Babowana said:
Very good information, thank you for providing that. I am learning alot researching the F5 prior to building it.
Thanks. 🙂
tinitus said:
Wow! I just read Papa's article on Power Supplies and that's almost an exact quote!
You have learned much, young Padawan.
tinitus said:
Sounds like a good idea to use them close to the source and destination.
Thanks for the input guys!
Ron
I found a good webpage with comparissions of various capacitors when used in power supplies. It is a subjective test with real listening values in the same power supply,
http://tech.juaneda.com/en/articles/electrolyticcapacitors.html
Ron
http://tech.juaneda.com/en/articles/electrolyticcapacitors.html
Ron
Hi,
there are two concerns in starting a power amplifier from cold.
1.) the start up current required to get the transformer working.
This requires a soft start in the primary circuit.
It can be either a Power Thermistor or a Power Resistor, selected to reduce the peak current at start up to a value acceptable to the designer and the fuse he/she decides to use. If a Resistor is chosen then a bypass that closes soon after start up must be incorporated. If a Thermistor is used the bypass can be optional, but I recommend a relay bypass across either the the Thermistor and/or the Resistor that is timed to ~200ms (about 10cycles of 50/60Hz). This soft start is almost independent of the capacitance that needs to be charged.
2.) the charging current to bring the PSU upto working voltage.
This requires a slow charge circuit if the inherent resistance in the transformer and wiring is so low that peak charging current is a concern.
I have charged +-75mF to +-50Vdc from a 1kVA transformer fused @ T2.5A on the primary with only a soft start circuit. The slow charge is not a problem for the fuse nor the transformer nor the 35A bridge rectifier nor the bank of 10*15mF capacitors.
However if one is worried about peak charging currents then a slow charge circuit should be fitted in the secondary side of the transformer.
The obtain good charge vs time, I consider it better to use a Power Thermistor here, one in each supply rail. Again it is optional to bypass or not. The time delay needs to be much longer than the soft start. Try 2s to 15s depending on what peak charging current you are prepared to tolerate.
there are two concerns in starting a power amplifier from cold.
1.) the start up current required to get the transformer working.
This requires a soft start in the primary circuit.
It can be either a Power Thermistor or a Power Resistor, selected to reduce the peak current at start up to a value acceptable to the designer and the fuse he/she decides to use. If a Resistor is chosen then a bypass that closes soon after start up must be incorporated. If a Thermistor is used the bypass can be optional, but I recommend a relay bypass across either the the Thermistor and/or the Resistor that is timed to ~200ms (about 10cycles of 50/60Hz). This soft start is almost independent of the capacitance that needs to be charged.
2.) the charging current to bring the PSU upto working voltage.
This requires a slow charge circuit if the inherent resistance in the transformer and wiring is so low that peak charging current is a concern.
I have charged +-75mF to +-50Vdc from a 1kVA transformer fused @ T2.5A on the primary with only a soft start circuit. The slow charge is not a problem for the fuse nor the transformer nor the 35A bridge rectifier nor the bank of 10*15mF capacitors.
However if one is worried about peak charging currents then a slow charge circuit should be fitted in the secondary side of the transformer.
The obtain good charge vs time, I consider it better to use a Power Thermistor here, one in each supply rail. Again it is optional to bypass or not. The time delay needs to be much longer than the soft start. Try 2s to 15s depending on what peak charging current you are prepared to tolerate.
it's not the capacitors that require the 13A mains (primary) fuse.nigelwright7557 said:
It's the start up current required to get the transformer working that requires the stupidly large fuse rating.
Fit a soft start and get that fuse down to something much more sensible. It is capable, at 13A, to pass >3kW for ever and will pass upto 6kW for quite a period of time before rupturing. This could extend to some minutes.
BTW,
the mains fuse MUST be rated lower than the lowest rated cable that is fed from it that is not separately fused. i.e. if a 6A cable is used anywhere in the mains wiring after that 13A plug top fuse then you MUST use <=5A plug top fuse, or separately fuse all the low rated cables.
I have always been told that up to 500watt trafo is ok without softstart
Thats for toroids
IE-core etc may be bigger
Thats probably just the safe limit with no issues of blowing fuses
Then theres theories that the statup current wears down the caps
Not sure whether its more theoretical, or really a fact
At least I havent seen any proof of it
Dont know why, but softstart kits for power transformers cost serious money
High ripple current is a widely used spec
I take it they mean high peak current
But ripple is also leftovers from not perfect bridge
What does it mean really
Thats for toroids
IE-core etc may be bigger
Thats probably just the safe limit with no issues of blowing fuses
Then theres theories that the statup current wears down the caps
Not sure whether its more theoretical, or really a fact
At least I havent seen any proof of it
Dont know why, but softstart kits for power transformers cost serious money
High ripple current is a widely used spec
I take it they mean high peak current
But ripple is also leftovers from not perfect bridge
What does it mean really
hi all
is the f5 scalable to higher power levels like the aleph series of amps
say more output devices and voltage with bigger sinks to match
if so would 100w into 4 or 8 ohm be doable
im currently using 2x f5 to biamp some esl and i love the sound at low volume but they do struggle at anything over half level
i know the f5 is not suitable to drive my esl but the sound is so nice id prefer to look at ways of getting more power than to build another amp
cheers
is the f5 scalable to higher power levels like the aleph series of amps
say more output devices and voltage with bigger sinks to match
if so would 100w into 4 or 8 ohm be doable
im currently using 2x f5 to biamp some esl and i love the sound at low volume but they do struggle at anything over half level
i know the f5 is not suitable to drive my esl but the sound is so nice id prefer to look at ways of getting more power than to build another amp
cheers
Maybe a preamp with gain will help some
I guess the safe limit with the input/driver stage is about 50watt, but Im only guessing
Do you realise the very expencive power consumption from 100watt classA
I guess the safe limit with the input/driver stage is about 50watt, but Im only guessing
Do you realise the very expencive power consumption from 100watt classA
hi mate
im running a meridian pre with adjustable gain on each output so i can easily dial up more gain if needed but im not sure if that would be ideal
i do have a thought though im hoping it may be doable
i have an aleph 5 built with brian gt mini aleph boards the ones that have flying leads to the seperate output boards 6 mosfets per channel
is it possible to adapt the aleph 5 output boards to the peter daniels f5 boards using fling leads the same as my aleph setup so id have 6 output devices per f5 instead of 2
regards sheafer
im running a meridian pre with adjustable gain on each output so i can easily dial up more gain if needed but im not sure if that would be ideal
i do have a thought though im hoping it may be doable
i have an aleph 5 built with brian gt mini aleph boards the ones that have flying leads to the seperate output boards 6 mosfets per channel
is it possible to adapt the aleph 5 output boards to the peter daniels f5 boards using fling leads the same as my aleph setup so id have 6 output devices per f5 instead of 2
regards sheafer
I can't confirm whether the electrical connections of your proposal will work.
I can confirm that three times as many output devices biased to the same level (1.3A) will require three times the heatsink capacity.
I can confirm that three times as many output devices biased to the same level (1.3A) will require three times the heatsink capacity.
An option with doubling output devices is carefully explained in previous posts(many)
With upped bias to 1.8amp
thanh shows schematic in post 4153
But post 4154 is yours, so I suppose you know that one
With upped bias to 1.8amp
thanh shows schematic in post 4153
But post 4154 is yours, so I suppose you know that one
slr 5000 said:
If you havent tried more pre gain, maybe you should
Nelson explains perfectly that F5 is a low/medium gain amp, best suited for higher sensitive speakers
If low sensitive speakers you need the pre gain
Whether its optimal or not is highly relative
Multiplying the output devices will not increase the available power into 8 ohms.
+/- 24 volts rails will never give more than 25 watts.
Bridging and paralleling operations have to be considered.
+/- 24 volts rails will never give more than 25 watts.
Bridging and paralleling operations have to be considered.
Hi all
A simple advice
I have 4x68000 25V siemens capacitors. It' better to use them 2 for channel, having a dual mono from the bridge rectifier, (trafo il 17-0-17 500VA) or 2 per rail with some resistor (0.1-0.2 hom) in beetwen?
Or maybe to use only 2 and save the other for athers applications?
Thank you
A simple advice
I have 4x68000 25V siemens capacitors. It' better to use them 2 for channel, having a dual mono from the bridge rectifier, (trafo il 17-0-17 500VA) or 2 per rail with some resistor (0.1-0.2 hom) in beetwen?
Or maybe to use only 2 and save the other for athers applications?
Thank you
Hi pro,
finally you got twin secondaries?
Are they 17 volts under load?
I would keep the pi filter.
finally you got twin secondaries?
Are they 17 volts under load?
I would keep the pi filter.
Hi Bobo
The other trafo is for another project
http://www.diyaudio.com/forums/showthread.php?s=&threadid=146459
F5 is not a project, is just something to build and listen to
just need money for heatsinking, then it will be ready in 2 or 3 days.
PI isn't a bit overkill is it?
The other trafo is for another project
http://www.diyaudio.com/forums/showthread.php?s=&threadid=146459
F5 is not a project, is just something to build and listen to
just need money for heatsinking, then it will be ready in 2 or 3 days.
PI isn't a bit overkill is it?
A little help from the experienced folks here on DiyAudio if you please.
I'm not sure of what I should be looking for when comparing capacitors for the Power supply bank.
Low ESR= equivelant series resistance.............lower is better.
Temp. 85c or 105c= 105c lasts longer
Unted Chemi Con KMH series 22,000uf with esr 19 mOhm ~$4.90
or
Panasonic TSHA series 22.000uf with esr 23 mOhm ~$7.75
or
CDE SLP series with esr 22,000uf with esr 24 mOhms~$6.30
All are rated at 105c temp.
All are from DK. Whom I trust.
Which one do you prefer? How much difference does ESR make?
No ripple current ratings that I could find.
Any opinions or recomendations in this price range and size?
I chose this size over the 15,000uf because it has better ESR and bigger IS better!
Thanks for always having an extra seat in electronics class.
Ron
Oh, one last thing,
Happy Independence Day to all the Yanks on the Forum.
No disrespect to the lads from the U.K. 😀
I'm not sure of what I should be looking for when comparing capacitors for the Power supply bank.
Low ESR= equivelant series resistance.............lower is better.
Temp. 85c or 105c= 105c lasts longer
Unted Chemi Con KMH series 22,000uf with esr 19 mOhm ~$4.90
or
Panasonic TSHA series 22.000uf with esr 23 mOhm ~$7.75
or
CDE SLP series with esr 22,000uf with esr 24 mOhms~$6.30
All are rated at 105c temp.
All are from DK. Whom I trust.
Which one do you prefer? How much difference does ESR make?
No ripple current ratings that I could find.
Any opinions or recomendations in this price range and size?
I chose this size over the 15,000uf because it has better ESR and bigger IS better!
Thanks for always having an extra seat in electronics class.
Ron
Oh, one last thing,
Happy Independence Day to all the Yanks on the Forum.
No disrespect to the lads from the U.K. 😀
I'm learning....
I found the datasheets and the ripple curent for all of them if fairly close...~5 Amps rms.
Not really sure what that means but if they are all very close, I suppose it does not matter much.
Ron
I found the datasheets and the ripple curent for all of them if fairly close...~5 Amps rms.
Not really sure what that means but if they are all very close, I suppose it does not matter much.
Ron
Hi,
the choice seems to be 22mF//22mF or 15mF//15mF//15mF.
Both give ~+-45mF per channel.
Both will perform well.
Both will have adequate ripple capacity.
Both will cost money.
Both will take up space.
What is much more likely to sound different is the decoupling you place on the PCB.
the choice seems to be 22mF//22mF or 15mF//15mF//15mF.
Both give ~+-45mF per channel.
Both will perform well.
Both will have adequate ripple capacity.
Both will cost money.
Both will take up space.
What is much more likely to sound different is the decoupling you place on the PCB.