| Posted by Steve L. | December 20, 2021 - 03:44 pm |
| This reply is for Mark W. who posted on 5/3/20 below, and other interested readers. He was asking for thoughts on "monoblocking [his] matching ST70s." The idea is to turn each ST70 into a monaural amp and I will presume that each will be used as a basic power amp, using the grids of V5A and V5B as inputs and cutting out all the preceding circuitry. Perhaps the simplest notion would be to drive the two channels of the ST70 from the same signal and to connect the outputs in parallel. That may be possible but I'm uncomfortable with it because each amp will attempt to force it's output to a specific voltage and they won't in general agree on what that should be, exactly. The problem is particularly acute if there is a lot of negative feedback (NFB), as there is in the ST70A covered by our article. It could cause unnecessary stress on the tubes and may cause distortion. The next simplest notion might be to drive the two amps with 180deg out-of-phase signals and then connect the speakers between the two outputs in a bridging mode. Each amp will see half the load impedance, so the range of loads supported would be 8-32ohms instead of the 4-16ohm normal range. The inversion of phase at the input could be handled by converting an unused tube, say V4, into a phase inverter or by employing a 1:1 transformer with a center-tapped secondary. A concern I have with this method is that since both sides of the speaker are driven, left and right output channels could not share a common ground with each other or with the input. This causes complications for test equipment, headphones and speaker switches. One feature of the ST70A article was to eliminate Eico's idea of grounding the 4-ohm tap, which caused such complications. The other idea would be to rewire the amp so that a single driver runs all 4 output tubes. They would still drive their existing two output transformers but the secondaries would be paralleled. The other channel's driver would be idle. The range of loads supported would be 2-8ohms instead of the normal 4-16ohm range. Alternatively, the secondaries could also be used in series. Then the range of loads supported would be 8-32ohms instead of the 4-16ohm normal range. Negative feedback may need to be adjusted, particularly for the series case. So you see that doing this well isn't an easy task. Mark (if you see this) I apologize for the inordinate delay in answering your query. It turned into a complex question which required some study and that resulted in the oversight. |
| Posted by Steve L. | December 21, 2021 - 07:16 pm |
| Hi Al, I'm sorry if you've tried the ST70A mods and had problems. From your description, I can only conclude that there must be something wrong and I'll be happy to help track that down and get it fixed. There should not be audible low frequency loss. The mod which reduces hiss is the gain reduction in the power amp. Note that this must be done in conjunction with the line stage mod to keep overall gain the same. If your noise increased, there was likely something wrong with the power amp mods. The reduction in the output coupling caps (C19-C22) was not primarily intended to deal with some "hunting or pumping issue." If you are using the power amp mods discussed in the article, I strongly advise against increasing those coupling caps. The reason is that the low frequency response shaping from reducing them is necessary for good low-frequency loop stability. There is a lot more loop gain in the modified power amp and a good deal of effort went into making it rock-stable both at low and high frequencies. I know that this might seem like we're giving up something at low frequency but it's not so. As shown in the test results, low frequency response and power are fine in the ST70A. Response at 20Hz was only 0.14dB down and it was 3dB down at 2.6Hz. As shown by the test data presented on p.9 of the article, 20Hz power increased 46%, so the driver has no problem delivering adequate voltage at low frequency. Also on p.9, you can see that signal to noise ratio (hiss) is improved by 6-12dB, depending on Level Control setting. Perhaps what you experienced is a result of doing the mods piecemeal. Remember that the line amp and power amp mods must be done together and that each mod must be done completely. It won't work to do some pieces and not others. If your amp didn't perform as documented, I will be happy to help find and fix the cause. First though, we need it to be brought to the properly modified condition. Since this discussion may need to get into arcane details, it would be best to carry that on privately. My email address is on the About page which is shown when you click the "About..." button near the top of the home page. Al, this is the first time in the ten years since this article was published that I've found someone who is unhappy with the results. But as far as I'm concerned, that's one too many! I hope that you'll get in touch so we can get to the bottom of this. Thank you for bringing this to my attention. Steve |
| Posted by Al | December 19, 2021 - 04:59 pm |
| Bass performance isnt as good as it can be--Even with these mods. Relying on neg feedback to compensate for poor low end response due to changing grid coupling from 0.1 to 0.047 therefore increasing the driving impedance to 169K ohm--which is in series with potential-divider caused by bias resistances of --about same value, (120K and bias pot,-say 25K) hence halving o/p tube drive at 20Hz... Not good practice IMHO to remove a non-issue. Yes! This L.F. loss is most certainly audible! Makes for a flat, lifeless amp, sorta hollow, shallow soundstage--Like a poor quality sand amp from '70's.... My method to address this, Change o/p coupling-caps to 0.15uF from 0.1uF, Increase bias resistors to from 150K to 280K. With global f/b you'll get below 10Hz and zero sign of the hunting or 'pumping' issue the original cap size reduction was supposed to cure--It was a non issue with my amplifier.... Also,--The mods to reduce the hiss--made my amp hiss Worse! Ive added a coupling-cap to vol cont wiper and grid resistor to improve things. |
| Posted by Steve L. | November 25, 2020 - 06:44 pm |
| Hi Loomis, Sorry you're in quarantine--hope all goes well, otherwise. Answering your questions: #1 - (Higher voltage caps okay?) - Yes, higher voltage ratings will be fine. Sorry that was left off of that line. #2 - (Use 56K instead of bridging?) - Yes, that will be fine. The only thing is, the method of using fixed resistors instead of a pot means that the setting won't be optimum. As mentioned in the text, the pots in my unit ended up about 420K and 290K. The best setting is highly dependent on the tube. I included the fixed 360K alternative as a rough shot at it for folks who can't do the adjustment. With an extreme tube, results might not be satisfactory. #3a - (remove all wires and/or switches etc going to the center speaker, the speaker phase switch, and the balance check switch?) That should be okay, as long as the circuit is left with the same connections it would have with those functions unused. #3b - (What about the hi and lo filter switches?) Again, that should be okay, as long as the circuit is left with the same connections it would have with those functions unused. #4 - (Will the loudness circuit correction mods interfere?) I would have to see the mod schematic to answer definitively but I did a brief search and what I saw at one link didn't seem to indicate a problem. Thank you for your interest and I will be happy to help if you have any other questions. |
| Posted by loomis | November 25, 2020 - 02:23 pm |
| Hello. Thank you so much for creating this detailed package of modifications. I have a few easy questions that I'm hoping you'll be able to answer, so that I can order parts and perform these mods during my quarantine. 1) Usually you comment that it's ok to use higher voltage rated parts. But you don't say that in the parts list next to C7-C10, the 0.1uf 400v film caps in the phono section. Can't I simply upgrade these, and the rest of the coupling caps in this unit, to Panasonic or Orange Drop 630v polypropylene film caps instead of 400v? I wasn't clear if there was some reason to specifically need no higher than 400v there, of if you simply forgot to type "higher voltage ok" next to the parts there. 2) If using fixed resistors, instead of bridging R27 and R28, can't I just remove R27 and R28 and put in something like 56K resistors, instead of bridging the existing resistors with 360K's? 3) After doing these mods, can I then remove all wires and/or switches etc going to the center speaker, the speaker phase switch, and the balance check switch? What about the hi and lo filter switches? 4) Will doing the loudness circuit correction mods interfere in any way with any of your modifications? Thanks again for all your help! |
| Posted by Mark W. | May 03, 2020 - 03:13 pm |
| Hi Steve, Really interesting Hot Rod st70 article well written too. I have decided that going the way of mono blocks is better fro me due to my low power draw Klipsch Cornwell II and Forte' I spkrs. I'm after sound rather than power. Yes even if power is lost to the achievement of better quiet space and definition of actual sound not fb. I use a local tech who has decades of exp. & very sharp so I'm sure he can get done whatever is doable. I'd be grateful to get your thoughts on mono blocking my matching ST70s. mlw |