I am planning a 3 way tri-amplified system. My drivers will be:
Bass - Monacor SPH-275C, 8 ohm, 91db sensitivity, up to 420Hz
Mid - Audax HM100Z0, 8 ohm, 93db sensitivity, up to 4.3Khz
Treble - Hi-Vi RT2E-A, 8 ohm, 93db sensitivity (more like 96 up to 10Khz)
On amplifier choice I was originally considering the ESP P101 amp for bass and mid, with lower supply rails on the mid (to give about 180WPC and 80WPC respective powers). For treble I was considering the ESP DOZ class A amp with 40v rail (about 15W RMS).
Now all of this together costs a lot in parts and in time (and construction of the chassis is going to be difficult without a known source of cut 4mm aluminium panels). I have seen these commercial amps which seem to respesent great value, the Alesis RA series:
http://www.alesis.com/product.php?id=32
With the RA500 on bass, RA300 on mid and RA150 on treble I can have a full amplification system for a great price. My question is how much worse (if at all) would it be? Would there be a more optimum combination of amps (like the studio amps for bass and mid and the class A on treble)? Any input welcome
Bass - Monacor SPH-275C, 8 ohm, 91db sensitivity, up to 420Hz
Mid - Audax HM100Z0, 8 ohm, 93db sensitivity, up to 4.3Khz
Treble - Hi-Vi RT2E-A, 8 ohm, 93db sensitivity (more like 96 up to 10Khz)
On amplifier choice I was originally considering the ESP P101 amp for bass and mid, with lower supply rails on the mid (to give about 180WPC and 80WPC respective powers). For treble I was considering the ESP DOZ class A amp with 40v rail (about 15W RMS).
Now all of this together costs a lot in parts and in time (and construction of the chassis is going to be difficult without a known source of cut 4mm aluminium panels). I have seen these commercial amps which seem to respesent great value, the Alesis RA series:
http://www.alesis.com/product.php?id=32
With the RA500 on bass, RA300 on mid and RA150 on treble I can have a full amplification system for a great price. My question is how much worse (if at all) would it be? Would there be a more optimum combination of amps (like the studio amps for bass and mid and the class A on treble)? Any input welcome
There are many nice amps(these amps receive good reviews ) around see these sites:
http://www.aksaonline.com/
and
http://www.ska-audio.com/diy/index.html
I don't think there are expensive, they reasonably priced, furthermore, your speakers are fairly effcient, and you are tri-ampling.
you don't need heaps of power. 100watts per channel is plenty.
http://www.aksaonline.com/
and
http://www.ska-audio.com/diy/index.html
I don't think there are expensive, they reasonably priced, furthermore, your speakers are fairly effcient, and you are tri-ampling.
you don't need heaps of power. 100watts per channel is plenty.
Thanks for the replies!
With Tri-amping I was under the impression that higher frequencies required less power since music signals have diminishing energy content above about 3Khz? Also the baffle step effects act to boost these frequencies futhermore (meaning less required power again with an active compensation)? I don't know, I get mixed views on this.
Perhaps the very best combo would be the class-A DOZ on top, the Symasym on mid and the P101 or RA500/300 on bass? Thats at least 2 DIY amps though, not a problem so long as I can figure out the panelling for the chassis.
Is this the Symasym you used Pinkmouse?:
http://www.lf-pro.net/mbittner/Sym5_Webpage/symasym5_3.html
It fits the bill but I'd need a PCB. You don't happen to have any left over or know where to get them? I can drill them myself if I need. Did you use ON-Semi devices or Toshiba at the output?
With Tri-amping I was under the impression that higher frequencies required less power since music signals have diminishing energy content above about 3Khz? Also the baffle step effects act to boost these frequencies futhermore (meaning less required power again with an active compensation)? I don't know, I get mixed views on this.
Perhaps the very best combo would be the class-A DOZ on top, the Symasym on mid and the P101 or RA500/300 on bass? Thats at least 2 DIY amps though, not a problem so long as I can figure out the panelling for the chassis.
Is this the Symasym you used Pinkmouse?:
http://www.lf-pro.net/mbittner/Sym5_Webpage/symasym5_3.html
It fits the bill but I'd need a PCB. You don't happen to have any left over or know where to get them? I can drill them myself if I need. Did you use ON-Semi devices or Toshiba at the output?
we continue to disagree on this. I say that peak SPL is the criterion. It is certainly not the energy content of time averaged music. This energy content can have a bearing on your choice of smoothing capacitance and heatsink size for each of the amplifiers.Dr.EM said:
you just missed a group buy on AAK's PCB.Dr.EM said:
try asking Mike if he knows where some suitable PCBs are.
The 50W to 70W available from a 1pair Symasym would be ideal (economic build) for all six channels of a tri-amped system.
Sorry, but I have no spares. My original two channel version used home etched boards to the design on Mike's site, with Onsemi outputs. It was a very easy job with toner transfer process. If you like, I can print off a couple of glossy prints for you to do yourself.
The next project is 10 channels for an active system using the Toshiba devices, with AAK's GB boards. I'm getting the parts together as we speak.
The next project is 10 channels for an active system using the Toshiba devices, with AAK's GB boards. I'm getting the parts together as we speak.
There is still one or 2 sets of the first run from the DXAMP groupbuy left if you are interested, at $20 a set it may be worth looking in to.... I got some nice caps for my boards yesterday... the four caps cost $30... which is more than the boards... of course it can be built with scheaper/smaller stuff too.
I guarantee it is miles better than P3.
I guarantee it is miles better than P3.
With regards to energy content i'm only going by what Rod Elliot says in his articles, I haven't actually experimented with it much myself. The only real experiment I conducted was playing a 15W amplifier with 24db highpass at 3khz unclipped (viewed waveform on scope) into a speaker around the same 93db sensitivity. By itself it was horribly loud and I had to wear ear protection in the same room as it, it was heard clearly downstairs. With the full frequency range it'd probably be just be heard as "loud" but still louder than I tend to play music.
Shame I missed that group buy . I haven't made boards myself but I could experiment with it easily enough. Can get press-n-peel and a "student etching kit" cheaply enough. Thanks for the offer on glossy prints, I may take you up on that if I do decide to do it.
Your project sounds very interesting, though I trust your getting a great sound already from your 3 way with HM100Z0 mids? You planning a 5 way stereo or surround setup with all those amps?
Finishing these in an attractive way will still be the biggest problem, which is a part of why those commercial amps are an attractive option (and cost of course).
Shame I missed that group buy
. I haven't made boards myself but I could experiment with it easily enough. Can get press-n-peel and a "student etching kit" cheaply enough. Thanks for the offer on glossy prints, I may take you up on that if I do decide to do it. Your project sounds very interesting, though I trust your getting a great sound already from your 3 way with HM100Z0 mids? You planning a 5 way stereo or surround setup with all those amps?
Finishing these in an attractive way will still be the biggest problem, which is a part of why those commercial amps are an attractive option (and cost of course).
FWIW I just picked up an Alesis RA150. My original intent was simply to use it as a donor chassis for a fully-balanced UcD project, but I have it hooked up as the main amp in the system right now. I won't claim that it's perfect, but I've been very pleasantly surprised by it.
In my system (98dB/W efficiency) it seems to at least equal a Trends TA-10.1 in finesse/presentation and absolutely betters it in dynamics. Some of this may be due to the fact that I have a balanced source, and have to use balanced-> RCA cables with the Trends.
I haven't tried a single-ended source with the Alesis - if the SE=>BAL stage is marginal it may not perform as well.
The layout/construction is pretty good, although given the price point and target audience there are some obvious compromises. It is an all-discrete dual-differential topology (according to the manual - I haven't traced the circuit) with two pairs of output devices per side. I am now seriously thinking that mildly massaging the Alesis with PS cap upgrades, input stage improvements (it has the typical 'lytic coupling caps plus op-amp input buffer) etc might be a viable path to a very good amp.
As for my original idea, I think the RA150 is an ideal donor chassis for something like a UcD. Balanced detented volume pots, 42V rails and decent heat-sinking and (IMHO) a perfectly acceptable cosmetic package all for $200 - not bad in my books. (note - given the 75/45 power rating for the RA150, the 42V rails surprised me. It would work well for a UcD, though)
In my system (98dB/W efficiency) it seems to at least equal a Trends TA-10.1 in finesse/presentation and absolutely betters it in dynamics. Some of this may be due to the fact that I have a balanced source, and have to use balanced-> RCA cables with the Trends.
I haven't tried a single-ended source with the Alesis - if the SE=>BAL stage is marginal it may not perform as well.
The layout/construction is pretty good, although given the price point and target audience there are some obvious compromises. It is an all-discrete dual-differential topology (according to the manual - I haven't traced the circuit) with two pairs of output devices per side. I am now seriously thinking that mildly massaging the Alesis with PS cap upgrades, input stage improvements (it has the typical 'lytic coupling caps plus op-amp input buffer) etc might be a viable path to a very good amp.
As for my original idea, I think the RA150 is an ideal donor chassis for something like a UcD. Balanced detented volume pots, 42V rails and decent heat-sinking and (IMHO) a perfectly acceptable cosmetic package all for $200 - not bad in my books. (note - given the 75/45 power rating for the RA150, the 42V rails surprised me. It would work well for a UcD, though)
The groupbuy is not missed yet... I am still shipping...
And I intend to keep stock for the forseeable future...
I just checked online status, it would appear the PSU PCBs are also completed at the factory now.
Hit a little glitch when I wanted to matchthe last transistor pairs... The shop gave me 2 tubes of BD140... instead of BD139 + BD140... have to wait for them to be flown in now (expected tommorrow), as I bought out just about all stock in the Western Cape for this group buy.
95% of envelopes are ready and have their address lables on, just waiting for the 2 transistors need for each packet.
The delays had their benefits as I also found the bridges that goes on the jumper pins, and included those with the PCB's
Will try to take everything to the post office tommorrow in one batch.. trying to reduce unneccessary trips... as this whole thing does not even = non-profit buy... its more like Nordic's christmas gift to DIY audio.
And I intend to keep stock for the forseeable future...
I just checked online status, it would appear the PSU PCBs are also completed at the factory now.
Hit a little glitch when I wanted to matchthe last transistor pairs... The shop gave me 2 tubes of BD140... instead of BD139 + BD140... have to wait for them to be flown in now (expected tommorrow), as I bought out just about all stock in the Western Cape for this group buy.
95% of envelopes are ready and have their address lables on, just waiting for the 2 transistors need for each packet.
The delays had their benefits as I also found the bridges that goes on the jumper pins, and included those with the PCB's
Will try to take everything to the post office tommorrow in one batch.. trying to reduce unneccessary trips... as this whole thing does not even = non-profit buy... its more like Nordic's christmas gift to DIY audio.
Ok, I'm pretty certain I will go ahead with building my own amps. I'm just looking around for parts and adding things up now, should be under £500 for all 3 amps which isn't bad considering the quality (and compared to the commercial cost of such units).
With the Symasym amp which transistors will be important to match? Will the MPSA18 be most critical? Also is it HFE I should match. I ask because I can only get those from the US and wonder if it's worth getting 100 from there or just using BC550C which are readily available.
I see the DX amp on your web shop Nordic. It looks like really great value for a pair of boards with transistors and is also being considered as my mid amp.
My chassis woes may be solvable with these:
http://www.gumboot.com/cgi-bin/ezshop/shop/2052?prodcode=52
Noticed the ad on the index, part 432711 looks pretty ideal (might have prefered plain alu to match heatsinks but it should be fine for top and bottom panels). Anybody used this company, any opinions, quality of products?
With the Symasym amp which transistors will be important to match? Will the MPSA18 be most critical? Also is it HFE I should match. I ask because I can only get those from the US and wonder if it's worth getting 100 from there or just using BC550C which are readily available.
I see the DX amp on your web shop Nordic. It looks like really great value for a pair of boards with transistors and is also being considered as my mid amp.
My chassis woes may be solvable with these:
http://www.gumboot.com/cgi-bin/ezshop/shop/2052?prodcode=52
Noticed the ad on the index, part 432711 looks pretty ideal (might have prefered plain alu to match heatsinks but it should be fine for top and bottom panels). Anybody used this company, any opinions, quality of products?
Oops, must have glossed over that in the article . With hFE I suppose I want to aim to use the higher number units? I remember matching Vbe for transistors in a moog filter I built, so I can either use that circuit or the one you suggested, which is probably a lot more up to date.
Also, the P101 amp can use Exicon MOSFETs in place of the Hitachi/Renesas ones. I have these but they are in the T03 case, will mounting a short distance off board cause any trouble? I only have them because the deal was so good, 18 (9 pairs) for 86p . Discontinued item at Maplins, codes AY54 and AY56 if I remember correctly (were £6.50 each, now only 4p!) but I don't think many/any stores have them left
. With hFE I suppose I want to aim to use the higher number units? I remember matching Vbe for transistors in a moog filter I built, so I can either use that circuit or the one you suggested, which is probably a lot more up to date. Also, the P101 amp can use Exicon MOSFETs in place of the Hitachi/Renesas ones. I have these but they are in the T03 case, will mounting a short distance off board cause any trouble? I only have them because the deal was so good, 18 (9 pairs) for 86p
. Discontinued item at Maplins, codes AY54 and AY56 if I remember correctly (were £6.50 each, now only 4p!) but I don't think many/any stores have them left
Ok, just getting bits together now.
First will be the DoZ amp. I can use 2SC5200 as the output devices right? This is my plan anyhow, with 35V supplies.
Among other bits I have the heatsinks already. The class-A amp calls for fairly large units and there was a price break at 5 on the 0.4c/w ones, so I've ordered 6 with which I can do all the amps. Oversized for the symasym and P101 (at 80W, 42v) but hey they can never be too big right. Also gives the amps similar (identical) appearance with these sinks forming the chassis sides. Found a local supplier of aluminium for the top and bottom too, 4mm planned thickness.
Once the bits are in hand and with some more planning I'm hoping for a quick and straightforward construction. I don't say that with huge confidence however as that is not usually the case
First will be the DoZ amp. I can use 2SC5200 as the output devices right? This is my plan anyhow, with 35V supplies.
Among other bits I have the heatsinks already. The class-A amp calls for fairly large units and there was a price break at 5 on the 0.4c/w ones, so I've ordered 6 with which I can do all the amps. Oversized for the symasym and P101 (at 80W, 42v) but hey they can never be too big right. Also gives the amps similar (identical) appearance with these sinks forming the chassis sides. Found a local supplier of aluminium for the top and bottom too, 4mm planned thickness.
Once the bits are in hand and with some more planning I'm hoping for a quick and straightforward construction. I don't say that with huge confidence however as that is not usually the case
Attachments
Cheers, got them from Farnell:
http://uk.farnell.com/jsp/search/productdetail.jsp?sku=1211720
Reading the description you'd expect plain unfinished aluminium, I know I did. The anodized black is nicer though, shame the other panels won't be matching. Fronts will probably be plywood to save them looking too plain/industrial overall.
http://uk.farnell.com/jsp/search/productdetail.jsp?sku=1211720
Reading the description you'd expect plain unfinished aluminium, I know I did. The anodized black is nicer though, shame the other panels won't be matching. Fronts will probably be plywood to save them looking too plain/industrial overall.
Hmm, just built the DoZ amp and was testing it. I used the PCB and about 33V rails. I also used the 2SC5200 as output transistors but am worried because I just burnt my finger on the Zobel . It actually overall seemed to be working fine and I was setting the bias up but there was a nasty smell, I realised what it was when I touched that 10 ohm 2W resistor. Presumably there is some serious oscillation happening . Any help greatly appreciated 
. It actually overall seemed to be working fine and I was setting the bias up but there was a nasty smell, I realised what it was when I touched that 10 ohm 2W resistor. Presumably there is some serious oscillation happening . Any help greatly appreciated
I've checked this so many times and it all looks fine. Heres a few "stab in the dark" suggestions, I haven't earthed the heatsink yet, could that be a problem? Also the input is left "open", not grounded or connected to a signal source.
I've looked through and I'm not the first to have tried 2SC5200 in this design so I know it's possible. The devices are from Rapid electronics, surely there isn't much chance of them being fakes? Themselves they arn't getting excessively hot anyway but mabye fakes would cause an instability? To give more detail the driver transistors are the reccomended BD139 mounted on the suggested small heatsink. The bias control acts a bit erratically, I was aware of the delay in it's effect but it becomes very sensitive all of a sudden (ie. nothing for many turns then a sudden increase in one turn or so). Its not difficult to use because of that but is this normal? I used suggested BC559 as signal transistor. Any ideas anyone
I've looked through and I'm not the first to have tried 2SC5200 in this design so I know it's possible. The devices are from Rapid electronics, surely there isn't much chance of them being fakes? Themselves they arn't getting excessively hot anyway but mabye fakes would cause an instability? To give more detail the driver transistors are the reccomended BD139 mounted on the suggested small heatsink. The bias control acts a bit erratically, I was aware of the delay in it's effect but it becomes very sensitive all of a sudden (ie. nothing for many turns then a sudden increase in one turn or so). Its not difficult to use because of that but is this normal? I used suggested BC559 as signal transistor. Any ideas anyone
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.