DIY Audio Projects Forum

Hi, please PM me, I have some thoughts on this that could easily tie up the thread more than necessary.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

I've been enjoying learning about this tube amp. Many, many years ago I was quite adept at constructing electronics and as my main audio amp died hard (after only 24 years) I'm giving it a go. I never did any work with tubes. I don't fully understand the effect of the capacitance rating on the cap between the ground bus and the chasis. Also, as far as the line filter, would the filter from the KT88 project suffice? Yea I know I have a lot to learn but I can, at least, safely handle electricity.

Thanks

Hi, I presume that the cap you are referring to is the type X2. I also have to assume that your electrical code (don't know what country you are in) allows this type of connection. In the US external metal parts of equipment must either be double isolated (like some power tools) or have the metal parts connected to an earth ground. This is done as a protective measure in case of internal faults. Audio gear (and lots of other stuff) is sensitive to external noise sources and to prevent pick up of the noise a metal case is often used. A second problem is that if you actually use the case as a signal ground, there is a great tendency for it to develop alternate ground loops and cause hum. The solution I (and many other use) is to keep the signal and case grounds separate. Unfortunately if they are completely isolated there are again noise and hum problems. So the solution is to connect the two grounds at a single point through a capacitor and resistor in parallel. I have seen designs that use only cap and ones that use only resistors. I find that using both is better. The resistor is generally in the range of 50 to 150 ohms and the cap from .047 to .22uf. I use both .1 and .2 values. Which one I choose is based on noise measurements. I use which ever is quieter. The cap is a special type called an X2. it is specifically designed for this application as it may have to withstand full AC mains voltage in the event of a fault and also must be self healing. Several companies make them and they are not costly.

The AC mains line filters are built into the IEC entry connector. There are numerous values of the individual components inside them. The filters are also available as a separate component. Select one based on the current demand. I use ones rated at 5 amps AC minimum. There is no particular advantage to using ones that are much higher than the expected demand.

Hope this helps
Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Thanks for your reply. Your (extremely clear) expanation is what I figured was going on. I was looking for conformation as I'm re-learning a lot of electonics. It will probably take me a while but I will post my experience building your creation.

I'm in Oregon, USA.

Glad to help. I didn't know where you were and some countries have different electrical codes and might not require case grounding or perhaps have different specs on the parts needed. I try to design for a world wide audience, but it is not always possible. I do rely on several international contacts for input on projects - particularly the power supplies and interconnects.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

This is something that has confused me for some time - glad it came up here. I understand all of the above explanation - it's very concise and clear and it makes sense. However, there are two unanswered questions in my mind concerning this. The first is why not simply tie the grounds together? Don't isolate the case, but use ONLY ONE WIRE to tie the case at only one point to the signal and incoming ground so there are no ground loops in the case - use a star grounding system and do NOT use the case as a signal ground - only tie the case to the signal ground at a single point to prevent ground loops in the case.

The second question is, if the case shorts to high voltage and is only grounded through a resistor of 100 ohms or so, how will the circuit breaker trip? It would take 1,500 volts on the chassis to put 15 amps through 100 ohms - barely enough to trip a 15 amp overcurrent breaker in a typical home in the US based on our code. Any less voltage and the case sits there hot at 120 VAC, does it not? Say the case shorts to the incoming 120 VAC line. I=V/R or in this case 1.2 amps flows through the 100 ohm resistor. It just sits there indefinitley at 120 VAC??? It seems to me the resistor effectively prevents overcurrent protective devices on the branch circuit from operating in the event of a direct internal fault to the case of the incoming supply line as well as the internal fuse on the line. What the heck am I missing? What's the purpose of the resistor and cap instead of a short - why won't a hard wire accomplish the same thing without the disadvantages above and without creating loops in the case? (OK, four questions, sorry!). I didn't do the calcs, but the caps don't seem large enough to draw the reactive current required to trip breakers either.

Hope the questions make sense. Thanks.

_________________
Stephen

Hi Bruce, great explanation. Are there any requirements / rating for the resistor?
Cheers

_________________
[ DIY Mains AC Power Cable Cord ] - [ Gobo LM1875 Amp Kit ] - [ Tang Band D4-1 Horn Speaker Kit ] - [ Monoblock Push-Pull KT88 Tube Amp Kit ]

A 50 ohm resistor will need to handle 288 watts (but only 2.4 amps at 120VAC) if the chassis faults to the incoming hot wire or it will explode unless the power is removed somehow.

I still don't see how this protects a metal case from being energized by an internal fault in the amp and creating a safety hazard. How is the branch circuit breaker going to trip? What will de-energize the case other than the circuit breaker or the fuse when the resistor is limiting the fault current to very low levels?

I'm not trying to be a pain - I simply do not understand. BTW, X2 caps main feature is that they can take very large voltage spikes (2.5kV) without failing. I also calculate the impedance of a .22uf cap to be about 120 ohms (reactive) at 60Hz (it's an open circuit if the fault is DC). This in parallel with a 150 ohm resistor still won't come close to providing a low enough impedance path to ground to blow fuses or trip breakers.

I didn't think that an amp having a single uninsulated metal chassis that isn't directly bonded to earth could meet code and be sold today (but I could be wrong and if so I'd like to find out!). I though you either had to bond the chassis to earth ground or use a double insulated chassis. I've seen plenty of very old equipment that doesn't meet this criteria but nothing new.

Thanks.

_________________
Stephen

Hi all, Hi Steven .

LONG but read it:

I dunno if i'm right now, but i think there are some basic misunderstandings on decoupling the "mains earth" from the amp circuit.

A METAL enclosure or METAL chassis surely must be hard wired to ground. Directly - with NO resistance inbetween the chassis, metal parts that could be touched by a user and the mains earth - where i presume that "mains earth" is the grounding from the power line. A circuit breaker or mains fuse in the household can not trip with a resistance that is connected between the AC line and the breaker. Before this happens, the resistors or caps burn down as they can not withstand 120VAC @ 16 A for a second... A pure short circuit has a power of 2000 Watts (in our location it's 3600 Watt). What should a 100 to 150 Ohm resistance do here? Nothing else than fail!

Connect one 150Ohm / 5 Watt resistor, one 10A circuit breaker and a switched 60W light bulb in series and see what happens.. Don't forget wearing your protection glasses! Even the fact that an AC current is rated as impedance and not as ohm's resistance doesn't help in any way. I hope that makes sense..

Any resistance between AC powered devices and chassis or enclosures made of metal must be avoided. In any case. Transformers and transformer screens, line filters, output transformers must be connected to earth too. Anyways anytime. Even if you use chassis or enclosures made of wood or plastic.

A short circuit on the "DC side" basically is protected by the main-transformer's primary or secondary fuse, if the current of the load overrides the fuse's current rating. The DC circuit also has NO reference to mains earth. The DC circuit is isolated from earth by the main-transformer. This is beacuse the secondary transformer windings have no point where they are connected to earth, thus they have no reference to ground. Take a PSU and a DMM and a wire with clamps. Connect the DMM to the DC output of the PSU
and the DMM COM to the mains earth and measure the DC. You'll get no measurable result because of the fact that there is no earth reference - as meant before.

But, HV circuits are also attached to the mains ground for several reasons. One of the reasons is the GCFI protection. If a short occurs there is also a chance that the GFCI will trip. The circuit breaker surely doesn't.
What and why?! - Because the mains earth / power line earth and the NEUTRAL basically have the same reference. Earth and the NEUTRAL are 0 (NULL). But - they are splitted in the household.

The NEUTRAL works as "0" in a circuit breaker and GFCI balancer (The GFCI is a magnetic, balanced switch which shunts to earth). The earthing connects all to earth and at the end to the power line. If a AC / DC current (40mA - 80mA) runs into the neutral, the GFCI will break the mains power in the household at the point where the GFCI is mounted.

So - why the heck do some builders or schematics use resistors and capacitors to decouple the DC GROUND from MAINS EARTH??

Because it's OK and works. AC hum and buzz is an pretty much unbeloved child in the amplifier world.. Everyone tries to get rid of it. If you have coupled the DC grounding with the mains earthing the chance of hum and buzz is very high. There are ground loops and device loops (potential differences) which cause hum in several cases. Now - here comes the point why DC ground decoupling is used.

We can decouple the "DC ground" from "mains ground / earth" by using an 100 to 150 Ohm resistor and 100n / 630V cap. That will say that the DC circuit ground is connected by using an resistor and cap to the point where the mains earth connects to the chassis. At only one point. This will provide a way of secure amplifier (GFCI trip) and also decreases ground loops. That's the only way to use the resistor and cap. Nothing others.

Now to the conclusion:

The PSU from this project and the connections are OK! Bruce did the same like i mentioned before. He decoupled the amplifier ground (DC ground) from the mains earthing / grounding. This is not very unusual and works fine to me and for many others. As you can see on the schematics, a mains filter is used in the circuit. The filter package is directly connected to earth and the chassis, thus the chassis is directly hooked up to earth. The amplifier ground is attached to the earth via the decoupling network. There is NO resistance between earth and the amp!

I hope that makes sense to you.

Have fun

_________________
Cheers,
Tom.

Some of my projects: TDA2050 Chip Amp, the LM3886 Gainclone Thread and the Szekeres Headamp Thread.

Tom, your response is awesome and makes total sense. I simply misunderstood. The schematics don't really show the connection to the chassis unless I missed it. I made the assumption the chassis was connected to ground through the resistor and cap. Duh! I don't know why I thought that. Temporary (I hope) idiotitis. I deal with grounding both as an engineer in the utility industry as well as in a recording studio, and I'm reasonably familiar with both the NEC and the NESC. You know your stuff because everything you said was correct (or at least it's what I believe to be correct). Most people (not on this forum) don't know that the neutral and ground never come into contact inside a residence or even realize that only one wire carries voltage. Nor do they have any concept of how GFCIs work (5ma trip I thought).

Thanks very much for taking the time to write such a detailed and crystal clear answer. I left out the resistor and cap in both my amps and just used a wire because I misunderstood the schematic and nobody but you until now understood my questions regarding these. The part I needed to hear was that the chassis is connected directly to the earth ground and the ground of the amp circuit is connected the the earth ground through a resistor and capacitor. The rest I knew, but there was no way for anybody to know what part I didn't "get". I guess I could have phrased my questions better. My amps have some noise and hum, not much, but a little. It's time I put the resistor and cap in there. Looks like I have a project for tomorrow.

Sorry for being such a pain in the ass. Thanks again!

_________________
Stephen