Carlos,
These fastons from PE are a nice quality... http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=095-302
These fastons from PE are a nice quality... http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=095-302
Hi guys, just spend most of the weekend ripping the guts out of my Akai AM55 - nothing was left except the chassis.
Thought it was a cool idea to build a state of the art modern amp into a vintage integrated amp 😀
I've built the following in the case:
* Hypex Softstart
* Hypex HG Supply
* Hypex TR501 (what a large beast!)
* 2xUcD400AD
* A few connectors & leds & switches😀
Todo:
* Four Input cinches, I couldn't recover the existing ones on the AM55.
* Switch for direct amp input or an input going to the volume control (simple pot now, something better later)
Sound reports come later - there's a little more rest in the sound, and bass is lots tighter (not louder)
Everything works as it should, but had a few problems (mostly fixed), can anyone help me with these:
* With open or connected input leads (i'm using unbalanced) I had the dreaded 50hz hum. Spent a lot of time looking for loose leads & such, until (almost by accident) I connected the signal grounds of both input leads together. Hum gone?! I disconnect the grounds from eachother (while still connected to any source) -> hum. What gives? In my test setup with my homebrew supply I didn't have this problem. Might I have an error somewhere?
* Softstart & HG supply work fine together. I can switch on and off (almost) without a click. HOWEVER, when the caps are drained and the amp is disconnected from mains, and the SS starts the first time, I get quite a dull click in my speakers. Subsequent on/off cycles aren't a problem. I guess the caps are still loading when the amps come on the first time.
Can I delay the start of the modules somewhat, one way or the other? I was thinking a little cap or something, but the on signal is done by grounding, how can I slow this for say, half a second or something, with a minimum of components? Does anyone else with the Hype HG PS + SS have this?
Thought it was a cool idea to build a state of the art modern amp into a vintage integrated amp 😀
I've built the following in the case:
* Hypex Softstart
* Hypex HG Supply
* Hypex TR501 (what a large beast!)
* 2xUcD400AD
* A few connectors & leds & switches😀
Todo:
* Four Input cinches, I couldn't recover the existing ones on the AM55.
* Switch for direct amp input or an input going to the volume control (simple pot now, something better later)
Sound reports come later - there's a little more rest in the sound, and bass is lots tighter (not louder)
Everything works as it should, but had a few problems (mostly fixed), can anyone help me with these:
* With open or connected input leads (i'm using unbalanced) I had the dreaded 50hz hum. Spent a lot of time looking for loose leads & such, until (almost by accident) I connected the signal grounds of both input leads together. Hum gone?! I disconnect the grounds from eachother (while still connected to any source) -> hum. What gives? In my test setup with my homebrew supply I didn't have this problem. Might I have an error somewhere?
* Softstart & HG supply work fine together. I can switch on and off (almost) without a click. HOWEVER, when the caps are drained and the amp is disconnected from mains, and the SS starts the first time, I get quite a dull click in my speakers. Subsequent on/off cycles aren't a problem. I guess the caps are still loading when the amps come on the first time.
Can I delay the start of the modules somewhat, one way or the other? I was thinking a little cap or something, but the on signal is done by grounding, how can I slow this for say, half a second or something, with a minimum of components? Does anyone else with the Hype HG PS + SS have this?
Yves,
Can you post a drawing of how you have connect all wires?
We always use this XLR connector as an input connector; http://nl.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=724518&N=0
Thereby we have a XLR symmetrical to Cinch asymmetrical cable. How to make this cable is explained many times.
Jan-Peter
Can you post a drawing of how you have connect all wires?
We always use this XLR connector as an input connector; http://nl.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=724518&N=0
Thereby we have a XLR symmetrical to Cinch asymmetrical cable. How to make this cable is explained many times.
Jan-Peter
Yves Smolders said:
Mine did the same when I was using non insulated cinchs and had grounded the cabinet. After switching to insulated RCA's and not grounding the cabinet, (only one ground connection in the middle of my capacitors) everything works great.
Groundloops s....s
/avr300
No grounded cabinet here. For the moment everything is elecrically disconnected from the cabinet, not even a ground cable to the mains for the moment. After putting the grounds of the signals together at the far end of the cable, hum is gone completely, nothing left. I'm going to keep it like this, as my receivers' pre-outs are also common grounded.
More info on the switching on - after a while in standby, when switching on I still have a small click in the speakers (after the softstart delay), this is really bothering me - one of the reasons of buying the Hypex HG PS is that it has a pin for controlling the modules' mute... I'm going to contact Hypex to see if I might have done anything wrong.
The audioquality is amazing though - I went from a standard transformer with 35v secondaries and a pair of BC 156 22.000uF's to 500VA toroidal & the hypex HG and it's an improvement so far. It is different - it's not an even further improvement from what I was used to, but like a "sideway" of some sorts. Maybe it's my head or burn-in or whatever, but the sound is silkier now, before with the other power supply it was more analytical, of course I need to listen for a few hours to be sure.
The massive transformer does help in the bass region for sure, very defined basslines, quick and good impact on the drums (even after quickly repeated "kicks")
Highs seem the same more or less, voices and woodwinds are very good, brass is different, not better or worse, just different.
I'm happy with the investment in the Hypex PS parts, I just hope to get rid of the power-on clicks.
More info on the switching on - after a while in standby, when switching on I still have a small click in the speakers (after the softstart delay), this is really bothering me - one of the reasons of buying the Hypex HG PS is that it has a pin for controlling the modules' mute... I'm going to contact Hypex to see if I might have done anything wrong.
The audioquality is amazing though - I went from a standard transformer with 35v secondaries and a pair of BC 156 22.000uF's to 500VA toroidal & the hypex HG and it's an improvement so far. It is different - it's not an even further improvement from what I was used to, but like a "sideway" of some sorts. Maybe it's my head or burn-in or whatever, but the sound is silkier now, before with the other power supply it was more analytical, of course I need to listen for a few hours to be sure.
The massive transformer does help in the bass region for sure, very defined basslines, quick and good impact on the drums (even after quickly repeated "kicks")
Highs seem the same more or less, voices and woodwinds are very good, brass is different, not better or worse, just different.
I'm happy with the investment in the Hypex PS parts, I just hope to get rid of the power-on clicks.
I have five 1000VA potted transformers with 2x40Aac secondarys which would give a 55Vdc and several 15,000uF @ 80V caps. How much should I use per rail with the UCD400 given this unusually oversized transformer assuming only one UDC400 per transformer? The options are 15,000uF, 30,000uF, 45,000uF, 60,000uF….
oversized transformers
Hi LBHajdu
Well, one of those transformers with the 2 secondaries will be enough for 2 UCD400's. Wanta sell one of those transformers?
Ray
Hi LBHajdu
Well, one of those transformers with the 2 secondaries will be enough for 2 UCD400's. Wanta sell one of those transformers?
Ray
LBHajdu said:
15 000uF sounds good, but it won't hurt to test with higher, use what sounds best for you.
Regards,
Chris
LBHajdu said:
As I mentioned in another thread, split them up and stick a 0.3 ohm resistor between the two caps on each rail. PSUDesigner simulations show that you can drop rail noise 15 dB for a 30,000 uf - 0.3 ohm - 30,000 uf combination, and higher-order harmonics look much nicer as well, which makes sense since that structure is effectively a capacitor-input supply followed by a 16 Hz lowpass filter. You do need a fairly hefty resistor, and limiting capacitor inrush current at turn-on is also a good idea to avoid overstressing the R.
Cheers,
Francois.
Francois,
Have you tried a before/after listening test with your power supply changes? Wouldn't that configuration limit the 'speed' of the power supply, reducing the peak instantaneous current it could deliver - ie. affecting the amps dynamics?
Regards,
Dean
Have you tried a before/after listening test with your power supply changes? Wouldn't that configuration limit the 'speed' of the power supply, reducing the peak instantaneous current it could deliver - ie. affecting the amps dynamics?
Regards,
Dean
Alright, I guess the consensus of opinion is that I have to test it out. I have to save up some money to order the UCD400’s now. If I am correct the price of the UCD400 buying direct from Hypex now is 100euro with exchange rates right now that’s only 120USD, that’s not bad.
I will also try the crc filter. But it will take me some time as I plan to design the inrush limiter myself.
Ray,
I would consider selling one but I don’t think you want it. I was really worried about noise both mechanical and electromagnetic radiated so I ask that the transform have 3 turns of silicon steal wrapped around it for shielding and then that it be epoxyed potting an a steel case under pressure. They where custom wound in England and shipped to me in the US.
It does not have independent secondarys for each amp as I planned it to be a mono block, And I didn’t originally plan on using it for this amp. The secondary layout is like this (0Vac – 40Vac – 50vAC) (0Vac – 40Vac – 50vAC).
You’re probably thinking, alright that sounds pretty good, so why don’t I want them. Answer, is because they cost me $380us each. With shipping to you it’s going to cost like 400.
The basic 1000VA vanilla plastic tape wrapped transformer from Plitron will only cost $120.00 + sh.
Leve
I will also try the crc filter. But it will take me some time as I plan to design the inrush limiter myself.
Ray,
I would consider selling one but I don’t think you want it. I was really worried about noise both mechanical and electromagnetic radiated so I ask that the transform have 3 turns of silicon steal wrapped around it for shielding and then that it be epoxyed potting an a steel case under pressure. They where custom wound in England and shipped to me in the US.
It does not have independent secondarys for each amp as I planned it to be a mono block, And I didn’t originally plan on using it for this amp. The secondary layout is like this (0Vac – 40Vac – 50vAC) (0Vac – 40Vac – 50vAC).
You’re probably thinking, alright that sounds pretty good, so why don’t I want them. Answer, is because they cost me $380us each. With shipping to you it’s going to cost like 400.
The basic 1000VA vanilla plastic tape wrapped transformer from Plitron will only cost $120.00 + sh.
Leve
I'm a UK resident. Can I ask you who you used to get your transformers from please?
Thanks
Ed W
Thanks
Ed W
ewildgoose,
I know if they will sell to you, but give it a try anyway. Look up Siga Electronics.
Leve
I know if they will sell to you, but give it a try anyway. Look up Siga Electronics.
Leve
deandob said:
Good point. No, I haven't done a side-by-side. I'll consider it in my copious free time.
The instantaneous power available from the 30,000 uF cap after the resistor is still considerable - at 50 V out it's 37.5 joules per rail post-R. It doesn't sound like much, but it's sufficient when the supply is replenished 120 times a second, and really, most amps out there don't have 30,000 uF after the rectifiers anyway, let alone in a two-stage configuration. Also remember the resistor is comparable to the transformer equivalent resistance, but the transformer resistance ends up being more important since the current the transformer delivers to the rectifier is a series of narrow pulses with high peak current, whereas the current flow through the inter-cap R is comparatively steady, hence the voltage drop through the secondary is quite a bit higher than the drop through the resistor. At that point I'd say the transformer would be limiting dynamics more than the resistor would.
My experience tells me I perceive dynamics better against a background that's as black as possible, so my real concern here is to reduce rail noise. Since the RC combination is a 16 Hz first-order lowpass, the 120 Hz component is down about 17 dB, 180 Hz 21 dB, etc. The real win comes in the higher buzz frequencies, which get attenuated right when they start getting more perceptible. I mean, these modules have decent PSRR anyway, but why not help them along a little bit, especially since the speaker is 94 dB/W/m.
Cheers,
Francois.
Interesting analysis, thanks Francois.
Let us know how you go with an A/B comparison with and without the series resistor.
Regards,
Dean
Let us know how you go with an A/B comparison with and without the series resistor.
Regards,
Dean
RC-filtering
In all our designs, and recently in the Hypex application as well, we use two resistors: one between the transformer winding and one between the first and the second capacitor. In power amps we mostly use 0,1 Ohm and this hardly compromises the dynamic character but it improves isolation from the mains disturbances and the "sparks" from the rectifier as well.
Have a look at http://www.hawkaudio.com/tips.htm (fig. 3)
In all our designs, and recently in the Hypex application as well, we use two resistors: one between the transformer winding and one between the first and the second capacitor. In power amps we mostly use 0,1 Ohm and this hardly compromises the dynamic character but it improves isolation from the mains disturbances and the "sparks" from the rectifier as well.
Have a look at http://www.hawkaudio.com/tips.htm (fig. 3)
On a semi - related question, I don’t understand why Hypex puts a standby pin on these amps if they use up such little power as is? Was it intended as a mute for subwoofer applications when the thing is in stand by? Does it help with turn on thumps? Or was it just super easy to implement, and so they did.
The Hypex HG supply has connections to control the modules muting. So when the power goes off, the modules are muted. In a test setup, I didn't have this, and when the power dies away from the caps, you hear the weirdest sounds in your speakers before the modules cut out.
Yves Smolders said:
Yes, I noticed that with the UcD400 as well. The old UcD180s that I have switch off very gently when the power supply voltage drops so I have no issue to drive my tweeters directly with an UcD180 and without muting. With the UcD400 I think I will need to use the muting option to protect my tweeters during switch-off. So I need to make something for that. I think I will use one of the signals coming from my SMPS to make a switching signal for the mute.
Best regards
Gertjan