| |||||||||||||||||||||||||||
| Moderators: cmoy | Welcome. Please log in. | ||||||||||||||||||||||||||
Registration is required to post a new topic or a reply. 
| |||||||||||||||||||||||||||
DIY Workshop » β22: A discrete, cascoded, fully complementary, pure class A amplifier (part 3) | johnwmclean Member Joined: Dec. 8, 2008 |  Message [#300] posted on: 01-29-2009 03:28 PM CST (US).   |  |  Thanks AMB, The amp will be purely balanced, so the wiring to the headphone jacks will be removed. The input ground tied to the chassis is not indicated on the AMB site diagram or literature, or did I miss this? |
amb  Headphone Council Joined: Apr. 1, 2004 | Message [#301] posted on: 01-29-2009 03:37 PM CST (US). | |
In the β22 website "Wiring & ground" section, look at diagram "Option 1" showing the input jacks and volume pot body connected to the chassis. That is for a 3-channel active ground configuration, but it's no different from 4-channel balanced. |
| mrmajestic2 Member Joined: Jun. 5, 2007 | Message [#302] posted on: 01-29-2009 03:39 PM CST (US). | |
What would happen if its not connected to the chassis? |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#303] posted on: 01-29-2009 03:43 PM CST (US). | |
You lose the shielding properties that the chassis provides. |
| johnwmclean Member Joined: Dec. 8, 2008 | Message [#304] posted on: 01-29-2009 05:33 PM CST (US). | | Totally makes sense now! Thanks AMB. |
| mrmajestic2 Member Joined: Jun. 5, 2007 | Message [#305] posted on: 01-30-2009 01:03 PM CST (US). | | Did anyone ever build a 5 channel B22? that would give either 4 channel balanced or on stereo amp with active ground and one stereo channel with passive ground. Any drawbacks in doing it this way? Just a crazy idea maybe. |
| johnwmclean Member Joined: Dec. 8, 2008 | Message [#306] posted on: 01-31-2009 02:58 AM CST (US). | | I’m a little confused by zobel network. Is there a need for this if the build is purely a headphone amp? As the amb site states it can be omitted if this is the case. Yet reading various posts in these threads opinions vary and the area at least for me remains sketchy. Also if a zobel network for a headphone only amp is to used, where would this be located, obviously it wouldn’t be the output binding posts. I’m guessing the headphone jacks? My build will be a balanced only with the psu’s and trannies in a separate case. Headphones will be the 650’s so will a zobel be needed, or is it case of wait till it’s completed first, then troubleshoot for a zobel network if necessary. [Edited by johnwmclean on 01-31-2009 at 03:00 AM.] |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#307] posted on: 01-31-2009 05:38 AM CST (US). | | Speakers typically are much more inductive than headphones and the amp must have zobels at the output to maintain stability. Some headphones are also inductive enough to cause problems, but many don't. Also depending on the wiring, a particular model of headphone may cause instability on one β22 but not on another. I suppose I could recommend zobels on all builds, but in many cases they're not needed. My own 3-ch active ground β22 has zobels, and the amp we built for krmathis also has them. It's your choice to try without them first and see if there is a problem, or just install them anyway. There is no audible downside to them, just added parts... The best place to mount the zobel is on the speaker binding posts (if you have them), or on the headphone jack. |
| ZMN Member Joined: Nov. 13, 2004 | Message [#308] posted on: 02-01-2009 09:51 AM CST (US). | |
Hi Amb, Here is what I get so far. Step 1: No source connected, signal input conneted to ground (following your advice here) -> You are right. no noise, dead quiet. Step 2: IV stage connected to beta22, IV stage not hooked up to PSU, no inputs connected to IV stage. -> faint noise (TV snow like) Step 3: IV stage conneceted, IV stage hooked up to PSU and PSU on, no inputs connected to IV stage. -> faint noise (TV snow like) Step 4: If I connect the inputs on the IVY to IVY GND, the noise is much louder and a hum is added to it. Very sensitive to touching the wires on the input side of the IVY, distance (tested up to about 30cm (11")) does not seem to be of influence on the noise. The issue is indeed not the Beta22 itself, but the IVY or grounding. I have attached an overview of the wiring of the current setup. Is there something I could do to improve the grounding? /edit: I realize this is no longer a beta22 issue and actually should not end up in this thread. Feel free to ignore it and I'll try to post my questions somewhere more appropriate. Thanks. [Edited by ZMN on 02-01-2009 at 12:17 PM.] Attachment: C4607.pdf |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#309] posted on: 02-01-2009 02:34 PM CST (US). | | ZMN, there seems to be several issues with this setup: 1. The IVY has a non-trivial amount of DC offset. 2. The IVY seems to have a non-trivial amount of output noise. 3. The IVY hums when its input is grounded. Having a 4-gang volume pot between the IVY and the β22 boards should alleviate the first two issues by introducing attenuation, so that you're not running with the β22 inputs "wide open" at all times. As for the 3rd issue, this solution may also help but I think it is a problem that needs a different solution. How is your input jack ground and chassis grounding done? You may also want to add output coupling capacitors at the IVY to block its DC offset. |
| wnmnkh Member Joined: Apr. 20, 2008 | Message [#310] posted on: 02-01-2009 10:28 PM CST (US). | | Guess I am very unlucky person  I was checking all of the board careufully again, and found another big problem with other left channel board. Without load, everything seems fine. When connected with load (headphones) the current going through R34/R35 become crazy -it goes whooping 140~160mV! Without load it is normal again 75mV. Kan, can you guess what's happening with this board? |
| johnwmclean Member Joined: Dec. 8, 2008 | Message [#311] posted on: 02-01-2009 11:14 PM CST (US). | | Will a balanced 2 x sigma (headphone only, HD650) build yield benefits from transformers with a greater VA rating than the specified 50VA for two boards? This would be my choice Avel Y23 (Y236207). http://www.avell...l_y23_range.pdf [Edited by johnwmclean on 02-01-2009 at 11:28 PM.] |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#312] posted on: 02-01-2009 11:50 PM CST (US). | | wnmnkh, if you don't yet have a zobel network at the output, try adding one to see if it makes any difference. Otherwise, check the compensation caps C1-C5 to make sure they're installed correctly and have good joints. Note that these caps have values that must be changed with the gain setting of the amp. See the β22 website parts list. johnwmclean, for headphone use only, one suitably-sized transformer and one σ22 should be sufficient. However, going to two σ22 boards has the benefit of reducing the heat dissipation on each of the σ22's MOSFETs, and in many cases where ventilation is limited, this might be very good thing. |
| wnmnkh Member Joined: Apr. 20, 2008 | Message [#313] posted on: 02-01-2009 11:58 PM CST (US). | | You mean the one for the speaker output? Let me try this weekend then. XD |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#314] posted on: 02-02-2009 02:41 AM CST (US). | |
Yes, give it a try. |
| ZMN Member Joined: Nov. 13, 2004 | Message [#315] posted on: 02-02-2009 03:03 PM CST (US). | |
That is relieving news - there are solutions. The input jack is actually an optical SPDIF, for which the digital signal is converted (buffalo DAC) in the same chassis as the beta22s. This chassis is however not connected to ground or earth. Only the first chassis (see previous post attachment, the chassis with the toroids) is connected to ground/earth. I am looking in to the capacitors for IVY. Thanks! |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#316] posted on: 02-02-2009 04:29 PM CST (US). | | ZMN, try adding a wire to connect the IVY's ground to the amp chassis and see if it makes any difference. |
| tedsmith Member Joined: Feb. 3, 2009 | Message [#317] posted on: 02-03-2009 09:20 AM CST (US). | | Hi Guys, I'm building a B22 and have a problem with a ground channel board. The output is almost short to ground (3.62 Ohms). I performed the initial setup as I hadn't noticed the short (as it isn't a dead short) or the short developed during the initial setup. The board passes the initial setup fine, 4.5V on r9-12, 65mV on r34/r35, 0mV DC offset. I only noticed the problem only by chance when I couldn't stop my e12 from oscillating, so built another e12 and same problem, couldn't tune vr1 to 0mv, so I wondered is there something wrong with the ground I'm referenced to (i.e. output of ground board). So I disconnected the ground board and checked for shorts (v+ to g and v- to g ok). This could be just a bit more fine tuning needed on the e12, but I at least by chance found this other problem. If you have any tips on where to start troubleshooting that would be great, I'd like to minimise the amount of desoldering so would prefer advice from someone with experience in debugging this circuit. Many Thanks, [Edited by tedsmith on 02-03-2009 at 09:33 AM.] |
| mrmajestic2 Member Joined: Jun. 5, 2007 | Message [#318] posted on: 02-03-2009 09:24 AM CST (US). | | Im having problems with the fuse blowing on a B22 that I built. The problem started when it moved from a 230V country to a 240V country. I had a 2A fuse here and now it needs to run with a bigger fuse. Also, the turn-on circuit (e24) seems to shut off itself once in a while. Could this also be because of the higher voltage? This never happened here. The build is a 4-board B22 with dual S22 at 30V and two 120VA toroids. Also in the amp is a Joshua Tree with a separate small 20VA toroid and two 5V relay switch boards. |
| amb Headphone Council Joined: Apr. 1, 2004 | Message [#319] posted on: 02-03-2009 09:46 AM CST (US). | | tedsmith, examine your ground channel board carefully for solder bridge or a sliver of under-etched copper that might bridge the output pad or traces to the ground plane (look on both sides, using high magnification lenses). If you see anything suspicious, cut slong the edge of the trace with a sharp x-acto blade. mrmajestic2, if you're using a fast-acting fuse on the transformer primary side, you should change to a slow-blow. Two 120VA toroids plus a 20VA toroid could generate a fairly large turn-on surge. If you already have a slow-blow fuse, then change to a higher rated fuse. The voltage increase from 230V to 240V may just be enough to blow the fuse if your previous fuse was already operating near its rating. Power surges on the AC line can cause the ε24 to power off. |
| tedsmith Member Joined: Feb. 3, 2009 | Message [#320] posted on: 02-03-2009 12:17 PM CST (US). | | I think I need to remove the components covering the output trace's snake down the top side of the board. I've looked at the schematic and the resistance to ground is higher at the other side of every component directly connected to the output, so I think this must be the problem.. but 3.62 Ohms it is going to be a /tiny/ under-etched point or bridge Thanks. [Edited by tedsmith on 02-03-2009 at 12:17 PM.] |
| Page << < 15 16 17 18 19 20 21 22 23 24 25 | Next Page | Prev Topic | Next Topic | ||||
 (remove _nospam_ ) |
© Chu Moy, 2001.