Dear all,
I've purchased a dedicated web server for my website. Please direct link to www.diy-audio-guide.com where I will publish more interesting ideas and articles about DIY audio there.
Thank you all for your support!
www.diy-audio-guide.com
Ken
Saturday, March 19, 2011
Monday, November 17, 2008
Helping friend to clear some DIY stock
New DIY parts clearance from friend. All are sold even below market price sold in the USA. Imagine the savings on shipping, taxes, lost + damage during shipping and exchange rate losses!!!!
Guaranteed to work. Order quick, before they're gone! For a limited time only.
JJ 3ooB new-in-box - RM750/pair
* orig. price - U$220/pair
Tungsol 5687WA (military version, date code 5712), NOS - RM80/pc
* orig. price - U$34.95/pc
* exceptionally sweet and smooth, nice punchy bass
Auricap 600VDC, 0.22uf - RM60/pc
* orig. price - U$19.05/pc
Hammond 193L, 300mA DC, 5H - RM130
* orig. price - U$39/pc
Hammond 158Q, 150mA, 5H - RM70
* orig. price - U$22/pc
Hammond 1627SE - RM850/pair
* orig. price - U$135/pc
Rubicon YXF cap, 100V, 220uf - RM8 for 4pc
Panasonic FC cap, 10V, 220uf, 105 degC - RM5 for 10pc (2opc in total)
Panasonic FC cap, 16V, 180uf, 105 degC - RM2.5o for 5pc
Panasonic FC cap, 25V, 120uf, 105 degC - RM3.oo for 6pc
Panasonic FC cap, 100V, 220uf, 105 degC - RM5.oo per pc (21pc in total)
Panasonic TSUP cap, 16V, 15000uf - RM24 for 2pc
Panasonic TSHA cap, 400V, 100uf, 105 degC - RM13/pc (7pc in total)
Elna Starget cap, 35V, 100uf - RM2.50/pc (6pc in total)
Guaranteed to work. Order quick, before they're gone! For a limited time only.
JJ 3ooB new-in-box - RM750/pair
* orig. price - U$220/pair
Tungsol 5687WA (military version, date code 5712), NOS - RM80/pc
* orig. price - U$34.95/pc
* exceptionally sweet and smooth, nice punchy bass
Auricap 600VDC, 0.22uf - RM60/pc
* orig. price - U$19.05/pc
Hammond 193L, 300mA DC, 5H - RM130
* orig. price - U$39/pc
Hammond 158Q, 150mA, 5H - RM70
* orig. price - U$22/pc
Hammond 1627SE - RM850/pair
* orig. price - U$135/pc
Rubicon YXF cap, 100V, 220uf - RM8 for 4pc
Panasonic FC cap, 10V, 220uf, 105 degC - RM5 for 10pc (2opc in total)
Panasonic FC cap, 16V, 180uf, 105 degC - RM2.5o for 5pc
Panasonic FC cap, 25V, 120uf, 105 degC - RM3.oo for 6pc
Panasonic FC cap, 100V, 220uf, 105 degC - RM5.oo per pc (21pc in total)
Panasonic TSUP cap, 16V, 15000uf - RM24 for 2pc
Panasonic TSHA cap, 400V, 100uf, 105 degC - RM13/pc (7pc in total)
Elna Starget cap, 35V, 100uf - RM2.50/pc (6pc in total)
On-the-shelf price vs DIY price?
I've read a very interesting article from an online audio publication.
How does a DIY product compare to something bought retail? As you know, there are lots of overhead costs involved being a manufacturer. For those who'd been in an OEM line, they would know this very well. Also, the retailer is charging a whole lot on their side. 50% is not very abnormal, not to mention after going through the main distributor (if there is one, or more).
Let's do a litte maths. Say you are considering a U$10,000 amplifier. So, 1/2 of the price goes to the dealer. U$5,000 to the dealer, U$5,000 to the manufacturer. From there, the 1/3 is the actual cost of product. That leaves us with only U$1,666. From this number, almost 40% of the money goes to the miscellaneous components and only 60% of the $$$ are involved in the signal. So, what's left is only U$1,000.
I can only say: "WTF!!!". I went down the path of DIY-ing and never looked back ever since! If we do it the reverse way, if you pay U$2,000 for the component costs, you're actually getting yourself a U$20,000 gear! Who says DIY is not high end?
My manufacturer friend confirmed the same findings. His company is manufacturing cabinets for a few well known audio companies in the U.K.. The cost versus shelf price is at least 10x! So, your U$10,000 will only cost them U$1,000 or maybe less to manufacturer. So, the maths actually calculates.
Go figure.
Ken
How does a DIY product compare to something bought retail? As you know, there are lots of overhead costs involved being a manufacturer. For those who'd been in an OEM line, they would know this very well. Also, the retailer is charging a whole lot on their side. 50% is not very abnormal, not to mention after going through the main distributor (if there is one, or more).
Let's do a litte maths. Say you are considering a U$10,000 amplifier. So, 1/2 of the price goes to the dealer. U$5,000 to the dealer, U$5,000 to the manufacturer. From there, the 1/3 is the actual cost of product. That leaves us with only U$1,666. From this number, almost 40% of the money goes to the miscellaneous components and only 60% of the $$$ are involved in the signal. So, what's left is only U$1,000.
I can only say: "WTF!!!". I went down the path of DIY-ing and never looked back ever since! If we do it the reverse way, if you pay U$2,000 for the component costs, you're actually getting yourself a U$20,000 gear! Who says DIY is not high end?
My manufacturer friend confirmed the same findings. His company is manufacturing cabinets for a few well known audio companies in the U.K.. The cost versus shelf price is at least 10x! So, your U$10,000 will only cost them U$1,000 or maybe less to manufacturer. So, the maths actually calculates.
Go figure.
Ken
ULTIMA Preamp - Part 4
Rain is pouring outside. I'm stuck in office with nothing to do. I might as well continue my preamp project.
I can tell you that the ULTIMA project is taking a very sharp turn! I'm going towards BATTERY POWERED tube preamp. How's that? I've just read the reviews of a battery tube preamp in 6suns.com (you know where) and I just could not believe how good that preamp can be, with a U$4000 price tag!
It is SLA powered, uses JJ tubes, Jensen foil capacitors, Panasonic electrolytic, Alps motorized pot and a ver nice chassis. The Jensen caps are placed directly beside the JJ tubes, what a joke! I guess the manufacturer must either being stupid or they just want to suck blood out of their supports for replacing the foil caps in near future.
I can do better than that, for 1/2 the price! I've talked to my hardware (chassis) manufacturer friend and he should be able to supply me a nice chassis, with cryo-option for my tube preamp project. Now that's a good start!
So this is how the new tube preamp project is going to be, with 1/2 the price but better than the U$4000 one for sure!
Needless to say more.
SLA (2pc - 3pc) battery powered B+/HT and mains regulated filament heating
I'm choosing at least 2 SLA, 3 with even better results and drive strength for this preamp. User can select whether to power the tube filament with battery too but optional. The reasons for not using battery for filament are because:
1. The filament draws too much current.
2. Battery heating brings not much effect on indirectly heated triode. Mains
regulated or battery heating sounds almost the same.
3. Prolongs the lifespan of your SLA.
4. Extended play time - a 5AH SLA will last > 200 hours per charge.
Exterior charger circuit
The charger is of little consequences. You choose! I would prefer it to be external so that one can buy any charger off-shelf easily. This will remove additional noise from the inner of the chassis too.
Preamp with OPT from Sowter or capacitor coupled
A good OPT is alway better than capacitor coupled. It eliminates ground loop between your preamp and power amplifier as well. It has lower output impedance to be able to drive any interconnects. Also, I can wire it to be balanced if your power amplifier accepts balanced input. The unit I chosed has a very wide bandwidth of > 10Hz - 60kHz.
Aerospace grade aluminum chassis
Rigid. Low mechanical resonances. Sturdy. Reliable. Non-magnetic. Light.
Stepped attenuator
Others possible but stepped attenuator gives the best sonic results over brush-based pots. Remote controlled attenuator with relay switching is available too but with a premium price as they're not cheap themselves.
That's it! Anyone interested, drop me an email.
Ken
I can tell you that the ULTIMA project is taking a very sharp turn! I'm going towards BATTERY POWERED tube preamp. How's that? I've just read the reviews of a battery tube preamp in 6suns.com (you know where) and I just could not believe how good that preamp can be, with a U$4000 price tag!
It is SLA powered, uses JJ tubes, Jensen foil capacitors, Panasonic electrolytic, Alps motorized pot and a ver nice chassis. The Jensen caps are placed directly beside the JJ tubes, what a joke! I guess the manufacturer must either being stupid or they just want to suck blood out of their supports for replacing the foil caps in near future.
I can do better than that, for 1/2 the price! I've talked to my hardware (chassis) manufacturer friend and he should be able to supply me a nice chassis, with cryo-option for my tube preamp project. Now that's a good start!
So this is how the new tube preamp project is going to be, with 1/2 the price but better than the U$4000 one for sure!
- NOS Tungsol 5687WA tube
- SLA (2pc - 3pc) battery powered B+/HT
- Mains powered tube filament/heating
- Integrated or exterior charger circuit
- Preamp with OPT from Sowter or capacitor coupled
- Aerospace grade aluminum chassis
- Stepped attenuator or motorized (remote controlled) Alps pot or electronic attenuator with relays + resistors
Needless to say more.
SLA (2pc - 3pc) battery powered B+/HT and mains regulated filament heating
I'm choosing at least 2 SLA, 3 with even better results and drive strength for this preamp. User can select whether to power the tube filament with battery too but optional. The reasons for not using battery for filament are because:
1. The filament draws too much current.
2. Battery heating brings not much effect on indirectly heated triode. Mains
regulated or battery heating sounds almost the same.
3. Prolongs the lifespan of your SLA.
4. Extended play time - a 5AH SLA will last > 200 hours per charge.
Exterior charger circuit
The charger is of little consequences. You choose! I would prefer it to be external so that one can buy any charger off-shelf easily. This will remove additional noise from the inner of the chassis too.
Preamp with OPT from Sowter or capacitor coupled
A good OPT is alway better than capacitor coupled. It eliminates ground loop between your preamp and power amplifier as well. It has lower output impedance to be able to drive any interconnects. Also, I can wire it to be balanced if your power amplifier accepts balanced input. The unit I chosed has a very wide bandwidth of > 10Hz - 60kHz.
Aerospace grade aluminum chassis
Rigid. Low mechanical resonances. Sturdy. Reliable. Non-magnetic. Light.
Stepped attenuator
Others possible but stepped attenuator gives the best sonic results over brush-based pots. Remote controlled attenuator with relay switching is available too but with a premium price as they're not cheap themselves.
That's it! Anyone interested, drop me an email.
Ken
Friday, October 31, 2008
ULTIMA Preamp - Part ３
A friend read my ULTIMA blog and emailed me, are you able to upgrade the preamp you built for me and also what do you think of preamp output transformers (OPT)?
That's interesting. I did not want to include a preamp OPT as it would be costly. Since this is a cost-no-object design, I say why not? Preamp OPT itself is a much more difficult design than a normal power OPT or capacitor. It's easier to design a good capacitor or power OPT but preamp OPT is another different story at all.
A very, very good and wide bandwidth will be needed to do this job. If that is used, benefits are plenty. Here are a few recommended preamp OPTs you can choose:
Sowter type 9650
http://www.sowter.co.uk/specs/9560.htm
Better than 10 Hz - 60 kHz, 7k5:600
FEATURES A very large core to ensure a high bandwidth and low output impedance. High grade grain oriented silicon iron (M6) core for minimal losses and harmonic distortion. Multi section winding ensures minimal leakage inductance for an exceptional bandwidth. Low copper resistance which permits an output impedance of 203 ohms when driven from a valve with rp = 3000 ohms and a secondary load of 600 ohms.
Hashimoto HL-10k-6
http://www.eifl.co.jp/index/export/hashimoto/hashimoto.html
20 Hz - 45 kHz, pri:sec = 10k : 600/150
Tamura A-4714
http://www.eifl.co.jp/index/export/export2.htm
5k/60mA: 600/6 Ohm
Some other excellent OPTs vendors are not mentioned, like Lundahl, Magnequest, Tango, AudioNote UK, and etc. You can select from them as well. You decide!
Sunday, October 26, 2008
ULTIMA Preamp - Part 2
Alright boys, here comes the power supply!
It is a massive 4 choke supply. I believe many of the commercial tube preamps or even power amps don't even come close to this! Pseudo mono power supply is used to split the left and right channel B+ with chokes to reduce inter-channel crosstalk. Do remember this is a preamp and ripple on the B+ is highly undesirable, therefore the power supply has to be carefully designed. All the chokes have to be at least 20H 150mA for the 1st two, and 20H 80ma for the other 2. Since there is no output transformer used, the size and weight should still be bearable.
The filament supply for 5687 is AC and for 45, it is a special voltage-current regulated 2 stage DC supply. 5687 filament's supply is AC as it is indirectly heated. DC supply for 5687 is just an overkill. Some claim using DC as it reduces hum on indirectly heated tube, that's just plain BS as they do not know how to design a good power supply. As for directly heated tubes, this is another matter. For power amplifiers, AC heating on tubes like 300B (5Vac), 2A3 or 45 (2.5Vac) is still okay if users have medium efficiency speakers. For those with 94dB/w/m or above speakers, DC heating is a must to kill the AC hum. For preamps, DC is a must on directly heated triodes with whatever speakers.
Alright, that's all for PSU. Next, we're going to talk about parts... this is the interesting part... :)
It is a massive 4 choke supply. I believe many of the commercial tube preamps or even power amps don't even come close to this! Pseudo mono power supply is used to split the left and right channel B+ with chokes to reduce inter-channel crosstalk. Do remember this is a preamp and ripple on the B+ is highly undesirable, therefore the power supply has to be carefully designed. All the chokes have to be at least 20H 150mA for the 1st two, and 20H 80ma for the other 2. Since there is no output transformer used, the size and weight should still be bearable.
The filament supply for 5687 is AC and for 45, it is a special voltage-current regulated 2 stage DC supply. 5687 filament's supply is AC as it is indirectly heated. DC supply for 5687 is just an overkill. Some claim using DC as it reduces hum on indirectly heated tube, that's just plain BS as they do not know how to design a good power supply. As for directly heated tubes, this is another matter. For power amplifiers, AC heating on tubes like 300B (5Vac), 2A3 or 45 (2.5Vac) is still okay if users have medium efficiency speakers. For those with 94dB/w/m or above speakers, DC heating is a must to kill the AC hum. For preamps, DC is a must on directly heated triodes with whatever speakers.
Alright, that's all for PSU. Next, we're going to talk about parts... this is the interesting part... :)
ULTIMA Preamp - Part 1
ULTIMA design concept is determined. You do not need very complex circuitry and lots of components to make an amp sound good. Sometimes I just got so fed up looking at amps with such a complex design that are just simply an overkill, sound killer, noise generator, "body builder" and excuse for price increase, you know what I mean?!?! Simple is the key! Come on guys, not all audiophiles are stupid!
There shall be 2 flavors, 1 is for those with tube power amps, the other is for those with solid-state amps that desires more tube warmth and smoothness but also requires higher driving strength for the SS amp input stage.
Flavor 1 - two 5687 + two 45 + one 5U4G, 3 outputs preamp
Flavor 2 - two 5687 + 0ne 6X5, 2 outputs preamp
The 1st stage is a mu-stage, having a 5687 (other triodes are possible as well, like 6C45, 5842/WE417, 3A167M, etc) triode loaded with another triode that is configured as constant current source. The upper triode will act as a high impedance load for the lower triode. Such topology is much superior to simple resistor loaded or choke loaded types, with the cons of needing higher B+ and an additional tube in the circuit.
The 2nd stage is just a choke loaded gain stage with a direct-heated-triode like 45 or 2A3. Other choices could be 71A, 6B4G, 31, 01A, 801A and even the 300B! 45 is chosen as it fits nicey in to this application due to the sound character that is considered rather transparent and linear among the direct-heated-triodes. The current requirement to bias the 45 is just nice as well and not too taxing on the power design as very low ripple is needed to be used as preamps.
Some may ask why not just mu-stage the 45 as well? Of course you can! I'm not doing it as mu-staging the 45 will increase the power supply B+ voltage to over 500Vdc and there aren't many 500Vdc+ capacitors that sound that good compared to their normal 450Vdc/500Vdc brothers.
If you build the complete version, you shall get 3 outputs, 1 low and 2 medium output impedance outputs. The low output impedance channel can be used to drive subs as that is the worst sounding among the 3. The 2 medium output channels can be used to drive tube or SS power amps with input impedance of 25k or more, the higher the merrier. If your power amps just happen to have very low input impedance, then use the low output impedance channel, it is not as good as the other 2 but it is still a very good preamp output.
The frequency response of the preamp will be 5Hz - 100kHz @ 3dB. It also depends on the wiring methods and components used. One can copy the design but the results will differ depending on the skills of the builder. None is created equal. Don't blame if your version does not sound good. :)
The next article we shall talk more about ULTIMA in details.
There shall be 2 flavors, 1 is for those with tube power amps, the other is for those with solid-state amps that desires more tube warmth and smoothness but also requires higher driving strength for the SS amp input stage.
Flavor 1 - two 5687 + two 45 + one 5U4G, 3 outputs preamp
Flavor 2 - two 5687 + 0ne 6X5, 2 outputs preamp
The 1st stage is a mu-stage, having a 5687 (other triodes are possible as well, like 6C45, 5842/WE417, 3A167M, etc) triode loaded with another triode that is configured as constant current source. The upper triode will act as a high impedance load for the lower triode. Such topology is much superior to simple resistor loaded or choke loaded types, with the cons of needing higher B+ and an additional tube in the circuit.
The 2nd stage is just a choke loaded gain stage with a direct-heated-triode like 45 or 2A3. Other choices could be 71A, 6B4G, 31, 01A, 801A and even the 300B! 45 is chosen as it fits nicey in to this application due to the sound character that is considered rather transparent and linear among the direct-heated-triodes. The current requirement to bias the 45 is just nice as well and not too taxing on the power design as very low ripple is needed to be used as preamps.
Some may ask why not just mu-stage the 45 as well? Of course you can! I'm not doing it as mu-staging the 45 will increase the power supply B+ voltage to over 500Vdc and there aren't many 500Vdc+ capacitors that sound that good compared to their normal 450Vdc/500Vdc brothers.
If you build the complete version, you shall get 3 outputs, 1 low and 2 medium output impedance outputs. The low output impedance channel can be used to drive subs as that is the worst sounding among the 3. The 2 medium output channels can be used to drive tube or SS power amps with input impedance of 25k or more, the higher the merrier. If your power amps just happen to have very low input impedance, then use the low output impedance channel, it is not as good as the other 2 but it is still a very good preamp output.
The frequency response of the preamp will be 5Hz - 100kHz @ 3dB. It also depends on the wiring methods and components used. One can copy the design but the results will differ depending on the skills of the builder. None is created equal. Don't blame if your version does not sound good. :)
The next article we shall talk more about ULTIMA in details.
Subscribe to:
Posts (Atom)