Fine, I'll keep it then. 🙂
This is not the point... 🙄
Do you think it will be an improvement to have that ground separation?
Do you expect problems implementing it?
No problems...just use some anesthetic... 😀
Race update:
- BMCBob
- SoIL4x4
- Lehmanhill
- b.veneri
Last edited:
Yes, for the simple reason that I just saw that mistake made in a pre-production layout of a circuit I was responsible of designing, routed by a business partner. It was the same circuit basically, extra diodes from the secondary into a RC. A ripple current of a few mA caused measurable and audible hum/buzz (when putting your ear directly on the woofer cone, though). This tiny current just went along the path of a GND reference coming from the preamp section.
BTW, the emitter of Q1 (driving the relay) should as well return to the dirty GND, especially as it is so very close the GND sensing point at R11. That would be more effort, I will simply use a wire to check what it does...
In the turn-on delay and DC protection I see some minor issues, did anybody test it thorougly? I suspect it might oscillate at a specific negative DC voltage from the amp because the open relay will unload its +24V supply, pulling the base of Q2 higher causing it turn on again, all with slow slopes (no real Schmitt-trigger action with fast on/off transitions). Again this is nothing severe, but not a clean design either.
Very interesting...
Did that PCB use ground planes?
I ask because I tried to put my ear directly in contact with one of my woofers and I can hardly say I'm actually hearing it or not, BTW I would say it's real but very very very very low.
I'll chek if something can be done with the actual layout.
I really don't know, it's Penasa's design. I've simply used it.
Yes, but there were several errors at stategic points which caused the effect to be relevant. We were suprised ourselves that it made a difference where that current was returned to.... I said "it won't matter", but colleague tried it anyway and he was right...
Well, then it's OK. Depends on the efficiency of the speakers of course.
Interesting
Sure, mine are 89dB, maybe with more efficient speakers would be more clearly audible.
I've tried to extend the dirty ground plane to transistors but I'm not convinced it would not create new problems.
In that way the bridge ground plane woul be adiacent to the 0V one, the 1R and to the input cap.
Another possibility is to rearrange the DC protection but this would lead to the relay and output ground to be very near the 0V ground plane, the 1R and input cap.
Well then leave it as it is. I sure don't want to be responsible of messing up your design effort -- and slowing down the progress of the project -- with my (as of now) unsubstantiated suggestions from remote... that's why I ordered your boards to check the real thing on the bench, then we have a basis for reasonable improvements (if required, that is).
Well, in fact I'm not even sure that I've heard that hum...
For sure I didn't noticed any problem and I think Bob too (can you confirm Bob?)
BTW your suggestion is welcome and it would be interesting to investigate further for the ultimate performace.
Great!
Your impressions and measurements will be an intriguing read. 🙂
Race update:
- BMCBob
- SoIL4x4
- Lehmanhill
- b.veneri
- billo44
Well, it's actually yes and no.
With something plugged into the inputs, and turned on, the amps are dead silent. The Lightspeed, if powered off will create a hum. Dario, you might try this with your DCB-1. I'll double check when I get them back. I haven't put my ear super close to the wooofer but I will try that soon.
With something plugged into the inputs, and turned on, the amps are dead silent. The Lightspeed, if powered off will create a hum. Dario, you might try this with your DCB-1. I'll double check when I get them back. I haven't put my ear super close to the wooofer but I will try that soon.
Fine 🙂
It's pretty normal and you should experience the same with the stock My_Ref if not connected to safety earth.
With my DCB1 turned off the FE is deadly silent, with the pot used as passive preamp the same.
There is noise only if nothing is connected to the inputs, exactly like a stock My_Ref.
I possibly heard something (not completely sure about that) only when I put my ear in direct contact to the woofer membrane.
I'm using a two wire wall-wart to power the LS with DC. There is no safety/chassis ground back to the wall plug. That may contribute to the hum also.
Race update:
Partecipants should pay by this sunday, possibly.
- BMCBob
- SoIL4x4
- Lehmanhill
- b.veneri
- billo44
- arthur
Partecipants should pay by this sunday, possibly.
Last edited:
The lack of a PE does not affect the hum.
There are thousands, or should I guess millions, of Double Insulated audio kit out there and the lack of PE is supposed to make for easier hum attenuation, as well as cheaper.
The PE is a Safety issue. It is not needed for audio performance. Most equipment will perform better in the audio parameters without a PE connection.
Adding the PE can often make audio performance worse.
Race update:
Partecipants should pay by this sunday, possibly.
- BMCBob
- SoIL4x4
- Lehmanhill
- b.veneri
- billo44
- arthur
- StefanoSan73
- Suburra
Partecipants should pay by this sunday, possibly.
Attached is a LTspice file for the protection circuit. I have not seen misbehavior like you describe but maybe some alteration of the stimulus voltages would show something. If you try the model and find a circuit issue or even just a problem with the model, let us know.
In practice timing capacitor leakage has caused some circuits in the last group buy to malfunction. Also if the amp is stuck at -Vee the transistor Vbe's and timing capacitors will all be reverse biased. Clamping diodes fix that problem and I added those to my amp. I doubt that anyone has tested the protection circuit under fault conditions. If it does not interfere with normal operation of the amp most users let it go at that.
Thanks, Bill.
In fact it doesn't oscillate. When I slowly ramp the output voltage (of LM3386) to negative values at a certain point (-1.8V) the relay current starts to cease slowly as well, with a small ringing at the begining, depending on the size of C16. With 100uF it is well damped.
==> nothing to worry about, the relay just opens very gradually.
Trip points are not symmetrical, but that isn't that much of an issue.
EDIT: wrong observation removed
In fact it doesn't oscillate. When I slowly ramp the output voltage (of LM3386) to negative values at a certain point (-1.8V) the relay current starts to cease slowly as well, with a small ringing at the begining, depending on the size of C16. With 100uF it is well damped.
==> nothing to worry about, the relay just opens very gradually.
Trip points are not symmetrical, but that isn't that much of an issue.
EDIT: wrong observation removed
Last edited:
Race update:
Partecipants should pay by this sunday, possibly.
- BMCBob
- SoIL4x4
- Lehmanhill
- b.veneri
- billo44
- arthur
- StefanoSan73
- Suburra
- KSTR (ongoing)
- Randytsuch
Partecipants should pay by this sunday, possibly.
Question about C9
Will black gate 'F' 220uF 50v work well at C9, they are low ESR I think.
I tried to google but didn't come up with anything conclusive...
I can get these locally for about 23 euro each... a bit expensive if they are no use!
Thanks for any advice.
Bill
Will black gate 'F' 220uF 50v work well at C9, they are low ESR I think.
I tried to google but didn't come up with anything conclusive...
I can get these locally for about 23 euro each... a bit expensive if they are no use!
Thanks for any advice.
Bill
Hi Bill,
I think they'll be fine, better than BG standard.
Best would be BG FK (16V and up).
There is a Russian eBay seller that has them for about the same (high) price.
- Status
- Not open for further replies.