Re: Janneman...
I got mine from www.amplimo.nl, this is the company that is connected to Menno van der Veen who designs all these wonderfull transformers. Last time I bought them I paid about 5 US$ each. If you can't get in touch with them, let me know, I know the manager over there.
Jan Didden
mikek said:Would love to know source of your special relays...might get myself a handfull....
I got mine from www.amplimo.nl, this is the company that is connected to Menno van der Veen who designs all these wonderfull transformers. Last time I bought them I paid about 5 US$ each. If you can't get in touch with them, let me know, I know the manager over there.
Jan Didden
BOSOZ delay
My BoSoZ had such a turn-on thump that it blew the voice coil right out of an active subwoofer in a crappy computer speaker system I used for testing. So I installed some relays and do not connect the outputs until 1 second after the power supply comes up.
The circuit is basic but I don't have time to draw it. A FET switch has the source grounded and the drain connected to the low-voltage side of the relay coils. The gate is connected through a capacitor to ground. The capacitor is charged via a resistor from the 5V rail. R*C should be roughly the delay you want.
When the cap charges up enough to turn on the FET, it pulls the relay coils nearly to ground, thus turning them on. When you turn the power off, the outputs will disengage more or less immediately.
A few bucks in parts might save your speakers. I don't know why the BoSoZ has such a turn-on transient. It might be that my cascoded current source comes up before the +60V rail does.
My BoSoZ had such a turn-on thump that it blew the voice coil right out of an active subwoofer in a crappy computer speaker system I used for testing. So I installed some relays and do not connect the outputs until 1 second after the power supply comes up.
The circuit is basic but I don't have time to draw it. A FET switch has the source grounded and the drain connected to the low-voltage side of the relay coils. The gate is connected through a capacitor to ground. The capacitor is charged via a resistor from the 5V rail. R*C should be roughly the delay you want.
When the cap charges up enough to turn on the FET, it pulls the relay coils nearly to ground, thus turning them on. When you turn the power off, the outputs will disengage more or less immediately.
A few bucks in parts might save your speakers. I don't know why the BoSoZ has such a turn-on transient. It might be that my cascoded current source comes up before the +60V rail does.
I had the same issues with the JLH'95. The turn-on transient is awful, rising to almost 20v, yet inaudible (?!?!?). The turn-off thump is much less severe, yet audible (the classic "bass-boom"). The amp is dead stable in between but i don't like my speakers getting that kind of treatment, even if briefly.
Again, the ESP output protection takes care of that, shutting the outputs for 3sec at turn-on and INMEDIATLY after turning it off. Recomended!
but i don't like my speakers getting that kind of treatment, even if briefly.Again, the ESP output protection takes care of that, shutting the outputs for 3sec at turn-on and INMEDIATLY after turning it off. Recomended!
Bit slow in remembering to look up the details for this post but if anyone is interested....
There are also three IC's that will do all this for you, apart from the pilot light, for only a couple of euro and a few extra components. I built one on stripboard and it has been faultless. The IC's are uPC1237 TA7317P and HA12002 and the data sheets are all available on the web with a little effort.
I used uPC1237 and I'm very happy with it. Personally I prefer to spend my time working on amps and speakers and do things like this as quickly and simply as I can. I did build the slow start cct that Doug. Self published in one of his books before I discovered these IC's. It worked fine but it's too much effort.
Peter Daniel said:Since most of the interest seems to be in the realys and not the circuits
It has the following features:
Turn on delay with fast turn-off at power down
Amplifier offset shutdown
Adjustable thermal shutdown
Pilot light flasher indicating fault condition
Modular topology for easy adaptation
There are also three IC's that will do all this for you, apart from the pilot light, for only a couple of euro and a few extra components. I built one on stripboard and it has been faultless. The IC's are uPC1237 TA7317P and HA12002 and the data sheets are all available on the web with a little effort.
I used uPC1237 and I'm very happy with it. Personally I prefer to spend my time working on amps and speakers and do things like this as quickly and simply as I can. I did build the slow start cct that Doug. Self published in one of his books before I discovered these IC's. It worked fine but it's too much effort.
Hi all
I know that there are some amplifier designs that do not show any turn-on (and turn-off) thump, so they wouldn't need any DC output protection from this point of view. But this isn't true for all amps, and those that show a thump are not all bad designs !!
But, to use Nelson Pass' words: When solder joints or contacts are getting bad or any component fails "things can get interesting" (I really like this phrase ).
About 20 years ago I worked as a repair technician, having to repair quite a lot of failed power amps or output stages.
Many BJT stages failed hard i.e. producing short cuts or tying the putput voltage to one of the rails. In these cases the amps would certainly have killed the attached woofer's voice coil without output protection.
OTOH output relays can be detrimental to the sound.
So the best thing would be to have efficient output protection without having relay contacts in the supply lines or the speaker output connections.
I heard several times that MOSFETs fail soft, i.e. that they generate an open circuit rather than a short circuit if they fail. If this is really the case then it should be quite easy to apply a protection circuit, that isn't using any relays, at least with MOSFET amplifiers.
Because MOSFET amps were very rare twenty years ago I don't have enough experience to confirm this. The only two failed MOSFET amps I ever repaired were mine: One was a P.A. amp with Hitachi MOSFETS and the other one was a small Class-D amp. In both cases the devices failed soft indeed .
Is there anyone out there who had to deal with a lot of failed MOSFET output devices ???
If yes, can he/she confirm the story about failing soft ??
Regards
Charles
I know that there are some amplifier designs that do not show any turn-on (and turn-off) thump, so they wouldn't need any DC output protection from this point of view. But this isn't true for all amps, and those that show a thump are not all bad designs !!
But, to use Nelson Pass' words: When solder joints or contacts are getting bad or any component fails "things can get interesting" (I really like this phrase
).About 20 years ago I worked as a repair technician, having to repair quite a lot of failed power amps or output stages.
Many BJT stages failed hard i.e. producing short cuts or tying the putput voltage to one of the rails. In these cases the amps would certainly have killed the attached woofer's voice coil without output protection.
OTOH output relays can be detrimental to the sound.
So the best thing would be to have efficient output protection without having relay contacts in the supply lines or the speaker output connections.
I heard several times that MOSFETs fail soft, i.e. that they generate an open circuit rather than a short circuit if they fail. If this is really the case then it should be quite easy to apply a protection circuit, that isn't using any relays, at least with MOSFET amplifiers.
Because MOSFET amps were very rare twenty years ago I don't have enough experience to confirm this. The only two failed MOSFET amps I ever repaired were mine: One was a P.A. amp with Hitachi MOSFETS and the other one was a small Class-D amp. In both cases the devices failed soft indeed .
Is there anyone out there who had to deal with a lot of failed MOSFET output devices ???
If yes, can he/she confirm the story about failing soft ??
Regards
Charles
I once shorted outputs on my A75. One mosfet faild (IRF230), became short and tripped the fuse in the amp. No other mosfets failed.
As to avoiding relay at the output, it might be interesting to place the relay on primary with the same speaker protection circuit. Depending on your capacitance bank it might not work quick enough thou.
I wonder what Mr. Pass' experience and views are with regards to speaker protection. His commercial offerings don't have such (protection) and he's doing it long enough to know what failure ratio is.
As to avoiding relay at the output, it might be interesting to place the relay on primary with the same speaker protection circuit. Depending on your capacitance bank it might not work quick enough thou.
I wonder what Mr. Pass' experience and views are with regards to speaker protection. His commercial offerings don't have such (protection) and he's doing it long enough to know what failure ratio is.
Hi Peter Daniel
I agree with you on the point regarding the relay on the primary. A solid state relay might be nice here, but I am not sure how much audio-frequency hash they can generate that will feed through the transformer. When used with a passive speaker having a sturdy woofer (your Focal for instance) then even a large capacitor bank wouldn't be much of a problem with this solution (100 000 Microfarads and 6 Ohms give a time-constant of 0.6 seconds).
The real problem would be DC coupled midranges and tweeters in active speakers. Maybe a discharging circuit consisting of two mosfets and two resistors (and of course the necessary detection circuit and drivers) could help but might still be too slow for the tweeter ?
I am also very interested in Mr. Pass' experience not only regarding the behaviour of ageing equipment.
He will for sure have a lot of experience with failed new stuff. If you are developing power amps for years then you might kill a lot devices even when you take as much care - when designing circuits - as is expected from a good engineer.
Regards
Charles
I agree with you on the point regarding the relay on the primary. A solid state relay might be nice here, but I am not sure how much audio-frequency hash they can generate that will feed through the transformer. When used with a passive speaker having a sturdy woofer (your Focal for instance) then even a large capacitor bank wouldn't be much of a problem with this solution (100 000 Microfarads and 6 Ohms give a time-constant of 0.6 seconds).
The real problem would be DC coupled midranges and tweeters in active speakers. Maybe a discharging circuit consisting of two mosfets and two resistors (and of course the necessary detection circuit and drivers) could help but might still be too slow for the tweeter ?
I am also very interested in Mr. Pass' experience not only regarding the behaviour of ageing equipment.
He will for sure have a lot of experience with failed new stuff. If you are developing power amps for years then you might kill a lot devices even when you take as much care - when designing circuits - as is expected from a good engineer.
Regards
Charles
failure rate
As to Pass Labs amplifier failure rate (kind of related to the protection issues that we're talking about here), Mr. Pass has indicated (in one of the very old post - I can't find it) that his amp failure rate is almost zero and I guess you don't need protection if your amp don't fail. I remember making a comment that if your amp/design don't fail, how do you get people to buy more of your product and the money thing that goes with it.
As to Pass Labs amplifier failure rate (kind of related to the protection issues that we're talking about here), Mr. Pass has indicated (in one of the very old post - I can't find it) that his amp failure rate is almost zero and I guess you don't need protection if your amp don't fail. I remember making a comment that if your amp/design don't fail, how do you get people to buy more of your product and the money thing that goes with it.
Re: BOSOZ Pre-Amp turn on muting
This one does not introduce any degradation in the signal path.
http://mitglied.lycos.de/Promitheus/delay_circuit_for_toroids.htm
It's very small and fits everywhere. What it does is putting some resistance between main and transformer primary, so to charge power
supply condensers slowly. In about 2-5sec. (depends on the value of C1 and C2, the higher the value the longer it takes ) the relay switches, resistances are out of the way and power supply is directly connected to mains. I have it implemented in a valve amp and in the BOSOZ itself and it works nicely. It has also a place in a couple of monoblocks 5w SOZ that are in their final stages. In the BOSOZ, at full volume, woofer movement is still very pronounced but very very slow... no bump, no harm. You may want to use 22ohm power resistors instead of 10ohm to make it even less apparent. LCAudio circuit is based on the same idea, with the difference that timing is obtained electronically instead of being dependant on a charging condenser, and the single power resistor is of a much higher value, around 150ohm... so there is room to experiment here.
fcel said:
This one does not introduce any degradation in the signal path.
http://mitglied.lycos.de/Promitheus/delay_circuit_for_toroids.htm
It's very small and fits everywhere. What it does is putting some resistance between main and transformer primary, so to charge power
supply condensers slowly. In about 2-5sec. (depends on the value of C1 and C2, the higher the value the longer it takes ) the relay switches, resistances are out of the way and power supply is directly connected to mains. I have it implemented in a valve amp and in the BOSOZ itself and it works nicely. It has also a place in a couple of monoblocks 5w SOZ that are in their final stages. In the BOSOZ, at full volume, woofer movement is still very pronounced but very very slow... no bump, no harm. You may want to use 22ohm power resistors instead of 10ohm to make it even less apparent. LCAudio circuit is based on the same idea, with the difference that timing is obtained electronically instead of being dependant on a charging condenser, and the single power resistor is of a much higher value, around 150ohm... so there is room to experiment here.
Attachments
Hi guys!
i really don't use anything else then a line fuse & thermistor for protection. If anything blows, well, S#%T happens, and if you can build an amp and get a kick out of it, you should be happy too repairing it
As for inrush current i tested a simple, simple, simple soft start circuit on my future 15w SOZ with 300000uF/ch ( my design, with a little help from NP)
here is a pic of it:
Yes, thats just a 150W 220V lamp in series with the transformer primary and a light switch.
Turn the switch off and when you turn the amp on the house fuses don't blow.
Wait 10 secs and turn the switch on (lamp goes off), amp is runnig at desired volts
No signal mess and digital time delay circuits and stuff.
i really don't use anything else then a line fuse & thermistor for protection. If anything blows, well, S#%T happens, and if you can build an amp and get a kick out of it, you should be happy too repairing it
As for inrush current i tested a simple, simple, simple soft start circuit on my future 15w SOZ with 300000uF/ch ( my design, with a little help from NP)
here is a pic of it:
Yes, thats just a 150W 220V lamp in series with the transformer primary and a light switch.
Turn the switch off and when you turn the amp on the house fuses don't blow.
Wait 10 secs and turn the switch on (lamp goes off), amp is runnig at desired volts
No signal mess and digital time delay circuits and stuff.
Attachments
Peter,
I guess I'm aware that it's neither DC protection nor soft start circuit but as long as my speaker doesn't produce such a loud thump at turn on, I'm willing to try it.
Now that you bring up the word "thermister" ... how come the thermister that was already in my BOSOZ does not perform this delay function? May be it does but provides delay in milliseconds and since the delay is so short, the speaker will still produce a thump at turn on?
I guess I'm aware that it's neither DC protection nor soft start circuit but as long as my speaker doesn't produce such a loud thump at turn on, I'm willing to try it.
Now that you bring up the word "thermister" ... how come the thermister that was already in my BOSOZ does not perform this delay function? May be it does but provides delay in milliseconds and since the delay is so short, the speaker will still produce a thump at turn on?
Below is the schematic for a speaker protection circuit:
Power on delay (2Sec), DC protection and overload protection.
An externally hosted image should be here but it was not working when we last tested it.
Power on delay (2Sec), DC protection and overload protection.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.