Grounding old equipment

[John B]

Here's a stereo that I am repairing. Looks like 1990's stuff? Anyhow, one thing that surprised me is that the outer shell is metal, yet it only has a 2 prong plug, no ground. As you can see from the picture, the mains wiring sits rather close to part of the metal chassis. The earth on the RCA inputs and the earth on the headphone socket all connect to the chassis, so even if you cannot reach the metal chassis, there's still a few common human contact areas that could be a shock hazard. I get the heebie-jeebies just looking at. Is there a reason why I shouldn't ground the metal chassis with a 3 prong lead, and why wasn't it done by design in the first place?

[HackedFridgeMagnet]

not it is not very safe. doesn't satisfy class I or class II insulation requirements.
see this thread
https://www.eevblog.com/forum/projects/audio-grounding-xlr/

also
https://en.wikipedia.org/wiki/Appliance_classes

It would be ok if you ground connect the chassis to protective earth but depending on what is connected and how it is connected internally you could end up with ground loops.

edit:
oops Noticed I had used ambiguous wording.

[WastelandTek]

kids these days, we used to run AA5 radios with nonpolarized plugs and we only got shocked once in a while geeezzz




yeah, that's pretty ghetto by modern standards

[Electro Detective]

yeah, a lot of 240 volt 50 hz two prong video recorders and DVD players were/are good like that. 

You get a nice tingle from the metal when moving them about in a cabinet if they are still plugged into the mains and or TV antenna cable   

Once they are hooked up with other gear via RCA cords, no tingle, no problem,
even if all the gear is 2 prong powered

I assume all those combined 'floating' neutrals and chassis vote they are part of the Neutral/Earth/Ground link at the electrical switchboard, and settle down

If any one of a group of devices does come standard with a earth/ground 3rd prong, all the others connected via RCA cords get magically earth/grounded by the combined RCA shields AFAIK


The main concern with two prong gear is to ensure the Active wire going into the device is connected to the internal FUSE, not the neutral. 

[capt bullshot]

Don't ground it.
Grounding the chassis places all kind of ground loops on the system, resulting in line frequency hum.

These things were designed to work this way and to be safe according to the standards that were active then.  They aren't getting unsafe over time, it's the standards that change, if you build such things from scratch today, it would look different, but still won't have a grounded chassis. As long as you do your leakage current and isolation properly, there's no reason to ground a chassis today and wasn't back then. The isolation and leakage was done properly back then is still okay for todays usage.

[Zero999]

yeah, a lot of 240 volt 50 hz two prong video recorders and DVD players were/are good like that. 

You get a nice tingle when moving them about in a cabinet if they are still plugged into the mains and or TV antenna cable   

Once they are hooked up with other gear via RCA cords, no tingle, no problem,
even if all the gear is 2 prong powered

I assume all those combined 'floating' neutrals and chassis vote they are part of the Neutral/Earth/Ground link at the electrical switchboard, and settle down

If any one of a group of devices does come standard with a earth/ground 3rd prong, all the others connected via RCA cords get magically earth/grounded by the combined RCA shields AFAIK


The main concern with two prong gear is to ensure the Active wire going into the device is connected to the internal FUSE, not the neutral. 

The appliances that gave you that tingle probably have a switched mode power supply. The leakage across the Y capacitors can give you a tingle. Normally it's not an issue but if several appliances are connected together or their chassis touch, it can become more of a safety hazard, as the leakage currents add together.

The device shown in the original post, appears to have mains frequency transformers, which leak far less current than an SMPS with Y capacitors bridging the mains and output.

not it is not very safe. doesn't satisfy class I or class II insulation requirements.

I don't think there's enough information to come to that conclusion. We don't know where the mains tracks are on the other side of the board or the creepage and clearances between them and the case/secondary.

[HackedFridgeMagnet]

not it is not very safe. doesn't satisfy class I or class II insulation requirements.

I don't think there's enough information to come to that conclusion. We don't know where the mains tracks are on the other side of the board or the creepage and clearances between them and the case/secondary.

Was just going off the fact that there is only one layer of insulation and they put two layers on any two wire mains these days. No doubt to get a compliance badge.
That said I probably have an amp done to a similar standard very close to me now and am not about to change it.

[John B]

The unit does use a pair of transformers (one switched for low power standby). But before the isolating transformers, the posts that the mains wire are soldered to go through the board and out the other side. So there is about 8mm of clearance between the flimsy, flexible chassis and the wires straight from the wall. I suppose it hasn't killed anyone yet...but still.

[RGB255_0_0]

The unit does use a pair of transformers (one switched for low power standby). But before the isolating transformers, the posts that the mains wire are soldered to go through the board and out the other side. So there is about 8mm of clearance between the flimsy, flexible chassis and the wires straight from the wall. I suppose it hasn't killed anyone yet...but still.

Maybe if it gets dropped or kicked it would be a concern. Maybe have the cable routed underneath but that doesn't look possible to solder that way. As a compromise, put some Kapton tape directly below the mains connection joint.

[Nusa]

The stamped steel I see in the picture is not nearly as flexible or fragile as you imply. You'd literally need to use a hammer or beat the unit against a rock to compromise the assembled case enough to contact the AC input.

[RGB255_0_0]

Delivery drivers have strong boots. 

In any case, it'd be a couple of minutes to add some kind of insulating sheet there if you were inclined to worry about it.

[Zero999]

not it is not very safe. doesn't satisfy class I or class II insulation requirements.

I don't think there's enough information to come to that conclusion. We don't know where the mains tracks are on the other side of the board or the creepage and clearances between them and the case/secondary.

Was just going off the fact that there is only one layer of insulation and they put two layers on any two wire mains these days. No doubt to get a compliance badge.
That said I probably have an amp done to a similar standard very close to me now and am not about to change it.

There are two layers of insulation on the mains cable: the black over jacket on top of the brown and blue insulation on the two cores.

[b_force]

Don't ground it.
Grounding the chassis places all kind of ground loops on the system, resulting in line frequency hum.

These things were designed to work this way and to be safe according to the standards that were active then.  They aren't getting unsafe over time, it's the standards that change, if you build such things from scratch today, it would look different, but still won't have a grounded chassis. As long as you do your leakage current and isolation properly, there's no reason to ground a chassis today and wasn't back then. The isolation and leakage was done properly back then is still okay for todays usage.

I agree with this!

Worked on a lot of audio gear and as long as the clearances/creepage is good enough, the product is safe.
In fact, most gear from the 80's/90's is totally safe.
It's only the super old tube amplifiers which didn't had a isolation transformer.

[Richard Crowley]

Don't ground it.
Grounding the chassis places all kind of ground loops on the system, resulting in line frequency hum.

These things were designed to work this way and to be safe according to the standards that were active then.  They aren't getting unsafe over time, it's the standards that change, if you build such things from scratch today, it would look different, but still won't have a grounded chassis. As long as you do your leakage current and isolation properly, there's no reason to ground a chassis today and wasn't back then. The isolation and leakage was done properly back then is still okay for todays usage.

I disagree with this!

If you can't fix a ground loop problem, you don't belong in this forum.  But none of us (even the physicians) can fix electrocution.

[b_force]

Don't ground it.
Grounding the chassis places all kind of ground loops on the system, resulting in line frequency hum.

These things were designed to work this way and to be safe according to the standards that were active then.  They aren't getting unsafe over time, it's the standards that change, if you build such things from scratch today, it would look different, but still won't have a grounded chassis. As long as you do your leakage current and isolation properly, there's no reason to ground a chassis today and wasn't back then. The isolation and leakage was done properly back then is still okay for todays usage.

I disagree with this!

If you can't fix a ground loop problem, you don't belong in this forum.  But none of us (even the physicians) can fix electrocution.

Since when is not-grounding equal to electrocution?

In fact, as far as I know, most audio gear, especially CD-players etc, are never grounded.

[RGB255_0_0]

As said before though, the likelihood is eventually the ungrounded equipment does end up grounded; via HDMI and so forth to the TV or receiver.

What I don't like is figure of eight plugs because you have no control whether the fuse is live or neutral (pedant: yes ok after bridge...). Should move to Mickey Mouse plug and just ground equipment properly.

[Gyro]

not it is not very safe. doesn't satisfy class I or class II insulation requirements.

I don't think there's enough information to come to that conclusion. We don't know where the mains tracks are on the other side of the board or the creepage and clearances between them and the case/secondary.

Was just going off the fact that there is only one layer of insulation and they put two layers on any two wire mains these days. No doubt to get a compliance badge.
That said I probably have an amp done to a similar standard very close to me now and am not about to change it.

There are two layers of insulation on the mains cable: the black over jacket on top of the brown and blue insulation on the two cores.

I'm uncomfortable with what they've done with the mains cable termination, there's obviously been some cost cutting.

The normal termination for mains cables in Class II products I've designed is for the cable to be crimped into a touchproof 2 pin socket which plugs onto the header on the PSU. Exposed inner cores (single insulation) are covered in heatshrink to maintain double insulation all the way to the crimps in the connector. The mains cable length inside the case is also minimised so that, in the unlikely event of the connector becoming unplugged, it can't contact secondary side components or chassis.

Now if you look at the photo, the mains input header is in place but the inner cores are soldered directly to the pins, there's no secondary insulation on the inner cores and the cable connection point results in quite a lot of cable inside the case. It's pretty unlikely that both cores would become unsoldered - unless maybe a fatigue problem caused by the cable flapping about, or a badly soldered repair, but if it did, then you have the possibility of the live conductor contacting the case. At least the inner core length has been kept to a minimum.

I'm not sure, off hand, whether this arrangement would pass approval, but at minimum, I would want to add some mechanical tie-down of the cable to the board. This should be close enough to the termination to prevent the cores touching the chassis on joint failure.

The choice of SRBP for the PCB is normally a dodgy one as it is mechanically brittle and can break away at the mounting points. cem-1 is more appropriate. It's forgivable in this case because the heavy components are screwed to the case, providing additional support. It has obviously heen through the value engineering mill though.


P.S. It's not the delivery driver's boots you need to worry about, it's the crane driver dropping the shipping container on the dock. I've seen cases where the bottom 2-3 box layers have become 'sacrificial'! It's normal for OEMs to ship 5-10% (negotiated) to distributors for spares.

[tronde]

I never stop being puzzled about electronics people and their fear for mains voltage. We are talking about maximum 250 volts. It's not kilovolts. Some of you do really need to get some realism into this.

This is a stereo. It's meant for a normal domestic environment, i.e. no excessive humidity or dust. It is not ment to be mechanically abused as power tools are.

According to European regulations, this kind of equipment will need at least 4mm of clearance and at least 5mm of creepage between mains voltage and "exposed parts". That is basically anything a user can touch.

The term "double insulated" does not necessarily mean "two layers of insulation". It means the insulation should give the same level of protection.

A normal Class-II equipment is allowed to leak 0.25mA, and if they use Y1 capacitors between each mains wire and 0V out as they often do on smaller chargers, this can be felt as a light shock.

The only suspect thing I can see here, is the same as Gyro mention; the mains wires should be tied to the PCB, but this was not normal back then, and it is not a big safety risk when the stereo is used as expected.

[RGB255_0_0]

I never stop being puzzled about electronics people and their fear for mains voltage. We are talking about maximum 250 volts. It's not kilovolts. Some of you do really need to get some realism into this.

Realism is that 250VAC indiscriminately kills. I think your sentence is dangerous in itself as newcomers may read it and then we're reading another thread about someone dying as a result of ignorance or negligence.

[Gyro]

This is a stereo. It's meant for a normal domestic environment, i.e. no excessive humidity or dust. It is not ment to be mechanically abused as power tools are.

As I alluded to in my last post, this type of consumer equipment has a pretty easy life in domestic use. The problem is getting it to the customer still sufficiently safe that it doesn't immediately try to kill him. Also to ensure that it remains safe during it's operational life (thermal, component, insulation degredation etc.)

Edit: and yes 240V AC will quite happily kill you under the right circumstances!

[floobydust]

OP's stereo, I wonder if it has any decent safety approvals in the first place.

That mains power cable, I recognize as a chinese special- I've never seen on a product with real safety approvals like UL/CSA.
Sure the wire's insulation can have UL/CSA written on it but as a power cable, it's a fail.

There are tug, abrasion, crush requirements for mains cables. Look at every one in your house and they have thicker jackets.
I saw it once on a Walmart DVD player, which only had "special inspection" stickers for safety agency. This is basically a hi-pot test.

[tronde]

I never stop being puzzled about electronics people and their fear for mains voltage. We are talking about maximum 250 volts. It's not kilovolts. Some of you do really need to get some realism into this.

Realism is that 250VAC indiscriminately kills. I think your sentence is dangerous in itself as newcomers may read it and then we're reading another thread about someone dying as a result of ignorance or negligence.

No, people must get realism into this. Confirm the regulations. 4mm clearance and 5mm creepage. This is for reinforced insulation. Normal insulation is half of that.

4 or 5mm is not much, but the most stringent regulations says it is enough.

[Gyro]

A self-respecting engineer designs for 8mm minimum. That includes board flex during mechanical shock too. Just because it's consumer doesn't mean that some prat isn't going to drop it while plugged in!


EDIT: Another point is that you aren't just designing one item - you are designing many (my highest volume Class II product shipped 60k/month). That really gets one in a million type problems, hamfisted operators... crane drivers stacking up against you.

[RGB255_0_0]

I never stop being puzzled about electronics people and their fear for mains voltage. We are talking about maximum 250 volts. It's not kilovolts. Some of you do really need to get some realism into this.

Realism is that 250VAC indiscriminately kills. I think your sentence is dangerous in itself as newcomers may read it and then we're reading another thread about someone dying as a result of ignorance or negligence.

No, people must get realism into this. Confirm the regulations. 4mm clearance and 5mm creepage. This is for reinforced insulation. Normal insulation is half of that.

4 or 5mm is not much, but the most stringent regulations says it is enough.

It's not that part I find abrasive. You wrote something I felt as badly worded and could easily be taken out of context.

[tronde]

That mains power cable, I recognize as a chinese special- I've never seen on a product with real safety approvals like UL/CSA.
Sure the wire's insulation can have UL/CSA written on it but as a power cable, it's a fail.

The cable is perfectly normal. In Europe it is today usually known as SKX or H03VVH2-F. It is 2 x 0,75 mm².