Crazoff,
Yes, you could drop the input cap. However, be aware that if you ever set any signal input with DC, the offset may seriously upset the AN output offset and damage your speaker. Be warned; dropping caps should be done carefully.
Valery's moniker is Vaichenko.
Otto,
Yes, cold in Melbourne, <3C at night, but no snow, and very little virus......
The amp was designed to go down to 6R that is why I suggest 15% only.
I would always recommend four MF35-100 heatsinks for two channels. Each Conrad heatsink will give you 0.3C/Watt, so with 55W dissipation on each device you have 16.5C over ambient. This assumes that the single output device on each heatsink is thermally distributing the heat evenly across the sink. This is not easy to achieve, of course.
With a hot 30C room in summer, you could assume that the amps will run at 47C, 17C over ambient. Note from the conradheatsink.com website that the temperature rating is at 80C, so for 25C you should derate by 1.33. This makes a 0.22 rating actually 0.29, so allow for lower temperatures.
With Class A you cannot keep temperatures too low. MTBF of the output devices is heavily influenced by junction temp which should be around 20C over case temperature.
HD
Yes, you could drop the input cap. However, be aware that if you ever set any signal input with DC, the offset may seriously upset the AN output offset and damage your speaker. Be warned; dropping caps should be done carefully.
Valery's moniker is Vaichenko.
Otto,
Yes, cold in Melbourne, <3C at night, but no snow, and very little virus......
The amp was designed to go down to 6R that is why I suggest 15% only.
I would always recommend four MF35-100 heatsinks for two channels. Each Conrad heatsink will give you 0.3C/Watt, so with 55W dissipation on each device you have 16.5C over ambient. This assumes that the single output device on each heatsink is thermally distributing the heat evenly across the sink. This is not easy to achieve, of course.
With a hot 30C room in summer, you could assume that the amps will run at 47C, 17C over ambient. Note from the conradheatsink.com website that the temperature rating is at 80C, so for 25C you should derate by 1.33. This makes a 0.22 rating actually 0.29, so allow for lower temperatures.
With Class A you cannot keep temperatures too low. MTBF of the output devices is heavily influenced by junction temp which should be around 20C over case temperature.
HD
Last edited:
Examine his amplifier designs.
He is a brilliant amp designer and a lovely chap.
Worth a long look, and since you live in his city, probably worth a visit too.
He may be travelling, wait a few days.
Hugh
He is a brilliant amp designer and a lovely chap.
Worth a long look, and since you live in his city, probably worth a visit too.
He may be travelling, wait a few days.
Hugh
I have built his VHEX+ amp. Easy to build and great sounding. Nice powerful 100w Class AB with a nice harmonic distortion profile.
But I don’t think he does Class A amps that I know of. If it’s Class A you want, this is the right thread.
But I don’t think he does Class A amps that I know of. If it’s Class A you want, this is the right thread.
(Sorry I took so long to respond, too much on)
Thank you, Hugh
I was aware that with Class A, heat sinking is far more critical - but you’ve given some quantitative facts for me to find an appropriate solution
… Which just leaves that question: into 6 ohms - what would be the approximate/ likely output?
Cheers
Thank you, Hugh
I was aware that with Class A, heat sinking is far more critical - but you’ve given some quantitative facts for me to find an appropriate solution
… Which just leaves that question: into 6 ohms - what would be the approximate/ likely output?
Cheers
AN into 6 ohm load:
almost the same, but clip at a lower voltage because the current limit is unchanged.
With 2A quiescent, into a 6R load you would limit at 12Vpeak (on the negative half cycle, not the positive, which is not limited by the CCS system).
So, that would be 24Vpp output into 6R, and this would be 12W.
However, speakers have widely varying impedance, so although it might be limited to 12W nominal, it would depend WHICH frequency is clipping. Generally, however, the highest levels in music are below 200Hz, but most speakers have very high impedance peaks at resonance points. At this frequency (depending on your speaker of course) the amp will be able to clip at far higher voltage if the resonant peak corresponds with the frequency of high level, although this is unlikely to happen.
Still, 12W ain't bad......... particularly if the speaker is better than 93dB/watt/metre.
Hugh
almost the same, but clip at a lower voltage because the current limit is unchanged.
With 2A quiescent, into a 6R load you would limit at 12Vpeak (on the negative half cycle, not the positive, which is not limited by the CCS system).
So, that would be 24Vpp output into 6R, and this would be 12W.
However, speakers have widely varying impedance, so although it might be limited to 12W nominal, it would depend WHICH frequency is clipping. Generally, however, the highest levels in music are below 200Hz, but most speakers have very high impedance peaks at resonance points. At this frequency (depending on your speaker of course) the amp will be able to clip at far higher voltage if the resonant peak corresponds with the frequency of high level, although this is unlikely to happen.
Still, 12W ain't bad......... particularly if the speaker is better than 93dB/watt/metre.
Hugh
Last edited:
I understand your very good points around impedances varying with frequencies including (I’d forgotten about resonant peaks) and where the greatest amounts of energy are, ie situations vary - but was very surprised that clipping occurs at just ~ 30 % of output …
As you of course know, the difference between 12 watts and around 40 watts of usable power is equivalent to a significant difference in speaker volume, ~ 5 dB
Is there any combination of settings of eg quiescent current and/ or rail voltages that mean clipping is closer to the actual output?
Cheers
Last edited:
Hello, everyone. Let me ask you a couple of questions. Can I use Toshiba 2SA970 transistors as V111/112? It seems like good transistors. How critical is the nominal value of the capacitors C114/115 (15 pF) and C125/126 (5pF). Cornell Dubilier 12pF and 4pF are available. And with regard to the capacitor C121/1210. The diagram shows the values of 4700pF or 0.033 uf. This is a very big difference in capacity.Thanks.
Last edited:
I would just go with the posted “as built” schematic in Post 1 of the GB thread. I used 4700pF (4.7nF). I am not sure where the 0.033uF (33nF value for the SMT version came from).
For the input LTP v111/112 almost any small signal audio PNP BJT can be used. Just observe the pin configurations and you might need to adjust the set points resistors to get about 2mA bias current through both legs. 2N5401 or BC560 could work as well. Just check pins and max voltage ratings.
For the input LTP v111/112 almost any small signal audio PNP BJT can be used. Just observe the pin configurations and you might need to adjust the set points resistors to get about 2mA bias current through both legs. 2N5401 or BC560 could work as well. Just check pins and max voltage ratings.
Last edited:
Hmm I had the same cap, replaced it with 4700 and went Silver mica on the 10 pf and changed bd140 to ks992. Now the upper end sounds right. Getting close to casing it up.
My picture rotate randomly when post if some can tell why (iPad) appreciate it
My picture rotate randomly when post if some can tell why (iPad) appreciate itAttachments
Last edited:
Nice bench Wirewiggler! I see we use the same DMM’s and DAPs 🙂 that’s a neat arrangement you have there. Is that a big EI transformer in the middle?
The cap choice you made and changing to KSA992 - I don’t recall when a BD140 was used because I used a 992 also.
The cap choice you made and changing to KSA992 - I don’t recall when a BD140 was used because I used a 992 also.
X,
It is a cpu cooler, I have a pair of SLB power supplies in one case and the Alpha Nirvana’s in their own case - part of my dare to be stupid themes. I was surprised how quiet and effective the cpu coolers are, garage was 93 degrees f and cpu coolers 45 c hanging just below full speed.
It is a cpu cooler, I have a pair of SLB power supplies in one case and the Alpha Nirvana’s in their own case - part of my dare to be stupid themes. I was surprised how quiet and effective the cpu coolers are, garage was 93 degrees f and cpu coolers 45 c hanging just below full speed.
Attachments
Neat, compact AN amplifier module Wirewiggler, looks great!
The Noctua fans are key for almost silent forced cooling.
The Noctua fans are key for almost silent forced cooling.
Hi, what's C134 for? I made various circuits with this kind of ccs and it wasn't needed for stability, C133 isn't sufficient?
Vunce,
Thanks, if my metal skills were better I would have made it a little more compact between the fan and cooler. With the foam shields in it is fine, though I was testing the second amp and it was 10 degrees hotter and realized I had not place the sensor probe on heat sink so it was running at 40% speed. First listening impression Joe Bonamassa - blues of desperation I don’t usually care much for synthesized drums but you have to admit they sound really good on this amp. In the seventies when I sold Hifi ( actually midfi) we used to call it slam but now I realize it is dampening factor. Sold a lot of loose Cerwin Vega speakers with amps with good dampening. Put on some Ted Nugent and they would walk out the store on their own.
Bill
Thanks, if my metal skills were better I would have made it a little more compact between the fan and cooler. With the foam shields in it is fine, though I was testing the second amp and it was 10 degrees hotter and realized I had not place the sensor probe on heat sink so it was running at 40% speed. First listening impression Joe Bonamassa - blues of desperation I don’t usually care much for synthesized drums but you have to admit they sound really good on this amp. In the seventies when I sold Hifi ( actually midfi) we used to call it slam but now I realize it is dampening factor. Sold a lot of loose Cerwin Vega speakers with amps with good dampening. Put on some Ted Nugent and they would walk out the store on their own.
Bill
Ok take that back about synth drummer it was Anton fig with a second drummer, just did not think it was humanly possible. His produce Keith Shirley said “ he wanted to through in a cat with the pigeons”.
Last edited:
