Look up the regulations in your part of the world.
Here is it 100 milliohms or lower between measured between the earth pin of the mains plug and the chassis. The resistance of the cable based on its length and rating is factored into this, so if the cable has 0.1 ohm resistance then a measured value of 0.2 ohms would be a pass.
Washing machines, fridges, microwaves etc are all in that bracket and anything the user can touch on these (controls etc) are either insulated or in direct contact with mains ground.
A CL60 between the PSU ground and the chassis earth isn't going to get in the way of that.
The chassis is always directly connected to safety earth. Always.
I don't consider a ground lift unsafe or dangerous. It is less safe than the proposed solution of a rectifier bridge instead of the CL60.
However, should there be some kind of primary to secondary short, and the audio side becomes live, this is a catastrophic failure, the amp would go bang, and possibly speakers.
I don't particularly see how a CL60 would prevent a properly rated fuse blowing, or the RCD tripping in the above, very unlikely scenario.
The only thing potentially live after all of that, wouldn't be the chassis, it would be the RCA sockets and possibly the Speaker terminals.
Obviously there are better solutions, but I don't consider every amplifier out there with a ground lift, or indeed the First Watt commercial amplifiers that use a CL60, inherently dangerous or unsafe.
I remember reading a series of posts including Andrew T et. al. from quite some time ago re: this very subject. At the time... it flew over my head. Today, it still flies a bit over my head.
Would anyone be so kind as to explain to a noob what the potential contact points / failure mechanisms may be in layperson's terms?
If this fails ... and you do this ... this could happen.
It seems that some (potentially unlikely event) must happen, then something else must happen, and then someone must touch the speaker terminals and/or the inputs.
I can't imagine that a commercial First Watt Product wouldn't meet US (or international) standards, but I could be incorrect. The "cold" resistance the way I understand it for the standard CL-60 is ~10 ohms. However, if the concern is that the chassis must be directly connected (< a few milliohms) resistance to the electrical GND - then mine is exactly that. The safety GND goes directly to chassis. The CL-60 is between "audio gnd" as I understand it and chassis gnd. So that seems fine.
I often have small children with pokey little fingers in the room with my gear, so if I can understand the risks - I can choose proper mitigation paths through human training and/or altering the circuit. I'm "one of those" people that also leaves the amp on 24/7.
I know there are finer points to this that may still fly over my head, but I'd appreciate any thoughts.
Would anyone be so kind as to explain to a noob what the potential contact points / failure mechanisms may be in layperson's terms?
If this fails ... and you do this ... this could happen.
It seems that some (potentially unlikely event) must happen, then something else must happen, and then someone must touch the speaker terminals and/or the inputs.
I can't imagine that a commercial First Watt Product wouldn't meet US (or international) standards, but I could be incorrect. The "cold" resistance the way I understand it for the standard CL-60 is ~10 ohms. However, if the concern is that the chassis must be directly connected (< a few milliohms) resistance to the electrical GND - then mine is exactly that. The safety GND goes directly to chassis. The CL-60 is between "audio gnd" as I understand it and chassis gnd. So that seems fine.
I often have small children with pokey little fingers in the room with my gear, so if I can understand the risks - I can choose proper mitigation paths through human training and/or altering the circuit. I'm "one of those" people that also leaves the amp on 24/7.
I know there are finer points to this that may still fly over my head, but I'd appreciate any thoughts.
This shows that you are not understanding the problem and the theory behind it.
The primary side thermistor has the protection (we are talking now of the protection of the thermistor) of the transformers primary winding impedance. Any initial surge current... and it is that current that develops the voltage across the thermistor... will fall within a few cycles of AC supply.
A fault on the secondary side causing full mains to be present on the secondary side thermistor would cause extremely high current to flow until either the thermistor or something else on the supply side blew.
hi - could i ask you a question ? in your photo, you are referring i guess to the bridge rectifier and the cl60 on the top right. In the schematic, there should be a 0.0033uf capacitor. How have you connected this? I also presume that since i am running a 240v, the earth for primary is slightly different to the 120v as in the schema.
It seems you are just trying to pick holes in the argument tbh.
The specific question asked was whether the thermistor on the secondary side would provide protection in the event of 'something bad happening' and the answer to that is no, it would not. The fuse may or it may not blow, each fault scenario is different.
I just find it interesting that an admin on this site says that the amp designs provided on this site and built by so many are inherently unsafe.
Look at the first PS schematic dated 2005 (post 3), and the second dated 2012 (post 16). Why do you think Nelson added the bridge rectifier in the ground lift in the 2012 schematic?
Simple answer is because it makes the circuit safer.
edit: also moderators are posting as normal members unless the use the cop emoticon. Mooly responded to a direct question about the safety of the CL60 as a ground lift. That is based on his experience.
Tony.
Simple answer is because it makes the circuit safer.
edit: also moderators are posting as normal members unless the use the cop emoticon. Mooly responded to a direct question about the safety of the CL60 as a ground lift. That is based on his experience.
Tony.
Last edited:
It is not inherently unsafe.
The mains IEC connector ground pin is connected via a wire straight to the chassis. There's your safety. In case that the MAINS voltage is shorted to ground, the household RCD will trip.
The CL-60 connected between the PS ground and chassis will also conduct a sufficient amount of fault current in case of a failure of the MOSFET, for example, in which case the most likely outcome would be a burnt woofer coil, PCB track, smoke, and possibly fire. But you will be protected in case you touch the chassis in this scenario. The fusible resistors/fusses should come in play in this case.
Adding the bridge rectifier puck in "parallel" to that CL-60 is to ensure a higher current will flow to the ground; it is an additional level of safety at the DC side that enures the fuses and/or fusible resistor will go open circuit.
You should, in all reality and if concerned about safety, be checking your household installation; grounding and a safe RCD operation in particular.
can i just double-check some spec?
The official build says the thermistors should be 4.7k not the CL60 type.
1) Why then are we using CL60?
2) The suggested bridge rectifier for the modified circuit. Is it 400v ?
3) is the cap on the primary 0.0033uf (what voltage)? i presume i only need one for the UK according to the schematic.
4) i am buying a speakers protection board which needs 18v AC. Could i use the transformer output or is that not advisable? i presume it should be ok because it's not a heavy additional load.
The official build says the thermistors should be 4.7k not the CL60 type.
1) Why then are we using CL60?
2) The suggested bridge rectifier for the modified circuit. Is it 400v ?
3) is the cap on the primary 0.0033uf (what voltage)? i presume i only need one for the UK according to the schematic.
4) i am buying a speakers protection board which needs 18v AC. Could i use the transformer output or is that not advisable? i presume it should be ok because it's not a heavy additional load.
Ideally 600V or higher. The rectified component of AC mains at 220-240V will be very close to the peak reverse voltage of a 400V bridge. For 110-120V mains a 400V type is sufficient, but I'd still put a 600V or 1kV type just to be on the safe side.
I don't see how a MOSFET failure would inject current between chassis/mains earth and the circuit ground. Unless of course it blows up and punches through the insulator, connecting the heatsink to the circuit.
The CL60 is inserted into the (otherwise mandatory solid wire connection) between chassis/mains earth and circuit gnd to isolate the circuit from mains earth by a small amount, to mitigate ground loops between equipment that share the earth prong and are also connected to each other.
The ground connection between chassis and circuit ground is mandated so that in the event of a mains level fault causing the circuit ground to go to live potential (e.g. a loose wire or a shorted transformer from overheating), any exposed metal parts will be at earth potential and the fuse would also blow.
The insertion of any element between these two is usually forbidden. The DIY community has come up with a few 'techniques', of which the CL60 is one (another one is Rod Eliot's three-component solution). The issue with only a CL60 is that the component is not designed to take 230V mains across itself even once (it needs a load). In the event that it is used in such an application the bridge rectifier is mandatory, and even then it remains only marginally safe.
The fact that nothing has ever happened does not eliminate the possibility that it might. I'm not trying to be a fearmonger but people need to know what they're dealing with and what the risks are. NTCs can be quite spectacular when they fail, you don't want to be near one when that happens. Simply put, we are using a component that is not optimal for the application, and the part of the circuit it is used in could be lethal.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.