hifryer said:
Hi hifryerfloat said:
Float is 100% correct ... thanks Float 😀 Also Greg is just an email, away, fortunately. For me though, just a short drive, even better 🙂
Hi AndrewTAndrewT said:
I didn't know that you are now have an SKA 😀 excellent 😀
Have you posted you thoughts of your SKA, yet?
If not yet, what are your thoughts of your SKA 🙂
Thanks
Re: silent fan, corrected diagram
Grrr, there is still an error: we got R04 two times!
The 2nd one, attached to R08, and pin 12 of IC1-D, should be labeled R07 and have 39k0 Ohms...
It was muuuch too hot today... soft errors kept kreeping in.
Sorry for that, guys....
stoeffle.++++
stoeffle said:Again,
here the corrected diagram...
stoeffle.++++
Grrr, there is still an error: we got R04 two times!
The 2nd one, attached to R08, and pin 12 of IC1-D, should be labeled R07 and have 39k0 Ohms...
It was muuuch too hot today... soft errors kept kreeping in.
Sorry for that, guys....
stoeffle.++++
jacco vermeulen said:
Haha jacco 😀
It took a couple of reads to get through my thick skull.
Now, if only one of my neighbours was that piano playing Delta Goodrem. Now... she lives 10kms or 15kms away, is a GB150 powerful enough? 😉
BTW: The SKA is back on my workbench....drilling a few holes in the heatsink.
regards
Tacking on a photo of another complete gbd150. Had finished a few months ago. I like it's smoothness the best. Seems that music never pushed and kind of just comes through. Analogy would be sitting in a boat as the waves come in, feeling the rise and fall, rather than a speed boat chopping over the top.
for the price, it is killer. very easy to put together. dead silent.... don't know why gaincloners don't try something different. have done many clones, sound good but only go so far on depth and high end detail...
i used a 500va 35-0-35 toroid...8 x 3900uf 80v caps
for the price, it is killer. very easy to put together. dead silent.... don't know why gaincloners don't try something different. have done many clones, sound good but only go so far on depth and high end detail...
i used a 500va 35-0-35 toroid...8 x 3900uf 80v caps
Attachments
Hi Traw,
your PSU, is it set up as +-2caps on all four supply rails or +-4 caps on a pair of shared supply rails?
For 8ohm speakers your smoothing capacitance is a bit light.
I would recommend +-10mF/channel and you may find (hear) some benefit by increasing this to +-15mF or even +-20mF/channel.
10mF gives a PSU time constant of 80mS. This compromises the bass response since it demands an NFB RC of about <=60mS (the GB=47mS) and an input high pass of about <=40mS.
If you want/need more deeper bass then beef up the supply capacitance.
your PSU, is it set up as +-2caps on all four supply rails or +-4 caps on a pair of shared supply rails?
For 8ohm speakers your smoothing capacitance is a bit light.
I would recommend +-10mF/channel and you may find (hear) some benefit by increasing this to +-15mF or even +-20mF/channel.
10mF gives a PSU time constant of 80mS. This compromises the bass response since it demands an NFB RC of about <=60mS (the GB=47mS) and an input high pass of about <=40mS.
If you want/need more deeper bass then beef up the supply capacitance.
capacitance
AndrewT,
Could you show us how you derived those time constants please?
I haven't seen it done before.
cheers
AndrewT,
Could you show us how you derived those time constants please?
I haven't seen it done before.
cheers
Hi Fryer,
sure, no problem.
The three time constants you should consider are:- on the input, the NFB loop and the PSU. These should be scaled together and also should meet a ratio criteria to minimise the risk of low frequency instability.
Lets start with the PSU:- the Rail capacitance times the load impedance should be set to a value to ensure adequate low frequency response. Assume load =8ohm and capacitance is +-7.8mF then RC= 8*7.8e-3=62.4mS.
The NFB loop is recommended at half to one octave above the PSU RC. So you need between 31.2mS (62.4/2) and 44.1mS (62.4/root2). Ignore my precision it's there simply to allow you to see where the numbers are coming from. You have 47mS (1k0 * 47uF) which is close enough.
The input high pass filter should be a further half to one octave above the NFB loop giving a range of 23.5mS to 33.2mS.
The -3db frequency for the single pole high pass filter is F=1 / [2*Pi*RC] = 1 / [2 *3.142*0.033]=4.8Hz. For a single pole filter the -1db frequency is approximately one octave away. i.e. about 10Hz. This implies good bass response down to 10Hz (23mS>=-1db@13Hz) and thinking you may have speakers that roll off at 40Hz, you might consider that you have two octaves to spare. Well no!
You will be able to hear response roll off, as lightened bass, all the way down to about a decade below your speakers response and possibly even lower. I recommend that the input filter be a decade below below your speaker to help remove the power amp's influence from the final output. This requires an input RC>=50mS. Then working back up through the NFB @ 70mS and the PSU @ 100mS you have a fairly balanced amplifier.
If you were running a sub-bass driver or a wideband speaker with sub-bass ability then you may hear a bass improvement by doubling these RC constants yet again.
sure, no problem.
The three time constants you should consider are:- on the input, the NFB loop and the PSU. These should be scaled together and also should meet a ratio criteria to minimise the risk of low frequency instability.
Lets start with the PSU:- the Rail capacitance times the load impedance should be set to a value to ensure adequate low frequency response. Assume load =8ohm and capacitance is +-7.8mF then RC= 8*7.8e-3=62.4mS.
The NFB loop is recommended at half to one octave above the PSU RC. So you need between 31.2mS (62.4/2) and 44.1mS (62.4/root2). Ignore my precision it's there simply to allow you to see where the numbers are coming from. You have 47mS (1k0 * 47uF) which is close enough.
The input high pass filter should be a further half to one octave above the NFB loop giving a range of 23.5mS to 33.2mS.
The -3db frequency for the single pole high pass filter is F=1 / [2*Pi*RC] = 1 / [2 *3.142*0.033]=4.8Hz. For a single pole filter the -1db frequency is approximately one octave away. i.e. about 10Hz. This implies good bass response down to 10Hz (23mS>=-1db@13Hz) and thinking you may have speakers that roll off at 40Hz, you might consider that you have two octaves to spare. Well no!
You will be able to hear response roll off, as lightened bass, all the way down to about a decade below your speakers response and possibly even lower. I recommend that the input filter be a decade below below your speaker to help remove the power amp's influence from the final output. This requires an input RC>=50mS. Then working back up through the NFB @ 70mS and the PSU @ 100mS you have a fairly balanced amplifier.
If you were running a sub-bass driver or a wideband speaker with sub-bass ability then you may hear a bass improvement by doubling these RC constants yet again.
Hmmm... i have some nichicon 10000uf 63v laying around, same size. I had rectifier into 4 x 3900, 2 per rail. then 0.1 ohm leading over to 2 more 3900, 1 per rail for each board.
I had arrived at that approach levering p101 recommended ps except with touch lower values but doubled up central/initial bank.
don't want to kill board just yet.... changing all over would be overkill and sonically harmful(?) with total 80000uf.... should just swap up mid / starting bank?
I had arrived at that approach levering p101 recommended ps except with touch lower values but doubled up central/initial bank.
don't want to kill board just yet.... changing all over would be overkill and sonically harmful(?) with total 80000uf.... should just swap up mid / starting bank?
Well I've listened to my GB150D with the zener PSRR mod done, for about a week now, and have formed some definite impressions.
First there is more air around the instruments and vocals and the background is darker . Cymbals seem to shimmer more and there's greater spaciousness and realism. It's effectively peeled back another layer revealing even more of the performance. A worthwhile mod for sure. I really didn't think CD's could sound this lifelike.
This amp would be hard to beat at any price - and it's cost me less than $500 all up - and so simple to build I'd even recommend it to a first time DIYer.
First there is more air around the instruments and vocals and the background is darker . Cymbals seem to shimmer more and there's greater spaciousness and realism. It's effectively peeled back another layer revealing even more of the performance. A worthwhile mod for sure. I really didn't think CD's could sound this lifelike.
This amp would be hard to beat at any price - and it's cost me less than $500 all up - and so simple to build I'd even recommend it to a first time DIYer.
My review:
I just finished my GB150D in the beginning of August. Wen I first turn it on after the tweaking, the sound was already promising.
My holidays came, and I couldn't finished all the tests on the amp.
But I must say that I went to holidays with is music in my hears.... Even in the state of "virgin" the musicality was remarkable.
Now, after coming from holidays I add all the time I needed to finish the tests and let it burn for at least two days.
I started my tests with 20.000 uF of capacitance (2x 10.000uF per rail), C6 and C7 bypassed with 100nF and Daewoo caps instead of the Lelon.
In the first 30m to 45m the amp as some detail but lacks some transparency in each regime( bass, med and treble ).
You can notice clearly the high frequencies, the bottom or the meds.... even my wife could identify the bass or the cymbal.
The problem is that in this first 30m you can notice the drums but they seem "sharp" and "rude" with the toms mixed up.
But as we all know, any amp sounds much better when warm! So what's the point of describing the first 30m ?
«It's just for the readers to notice that even cold, this amp smacks some "pseudo amps" that you find in the market.»
And how it sound after the first 45m ?
Ohhh I think it's better that you grab a coup of red wine, because this appeals to the senses.
As I told before, even at cold you can notice the different frequencies. Now the question is, what got better ?
Ohh my friends, the "sharp" and "rude" sound was "converted" to a song of a mermaid.
The voice is very detailed and you can easily notice it between the instruments, even the backing vocals gain a decent sound stage.
The drums became more clear and the toms got very refined.
I played some hard music ( Slayer - Christ Illusion ) were Dave Lombardo kick *** in the drums and all the sound was very clear and detailed.
It only lacked some deep bass(for my taste), easily resolved with more ~40.000uF per rail 🙂 (rubycon 6800uF 63V black caps in parallel)
After this mod the bass started to punch my stomach ( ohh yes! it kicked *** hehehehe )
After that, was time to move to another CD, something with great vocals and also great drums. The chosen was Patricia's Barber - Cafe Blue CD, great CD of Jazz.
Once again the the bass kicked and the sound was very detailed, also. But this time I notice how the amp performed with high frequencies vocals and once again, I could easily notice the cymbals from the voice at high tones.
Apart of the detail the musicality of the amp it's very good and in my case it only lacks in bass dynamics but this could be due to a test without pre amp 😉
Attention, I am not saying that the dynamics it's poor, it's just not at the same level as the detail, for example.
Resuming, it's definitively a great amp and I am just awaiting for the chance of comparing it with the Krell (KAV 300) of a friend.
I think will be a tough confront.... and if I am in doubt, imagine how good this amp is and range it reaches.
And this time I will take a pre amp 🙂
Will the Krell be put in a corner ? I hope to find the question to that answer in the next few days.
Till then, cheers.
PS:
Setup:
~60.000uF
650VA trafo, +42Vcc and -42Vcc
I just finished my GB150D in the beginning of August. Wen I first turn it on after the tweaking, the sound was already promising.
My holidays came, and I couldn't finished all the tests on the amp.
But I must say that I went to holidays with is music in my hears.... Even in the state of "virgin" the musicality was remarkable.
Now, after coming from holidays I add all the time I needed to finish the tests and let it burn for at least two days.
I started my tests with 20.000 uF of capacitance (2x 10.000uF per rail), C6 and C7 bypassed with 100nF and Daewoo caps instead of the Lelon.
In the first 30m to 45m the amp as some detail but lacks some transparency in each regime( bass, med and treble ).
You can notice clearly the high frequencies, the bottom or the meds.... even my wife could identify the bass or the cymbal.
The problem is that in this first 30m you can notice the drums but they seem "sharp" and "rude" with the toms mixed up.
But as we all know, any amp sounds much better when warm! So what's the point of describing the first 30m ?
«It's just for the readers to notice that even cold, this amp smacks some "pseudo amps" that you find in the market.»
And how it sound after the first 45m ?
Ohhh I think it's better that you grab a coup of red wine, because this appeals to the senses.
As I told before, even at cold you can notice the different frequencies. Now the question is, what got better ?
Ohh my friends, the "sharp" and "rude" sound was "converted" to a song of a mermaid.
The voice is very detailed and you can easily notice it between the instruments, even the backing vocals gain a decent sound stage.
The drums became more clear and the toms got very refined.
I played some hard music ( Slayer - Christ Illusion ) were Dave Lombardo kick *** in the drums and all the sound was very clear and detailed.
It only lacked some deep bass(for my taste), easily resolved with more ~40.000uF per rail 🙂 (rubycon 6800uF 63V black caps in parallel)
After this mod the bass started to punch my stomach ( ohh yes! it kicked *** hehehehe )
After that, was time to move to another CD, something with great vocals and also great drums. The chosen was Patricia's Barber - Cafe Blue CD, great CD of Jazz.
Once again the the bass kicked and the sound was very detailed, also. But this time I notice how the amp performed with high frequencies vocals and once again, I could easily notice the cymbals from the voice at high tones.
Apart of the detail the musicality of the amp it's very good and in my case it only lacks in bass dynamics but this could be due to a test without pre amp 😉
Attention, I am not saying that the dynamics it's poor, it's just not at the same level as the detail, for example.
Resuming, it's definitively a great amp and I am just awaiting for the chance of comparing it with the Krell (KAV 300) of a friend.
I think will be a tough confront.... and if I am in doubt, imagine how good this amp is and range it reaches.
And this time I will take a pre amp 🙂
Will the Krell be put in a corner ? I hope to find the question to that answer in the next few days.
Till then, cheers.
PS:
Setup:
~60.000uF
650VA trafo, +42Vcc and -42Vcc
GB150 - for portuguese
Hello all! 🙂
I'm a newcommer to this forum and I just ordered the GB150D kit from Greg. It will be my first DIY audio project and I'm very excited with the reviews and listening impressions I'm reading here. I will post mine to when the kits arrive and I build them.
However, and I apologize for this kinda off-topic, I have a question for the portuguese who received the kit from Greg in Australia:
How long did it take to arrive to you? I'm asking this because I've heard the worst of the Lisbon constumes service. Did it stayed very long in the "alfandega"? And did you pay import taxes?
I'd really apreciate an answer and I thank you in advance.
Regards,
Paulo.
Hello all! 🙂
I'm a newcommer to this forum and I just ordered the GB150D kit from Greg. It will be my first DIY audio project and I'm very excited with the reviews and listening impressions I'm reading here. I will post mine to when the kits arrive and I build them.
However, and I apologize for this kinda off-topic, I have a question for the portuguese who received the kit from Greg in Australia:
How long did it take to arrive to you? I'm asking this because I've heard the worst of the Lisbon constumes service. Did it stayed very long in the "alfandega"? And did you pay import taxes?
I'd really apreciate an answer and I thank you in advance.
Regards,
Paulo.
Re: GB150 - for portuguese
Hi Paulo.
Can't tell you exactly how long it took arriving, because Greg was out of kit's when I order.
But I think that in one week you will get your amp.
I didn't pay any taxes, the packaged came directly to my house
😎
BTW, where do you by your parts in Lisbon ?
Here is an old picture from the tweaking phase.
paufon said:
Hi Paulo.
Can't tell you exactly how long it took arriving, because Greg was out of kit's when I order.
But I think that in one week you will get your amp.
I didn't pay any taxes, the packaged came directly to my house
😎
BTW, where do you by your parts in Lisbon ?
Here is an old picture from the tweaking phase.
Attachments
Hi Xelb,
did you cut a gap between the heatsink insulators for the diode temp monitors to poke through?
or
did you drill the holes as being recommended by Greg?
did you cut a gap between the heatsink insulators for the diode temp monitors to poke through?
or
did you drill the holes as being recommended by Greg?
My heatsink was already driled.
D5 as D6 are in contact with the heatsink.
When I weld them, I forced them to be inclined in to the direction of the heatsink.
I did the same with the transistors.
This way they are in pressure with the heatsink and in full contact.
D5 as D6 are in contact with the heatsink.
When I weld them, I forced them to be inclined in to the direction of the heatsink.
I did the same with the transistors.
This way they are in pressure with the heatsink and in full contact.
You had luck. Not all packages are selected for inspection and taxes. But some are and probably mine has been selected because I'm waiting now for 8 working days and nothing. If the package is at costums storage I'll have to wait months! And pay a lot... 🙁
I'm starting to think the best now is to buy parts were in Portugal and make the pcbs myself because I don't feel like waiting so long to build the amp. Greg has send me the circuit diagram. May be I can build it. But where can I find matched transistors? Can I use unmatched ones?
Sometimes I buy parts at Dimofel but I think they are expensive. I also order were: www.leiritronica.com
BTW, nice picture of your amp! 🙂
Cheers,
Paulo.
Don't worry, you should be receiving your package.
Maybe on the next week you will get it.
The "alfandega" don't take long.
Sometimes I buy things from USA and they contact me when the package arrives.
Maybe it's taking more because of the airmail control 😉
I also go to Dimofel and to a small shop near "Alameda Afonso Henriques".
Do you have any ideia about the PSU you will use in this amp ?
It deserves good caps, big capacitance and a nice transformer 😎
Maybe on the next week you will get it.
The "alfandega" don't take long.
Sometimes I buy things from USA and they contact me when the package arrives.
Maybe it's taking more because of the airmail control 😉
I also go to Dimofel and to a small shop near "Alameda Afonso Henriques".
Do you have any ideia about the PSU you will use in this amp ?
It deserves good caps, big capacitance and a nice transformer 😎
GB150
Hello all,
I will be buidling the GB150.
Couple of questions on power supplies:
1. Is there a minimum V+- level required for the rails?
2. Anyone have an idea of what is optimum V+-?
3. Do bigger transformers have an impact on sound quality?Again, what is an optimum size?
4. Is it worthwhile (i.e. makes the sound better) having one transformer per channel?
5. Is there any benefit in regulating the power supply?
6. What about regulating just the front-end?
7. Does the use of discrete diodes (vs diode bridges) make a difference?
8. What about fast soft recovery diodes?
9. Anyone here tried carlosfm snubbers on the power supply?
Thanks for any guidance!
Ryan
Hello all,
I will be buidling the GB150.
Couple of questions on power supplies:
1. Is there a minimum V+- level required for the rails?
2. Anyone have an idea of what is optimum V+-?
3. Do bigger transformers have an impact on sound quality?Again, what is an optimum size?
4. Is it worthwhile (i.e. makes the sound better) having one transformer per channel?
5. Is there any benefit in regulating the power supply?
6. What about regulating just the front-end?
7. Does the use of discrete diodes (vs diode bridges) make a difference?
8. What about fast soft recovery diodes?
9. Anyone here tried carlosfm snubbers on the power supply?
Thanks for any guidance!
Ryan
- Status
- Not open for further replies.