| |||||||||||||||||||||||||||
| Moderators: cmoy | Welcome. Please log in. | ||||||||||||||||||||||||||
Registration is required to post a new topic or a reply. 
| |||||||||||||||||||||||||||
cmoy  Administrator Joined: Apr. 13, 1999 |  Message [#20] posted on: 10-25-2005 11:45 PM CST (US).   |  |  How do the protection diodes work in these circuits? What happens when the input to an opamp exceeds 12V? |
runeight Headphone Council Joined: Mar. 8, 2002 | Message [#21] posted on: 10-26-2005 12:15 AM CST (US). | | It's not a matter of not exceeding 12V, it's a matter of exceeding the rail voltages. The data sheet for the OPA2134 says that the maximim input voltage is: Input Voltage ...... (V–)–0.7V to (V+)+0.7V This means that the input swing can't exceed the supply voltage by more than one diode drop. The diodes ensure that this won't happen. In Batmans's amp the diodes will clip the input at +/- 12.6V and in Main's case it will be +/- 11.2V. If the input voltage exceeds the max, the opamp can be fried, but mostly it will stop acting like an opamp. Getting the input to over 12V will be hard to do because the amp will be very loud. But it is possible given the gain of the first stage. Opamps are so cheap, anyone should feel free to leave the diodes out. |
| cmoy Administrator Joined: Apr. 13, 1999 | Message [#22] posted on: 10-26-2005 12:34 AM CST (US). | | Ok, but let's say I feed it an input of 14V (+14/-14) via a function generator. The excess voltage must go somewhere? Is it dissipated as heat in the wire? |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#23] posted on: 10-26-2005 01:19 AM CST (US). | | Well, what you mean is, what happens when the diodes conduct? Where does the current go? It has to go into the bipolar power supply. What happens then depends on the ability of the source to source current and the ability of the PS to sink it. In this case, the tube will start sourcing hundreds of microamps (a guess) which will find their way to the output caps of the bipolar supply which are in turn supplying current to the opamp V+ and V-. The output of the regs will fluctuate some, but at this point the amp would not be operating properly anyway. [Edited by runeight on 10-26-2005 at 01:20 AM.] |
| cmoy Administrator Joined: Apr. 13, 1999 | Message [#24] posted on: 10-26-2005 01:36 AM CST (US). | | Thanks runeight. This scenario really interests me. What happens if I use a function generator outputting a 25V (+25/-25) signal with over 1A of current capability and use this to drive the opamp. Will I end up destroying the PS? Also, the lower diode conducts when the input goes more negative than the power supply. Is the excess current then flowing into the source: the function generator? Will the generator be destroyed? |
| jogor Member Joined: Oct. 26, 2005 | Message [#25] posted on: 10-26-2005 04:13 AM CST (US). | | just a suggestion i hope you could consider. to have the opamps on a separate (own) power supply. if the improvement is well worth it, i think people building this amp will not mind the additional cost of power transformer. thank you runeight, mains hum and batman for sharing this design, building a tube based amp is now a possibility for me being affordable and safe. like cmoy, your names will become legend in the headphone scene. |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#26] posted on: 10-26-2005 09:03 AM CST (US). | | Yes, there are some limitations on what this protection scheme can do. You do have to think through some of these issues. On the positive cycle, the current flows from the source (the tube/CCS combination) into to v+ where it promptly finds its way into the v+ of the opamp relieving the PS of having to supply a few hundred uA of current. The PS sees a slightly lighter load and the regulator does its thing. On the negative cycle, the tube/CCS sink a few hundred uA and the negative regulator sees a slightly heavier load and does its thing. If the source can source/sink huge amounts of current, then what happens will depend on what the PS can do given what it's normal load is also doing. Of course, this technique is really assuming interstage audio signal voltages and currents (at solid state levels). If your signal generator can supply 1A, you're going to need some rectifier diodes. And then, what's an opamp doing looking at a 1A signal generator in an actual circuit? jogor - good suggestion. we thought about this for a while. In this case, the design goal was to use one trafo to keep the amp as compact and simple as possible. Given that the tube is only drawing 2mA, there is no real downside to using the voltage multiplier. If you decide to build one, try using the voltage multiplier and single transformer. You may find that you don't really need the extra trafo. [Edited by runeight on 10-26-2005 at 09:06 AM.] |
| Steve Culton Headphone Council Joined: N/A | Message [#27] posted on: 10-26-2005 09:25 AM CST (US). | | This is in response to Batman's post a few posts back, but I wanted to thank you publicly for taking the time to do your wonderful wiring diagrams. Parts lists are also terrific. And of course, thanks to all the designers for sharing. But darn you all, because now I have to build this, and it's going to end up pushing some other projects to the side.
|
Batman HeadWizer Joined: May 12, 2003 | Message [#28] posted on: 10-26-2005 09:50 AM CST (US). | | Yes, you do have to build one and then you have to compare it to the Cavalli-Lovell MKII that you built and tell everyone about how they compare. Muahahahahahahahahahah  Oh, and by the way, one of the things we were aiming for is the use of NON-BOUTIQUE parts, so don't buy anything expensive We've got a little more tweaking to do before we finalize the parts list and I can make a drawing. |
| Steve Culton HeadWizer Joined: N/A | Message [#29] posted on: 10-26-2005 10:53 AM CST (US). | | We like cheap, non-boutique parts. THAT should also warm someone who will remain nameless' heart.  Hey, we could start a contest to see who can build the cheapest one! I love it's portability. I'm thinking for sturdyness that a Hammond box (black?) would do nicely. |
| evo HeadWizer Joined: Jul. 18, 2004 | Message [#30] posted on: 10-26-2005 11:04 AM CST (US). | | Oh damn looks like i'm gonna build another head-amp  |
ericj HeadWize Fanatic Joined: Aug. 1, 2005 | Message [#31] posted on: 10-26-2005 11:40 AM CST (US). | |
I've bought capacitors by the pound before. Of course, some of them are so old that they predate the shift to different powers of 10. So i have to look at my 50uf electrolytics with a bit of skepticism . . . . or maybe i should build a variable power supply so i can 're-form' them. |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#32] posted on: 10-26-2005 12:50 PM CST (US). | | BTW, here is what will probably be the next iteration of the design. Using JFETs in the CCSs. Everything else the same. |
| Steve Culton Headphone Council Joined: N/A | Message [#33] posted on: 10-26-2005 01:36 PM CST (US). | | You know I love CCSes. Not that I know what the hell is going on, but this is the first time (in my admittedly limited experience) that I've seen a resistor between the CCS and the tube. Is there a CCSes for Dummies explanation? |
| Steve Culton Headphone Council Joined: N/A | Message [#34] posted on: 10-26-2005 01:37 PM CST (US). | | Nevermind. It's part of the CCS, isn't it?
|
| Steve Culton Headphone Council Joined: N/A | Message [#35] posted on: 10-26-2005 01:45 PM CST (US). | | Sorry for the stream of consciousness posts...so in the SOHA, the CCS is a plate load? (And the CJ MkII CCS is a cathode follower?) |
| runeight Headphone Council Joined: Mar. 8, 2002 | Message [#36] posted on: 10-26-2005 02:07 PM CST (US). | | Almost. In the SOHA the CCS is a plate load. In the MK II the CCS is a cathode load for the diff amp. |
| cmoy Administrator Joined: Apr. 13, 1999 | Message [#37] posted on: 10-26-2005 02:34 PM CST (US). | | Thanks runeight! Please do try some high current opamps in these circuits too. |
| ericj HeadWize Fanatic Joined: Aug. 1, 2005 | Message [#38] posted on: 10-26-2005 02:40 PM CST (US). | |
I have to admit i've started slowly acquiring parts to build this amp. Bought some Sylvania 12au7's on ebay last night. Need to check my parts bin for a 15-0-15 transformer. |
mains_hum HeadWize Fanatic Joined: Aug. 19, 2003 | Message [#39] posted on: 10-26-2005 02:51 PM CST (US). | | Employing a CCS as the anode (plate) load means we can allow the B+ volts to wander about as the audio demand changes and not need to worry too much about regulation ... within reason that is. It keeps the B+ PS section simple. OK, high current opamps eh?. |
| 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? |
| 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.