| |||||||||||||||||||||||||||
| Moderators: cmoy | Welcome. Please log in. | ||||||||||||||||||||||||||
Registration is required to post a new topic or a reply. 
| |||||||||||||||||||||||||||
cmoy  Administrator Joined: Apr. 13, 1999 |  Message [#40] posted on: 10-26-2005 03:35 PM CST (US).   |  |  Re: high current opamps, how about the OPA551/552 or LM7171? |
ericj HeadWize Fanatic Joined: Aug. 1, 2005 | Message [#41] posted on: 10-26-2005 05:07 PM CST (US). | | 552 is only stable for gains of 5 and above, so 551 is one possible option. Of course, that's a Really High Current amp, 250ma was it? jrc4556 outputs, what, 70ma? I have to admit that the 4556 has character and it's not a bad opamp, but in my +/-15v cmoy the 2107 has so much more going on. It's the little things, like tracks recorded with a subtle echo on the vocals, as though they were recorded in a room that wasn't acoustically dampened - I can't clearly hear the echo with the 4556 but i can with the 2107. Somebody will probably use a 4556 at some point because some people like it and i can't say i blame them. It'd be prudent to determine if it's at all appropriate in this circuit. How much current do people honestly need even for grados? I never understood the appeal of these low-z headphones so it's a foreign concept for me. |
| cmoy Administrator Joined: Apr. 13, 1999 | Message [#42] posted on: 10-26-2005 05:32 PM CST (US). | | ericj, thanks for the reminder about 552 stability. Since the opamp is configured for unity gain, I suppose a standard high current buffer like the BUF634 could work too.
I guess the trend towards low impedance headphones follows the battery trend of portables - the lower the better. You can get sound out of a Grado plugged into a portable run by one AA battery, but for the low end to bloom, fans of Grados and other low Z headphones like plenty of current reserve in their amps. One other nice aspect of high current opamps: driving (the now discontinued AKG K1000). The K1000 requires a minimum 5W amp. A 250mA current buffer could deliver that much power if heatsinked. |
| ericj HeadWize Fanatic Joined: Aug. 1, 2005 | Message [#43] posted on: 10-26-2005 05:43 PM CST (US). | |
At any rate, buf634p and buf634u appear to be out of stock the world around, and people are resorting to adapting the opa551 for substitute for it. Ti says they are out of stock in all formfactors and expect a lead time of 12+ weeks. Somebody failed to anticipate demand, or something. |
| cmoy Administrator Joined: Apr. 13, 1999 | Message [#44] posted on: 10-26-2005 05:49 PM CST (US). | |
What is the reason for this drain on the BUF634? |
runeight Headphone Council Joined: Mar. 8, 2002 | Message [#45] posted on: 10-26-2005 06:38 PM CST (US). | | If it were available, a BUF634 could easily be put into the unity gain feedback loop of the OPA2134 to give super high output. OPA551 appears to be a FET input opamp. If so, it would drop right into the Stoopid. Except it only comes in single packages so you'd have to account for this with the build. One change that might be necessary if anyone really tries to pull 200mA would be to increase the size of the input caps in the bipolar PS to more like 2200u. Even bigger possible to get full bass. As I think I posted before, FET input opamps are preferred because the BJT inputs may tend to load down the tube and defeat the effect of the CCS. [Edited by runeight on 10-26-2005 at 07:10 PM.] |
| ericj HeadWize Fanatic Joined: Aug. 1, 2005 | Message [#46] posted on: 10-26-2005 09:37 PM CST (US). | |
|
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#47] posted on: 10-26-2005 10:49 PM CST (US). | | Here's another version with about 4db of bass boost. No one has tried this and I suspect that neither mains nor batman thinks its necessary. So you might want to ask one of them about the need for this before you try it. OTOH, you can always switch the bass boost out. The main drawback to this is that it makes the amp more complicated and one of the goals was to keep it as simple as possible. [Edit: See later posts for a better bass boost circuit] [Edited by runeight on 10-30-2005 at 05:09 PM.] |
Batman HeadWizer Joined: May 12, 2003 | Message [#48] posted on: 10-27-2005 09:18 AM CST (US). | | I'd like to caution all of you who are considering building this Stoopid amp that the filament dropping resistors get VERY hot (~110 C!!)  . As you plan enclosures, please keep this in mind and be sure to provide for adequate ventilation. [Edited by Batman on 10-28-2005 at 09:16 AM.] |
mains_hum HeadWize Fanatic Joined: Aug. 19, 2003 | Message [#49] posted on: 10-27-2005 01:30 PM CST (US). | | Ahh .. I see the 68Ω 3W are now 100Ω||220Ω 2W Yes, this helps as the absolute temperature for these parts drops to around 105°C from the 115°C I measured using the original resistor pair. Hence the four resistors shown in the prototype. Even so, just as Batman says, provide ventilation and keep these droppers away from the other components if you can to avoid heating the silicon. [Edited by mains_hum on 10-27-2005 at 01:37 PM.] |
| eric.w HeadWize Fanatic Joined: N/A | Message [#50] posted on: 10-27-2005 10:54 PM CST (US). | | Wow, this is a very neat design! I might try to build this.. do you think it might be feasible to build the SOHA with a 6111wa tube? Aside from how the filament is connected (I have a 6.3v transformer I could use), It seems to have fairly similar specifications, though the 6111's transconductance is higher (5 instead of 2.2 for the 12au7.) I might experiment with building it, because I have everything I would need except the correct tube. Anyway, looks like an exciting design  |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#51] posted on: 10-27-2005 11:30 PM CST (US). | | You should be able to use the 6111 with a separate heater supply. Once you take the heater off the B+ trafo secondary, it's voltage will go way up. To keep the trafo regulation within reason for this design you should use a 30VCT/20mA trafo or something close to this. Also the 6111 has a slightly higher mu which will increase the gain of the amp. If you could find a 12AU7 for the first build, it might be a little less tweaking. OTOH, tweaking is a big part of DIY. [Edited by runeight on 10-27-2005 at 11:31 PM.] |
| mains_hum HeadWize Fanatic Joined: Aug. 19, 2003 | Message [#52] posted on: 10-28-2005 11:11 AM CST (US). | | Yes, a 6111 is an interesting valve to try. Clearly, we are working right at the bottom of published curves here. From what I can tell/surmise/extrapolate it ought to sound similar to the 12AU7 we originally chose in that the likely compression you'll get (on paper that is) isn't too dissimilar. This just means it should work fine but we don't really know until you try. And as you realised, a separate 6.3Vac supply is needed. This will up B+ by a good 20V or so. |
| pho_boi Member Joined: Dec. 20, 2002 | Message [#53] posted on: 10-30-2005 06:25 AM CST (US). | | I was wondering if its ok to use a 2K trimpot instead of the 2K5? Because I'm having trouble finding a 2K5 one in Australia. Thank you. |
| mains_hum HeadWize Fanatic Joined: Aug. 19, 2003 | Message [#54] posted on: 10-30-2005 08:07 AM CST (US). | | Yes, 2kΩ is fine. I agree 2k5Ω is not so easy to get. I just searched my local suppliers and I don't see any as small trimpots which is what is intended here. 2kΩ no problemo. You might want to consider a multi-turn cermet (not cheap) for best long term stability, but a single turn trimmer will work fine .. just drift a little quicker over time. As long as you can measure 37V at the anode as a minimum and can keep it down to 45V as a max value (40V being nominal for this design) then all will be well. |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#55] posted on: 10-30-2005 08:48 AM CST (US). | | I've changed the schematics to use a 2k trimpot. 2k5 seems to be a problem everywhere. pho_boi, may I ask which version you have chosen to build? |
| pho_boi Member Joined: Dec. 20, 2002 | Message [#56] posted on: 10-30-2005 09:07 AM CST (US). | | I've chosen to build the Bass-boost version, I decided on this one because I remember the first hybrid I built using sijosae schem was very lacking in bass. And with this design I have the ability to turn bass-boost off or on depending if I like it or not. Thanks for the posting up the designs, its really great that we have this kind of community. Thanks again.
|
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#57] posted on: 10-30-2005 09:44 AM CST (US). | | OK. The break point for the boost circuit is about 300Hz. If you want to move it up in frequency use smaller capacitors where the 22n cap is. For example, 6n will put the break at 1kHz. 10n will put it at 600Hz. |
| mains_hum HeadWize Fanatic Joined: Aug. 19, 2003 | Message [#58] posted on: 10-30-2005 10:20 AM CST (US). | | pho_boi, The "bass boost" version is an interesting idea. It was originally put together after the original prototype was fired up for the first time and, to be honest, bass was a bit disappointing. However, the problem was tracked back to the voltage doubler capacitors not being big enough. This meant the CCS failed for low frequencies. Once , after some experimentation, we had changed the circuit to be as posted, bass is solid, clean, and tight. My personal view is you don't need the bass boost, but don't let me sway your decision |
| Batman HeadWizer Joined: May 12, 2003 | Message [#59] posted on: 10-30-2005 10:23 AM CST (US). | | pho_Boi I agree with mains_hum that the current version of the SOHA does not need the bass boost. While it won't hurt to build it in, leaving it out will keep the amp closer to stoopidly simple.  |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#60] posted on: 10-30-2005 03:11 PM CST (US). | | Mains and Batman are no doubt correct about the bass boost. They both did a fair amount of component adjusting to fine tune the response of the amp. However, if bass boost is still dear to your heart, here is probably a better way to do it. Typical opamp feedback loop compensation. This scheme doesn't load the tube as heavily. [Edited by runeight on 10-30-2005 at 06:13 PM.] |
| Page 1 2 3 4 5 6 7 8 9 10 11 12 13 14 > >> | Next Page | Prev Topic | Next Topic | ||||
 (remove _nospam_ ) |
© Chu Moy, 2001.