built guitar amps and both resulted in an inoperative amp BECAUSE the secondary was grounded, resulting in a blown fuse or fried rectifier tube.
More recently there have been two BRAND NEW Hammond power transformers discovered here on diyAudio that had a direct short between the 6.3 volt heater winding and the 5.0 volt heater winding. If this kind of short can occur in a brand new Hammond power transformer, it can also occur in a Hammond OPT, though I have never seen one.
Leaving the speaker winding floating does present a possibility for a shock hazard to a user, especially in a guitar amp situation where 1/4 inch jacks are often used for both the guitar and speaker connections. There is no reason NOT to ground one side of the OPT secondary (or a tap) except for a few extra pF in stray capacitance to ground reflected on the primary. Everything I build will have a path from the speaker circuit to ground capable of blowing the fuse or otherwise shutting down the amp should a short occur. Note that some push pull amps that use balanced GNFB ground the 4 ohm tap in a 0-4-8-16 ohm secondary.
I know, that is why I wanted to make this bait thread so I can find you. I heard you were hanging up the soldering iron.