I've pretty much completely rebuilt my SCA80Q with parts and help from Dan at updatemydynaco.com; all the big capacitors, fully repopulated both preamp boards, new input jacks, new speaker binding posts, new balance control, new caps and resistors on the front panel switches and pots, and new caps and transistors on the power amp boards (the power amp parts were from an eBay supplier).
The amp sounds better than ever, except for a few vagaries....
1. I have lost the left channel completely on two occasions. Lost, as in, NO output even at full volume. No hiss, hum... nothing. Swapping input channel leads confirms that the issue is internal in the SCA80Q. With volume full up, no inputs connected, there is the expected hiss and very slight hum in the right channel, but absolutely nothing in the left channel.
2. There has always been a significant turn-on "thump," and it is still present; if anything, it's even stronger than it was before the rebuild.
3. Here's where it gets weird. The solution to item number 1 above, on both occasions, has been to turn off the amp, wait a few seconds, then turn it back on. When doing so, the turn-on thump is severe... almost a turn-on "CRACK," meaning not just the "thud" usually heard, but a strong full-range POP. After that, both channels work and the amp sounds pretty good.
Have already received advice to swap the pad 19 leads from the pre boards to the amp boards to narrow down where the problem is, and will do that. Anyone have any further insights/places to look? I had originally thought that the issue may lie in the selector switch contacts, but the fact that turning the power off, then back on, seems to negate that. Who knows.
Thanks in advance for any suggestions.
The amp sounds better than ever, except for a few vagaries....
1. I have lost the left channel completely on two occasions. Lost, as in, NO output even at full volume. No hiss, hum... nothing. Swapping input channel leads confirms that the issue is internal in the SCA80Q. With volume full up, no inputs connected, there is the expected hiss and very slight hum in the right channel, but absolutely nothing in the left channel.
2. There has always been a significant turn-on "thump," and it is still present; if anything, it's even stronger than it was before the rebuild.
3. Here's where it gets weird. The solution to item number 1 above, on both occasions, has been to turn off the amp, wait a few seconds, then turn it back on. When doing so, the turn-on thump is severe... almost a turn-on "CRACK," meaning not just the "thud" usually heard, but a strong full-range POP. After that, both channels work and the amp sounds pretty good.
Have already received advice to swap the pad 19 leads from the pre boards to the amp boards to narrow down where the problem is, and will do that. Anyone have any further insights/places to look? I had originally thought that the issue may lie in the selector switch contacts, but the fact that turning the power off, then back on, seems to negate that. Who knows.
Thanks in advance for any suggestions.
If there's no normal noise at all in the bad channel, that does seem like it has to do with the speaker mode switch,
or maybe the headphone jack, both of which interrupt the output signal path. The problem will be after the
power amplifier, not in the preamp section. Odds are it's a bad connection or solder joint.
The loud thump is likely the output coupling capacitor charging. This is a single supply amplifier.
Perhaps the bad connection is at (or related to) the output inductor or the output coupling capacitor.
That would open the signal path and give no sound at all. The large solid wire of the inductor is
rather difficult to solder, so check that.
or maybe the headphone jack, both of which interrupt the output signal path. The problem will be after the
power amplifier, not in the preamp section. Odds are it's a bad connection or solder joint.
The loud thump is likely the output coupling capacitor charging. This is a single supply amplifier.
Perhaps the bad connection is at (or related to) the output inductor or the output coupling capacitor.
That would open the signal path and give no sound at all. The large solid wire of the inductor is
rather difficult to solder, so check that.
Last edited:
A bad connection is full of tarnish. It may be insulating (no sound). The oxide may be punctured by a LARGE signal, and may then work for the rest of the concert. Or not.
This is a little solid-state amp? Run it with small signal, open, and poke a wood/plastic chopstick around inside.
While I have got-through concerts on tarnished contacts (switches or plugs) by BANGing on the tarnish with big signal, it is nicer to find it and clean/fix/solder the contact as appropriate.
This is a little solid-state amp? Run it with small signal, open, and poke a wood/plastic chopstick around inside.
While I have got-through concerts on tarnished contacts (switches or plugs) by BANGing on the tarnish with big signal, it is nicer to find it and clean/fix/solder the contact as appropriate.
Thanks for the responses so far.
Just spent another hour or two tinkering. I THINK the problem of a channel cutting out may have been caused by one of bare jumper wires on the speaker mode switch shorting out, but I'm not sure. It hasn't happened again since I opened the amp up today.
I swapped the pin 19 leads from the preamp boards, and the issue remained in the left channel indicating, as suggested, that the issue may most likely lie within the left power amp board. Swapping physical speakers, followed by swapping left and right channel RCA input leads, continued to indicate that the problem stayed with the output of the left side.
I took all of the voltage test point measurements suggested in the Dynaco manual, amp powered up, inputs shorted, connected to a pair of 8ohm speakers, with the volume down. I've attached a pdf of the spreadsheet containing the results and some notes. Nothing jumped out at me as being way out of spec, I'm sure many folks here have more knowledge and experience than me. Of note was just how close the two fully repopulated preamp boards were... almost identical measurements. THAT was impressive. Both PC-18s measured under the expected voltages listed by Dynaco pretty much across the board (no pun intended), but well within 10% tolerances. The right channel measuring slightly lower than the left. The only thing that jumped out at me was the presence of a few millivolts on pins 1, 2, 11, and 12, all of which were supposed to show no voltage present. The highest reading of those four points I got was 3.4mV on PC-18 pin 2 of the right channel. All others were 2mV or less.
The left channel really does sound different than the right; the right sounds more extended and open, while in comparison the left channel sounds slightly rolled off at top and just dead and two-dimensional. My wife heard it right away, without prompting from me. She's a musician, but doesn't share my quest for audio... if it wasn't obvious, she wouldn't have noticed.
As mentioned in the notes on the attached spreadsheet, about the only things left in this SCA80Q that are original other than the switches, all of the pots except for the new balance control purchased from updatemydynaco.com, and most of the physical wire and hardware, are the components on PC-19, the components that are NOT capacitors on the PC-18 boards, and the Toshiba 2N3772 bipolar transistors and associated ceramic resistor on the left channel heat sink. Everything else has been replaced and upgraded. The right channel bipolar transistor pair are a SPC 571844/571845 pair with a carbon resistor instead of the original ceramic.
What effect DOES the selection and health of the bipolar transistors have on sound quality?
Would replacing the twisted pairs from the balance control to the preamp boards with shielded wires yield improvement in hum rejection? I noticed as I was tinkering around with the top off that getting my hand anywhere near those wires created a ton of hum.
Thanks again, in advance, for any further insight. My test tools are limited to a decent multimeter and a two channel function generator app that I can use to output any frequency sine or square wave at a given voltage.
Just spent another hour or two tinkering. I THINK the problem of a channel cutting out may have been caused by one of bare jumper wires on the speaker mode switch shorting out, but I'm not sure. It hasn't happened again since I opened the amp up today.
I swapped the pin 19 leads from the preamp boards, and the issue remained in the left channel indicating, as suggested, that the issue may most likely lie within the left power amp board. Swapping physical speakers, followed by swapping left and right channel RCA input leads, continued to indicate that the problem stayed with the output of the left side.
I took all of the voltage test point measurements suggested in the Dynaco manual, amp powered up, inputs shorted, connected to a pair of 8ohm speakers, with the volume down. I've attached a pdf of the spreadsheet containing the results and some notes. Nothing jumped out at me as being way out of spec, I'm sure many folks here have more knowledge and experience than me. Of note was just how close the two fully repopulated preamp boards were... almost identical measurements. THAT was impressive. Both PC-18s measured under the expected voltages listed by Dynaco pretty much across the board (no pun intended), but well within 10% tolerances. The right channel measuring slightly lower than the left. The only thing that jumped out at me was the presence of a few millivolts on pins 1, 2, 11, and 12, all of which were supposed to show no voltage present. The highest reading of those four points I got was 3.4mV on PC-18 pin 2 of the right channel. All others were 2mV or less.
The left channel really does sound different than the right; the right sounds more extended and open, while in comparison the left channel sounds slightly rolled off at top and just dead and two-dimensional. My wife heard it right away, without prompting from me. She's a musician, but doesn't share my quest for audio... if it wasn't obvious, she wouldn't have noticed.
As mentioned in the notes on the attached spreadsheet, about the only things left in this SCA80Q that are original other than the switches, all of the pots except for the new balance control purchased from updatemydynaco.com, and most of the physical wire and hardware, are the components on PC-19, the components that are NOT capacitors on the PC-18 boards, and the Toshiba 2N3772 bipolar transistors and associated ceramic resistor on the left channel heat sink. Everything else has been replaced and upgraded. The right channel bipolar transistor pair are a SPC 571844/571845 pair with a carbon resistor instead of the original ceramic.
What effect DOES the selection and health of the bipolar transistors have on sound quality?
Would replacing the twisted pairs from the balance control to the preamp boards with shielded wires yield improvement in hum rejection? I noticed as I was tinkering around with the top off that getting my hand anywhere near those wires created a ton of hum.
Thanks again, in advance, for any further insight. My test tools are limited to a decent multimeter and a two channel function generator app that I can use to output any frequency sine or square wave at a given voltage.
Output transistors, not much. Every TO3 transistor since 2n3772 has had Ft fast enough to reproduce music. Even tinkly bells & top octave grand piano. FT 2 mhz is fine for output transistors. My ST120 had 40636 originally, 400 khz ft, and those were NOT so good. Obsolete since 1980.
Where the highs can get lost in in the driver transistor. Q3 & Q4. I've proved by misapplication of TIP31c/32c that those have to be 20 mhz Ft or higher. Dynaco had access to RCA 50 mhz TO5 transistors for VAS & drivers, equivalent to 2n5320 & 2n5322. I have a pair in my ST120 that sound good. The other pair were blown up, and when I replaced them with NTE49/50 and later TIP31c/32c with Ft 6mhz the result was dull lifeless sound. VAS Q2 has to be pretty fast too. My replacement for that was a 20 mhz GE D44R2 which sounded just fine.
I suggest before you start replacing wiring for hum improvement you invest in a oscilloscope or at minimum a Simpson 380 VOM with 20 vac & 2 vac scales. Scopes are high tech, simpson meters have no electrolytic capacitors in them to get old & work the same 35 years after purchase as they did the day you bought it (unless damaged).
Scopes also require probes, that if stepped on, cost $50. Where the hum starts, that is where you need to apply improvements. AC voltmeters need a capacitor series the minus probe (to analog ground) to not read AC volts on DC sources. I use a .047 uf 400 v plastic film one. Also some alligator clip leads, like the 10 for sale at parts-express.com for $8. Newark has them from pamona, and they charge $70 for 10. Don't ever use two hands inside a powered up amp, voltage >25 across your heart can stop it. Use one hand and an alligator clip lead. If you go any smaller than these big parts, you need two pamona grabber probes or a queballs to not short across two points and blow up something.
Happy hunting.
Last edited:
Indianajo (Is that "Indiana Joe" or "Indianajo," pronounced like "Navajo?"), thanks for the reply. I didn't write back immediately because I'm kind of ignorant to a lot of this and didn't understand some of what you wrote, acronyms, etc., so it took me a day or two to digest it and look up some of the stuff. Can I ask you to elaborate a bit more on what you meant by "misapplication of TIP31c/32c?"
The more I read and understand the troubleshooting and the "For the Service Technician" portions of the Dynaco manual, and your post, the more I think that Q3/4 ARE the source of the problem I'm hearing - the left channel sounding significantly deader, flatter, and duller than the right.... as if the frequency response is narrower. A third party did some repair work on this amp a few decades ago, and it just hasn't sounded the same since. I was hoping that the full repopulations of the preamp boards, replacement of all the large caps, and the capacitors on the PC-18 boards would take care of it, but no dice (although improvements have been heard and noted). I just took some pictures of the differing components between the left and right channels. Keep in mind the left channel sounds worse, but the right channel voltage test point measurements listed on the spreadsheet in my previous post are actually further off than the left channel, with both being seemingly within tolerances. I have NOT measured the legs of Q3 or Q4, and probably should at this point. Do any of these components give clues as to why the left channel sounds worse?
Right channel Q3 -
Left channel Q3 ...
Right channel Q4 ...
Left channel Q4 ...
Right channel Q5/6 ...
Left channel Q5/6...
Right channel R19 ...
Left channel R19 ...
I know substitutions are made as various parts become obsolete and/or unavailable. Is there anything problematic about what you see in terms of substitutions?
Edit: In starting a preemptive search for replacement TIP31c/32c transistors, I don't see any on Digikey that are faster than 3mhz fT... ?!?!? Where does one find a 20mHz one?
Thank you very much!
The more I read and understand the troubleshooting and the "For the Service Technician" portions of the Dynaco manual, and your post, the more I think that Q3/4 ARE the source of the problem I'm hearing - the left channel sounding significantly deader, flatter, and duller than the right.... as if the frequency response is narrower. A third party did some repair work on this amp a few decades ago, and it just hasn't sounded the same since. I was hoping that the full repopulations of the preamp boards, replacement of all the large caps, and the capacitors on the PC-18 boards would take care of it, but no dice (although improvements have been heard and noted). I just took some pictures of the differing components between the left and right channels. Keep in mind the left channel sounds worse, but the right channel voltage test point measurements listed on the spreadsheet in my previous post are actually further off than the left channel, with both being seemingly within tolerances. I have NOT measured the legs of Q3 or Q4, and probably should at this point. Do any of these components give clues as to why the left channel sounds worse?
Right channel Q3 -
Left channel Q3 ...
Right channel Q4 ...
Left channel Q4 ...
Right channel Q5/6 ...
Left channel Q5/6...
Right channel R19 ...
Left channel R19 ...
I know substitutions are made as various parts become obsolete and/or unavailable. Is there anything problematic about what you see in terms of substitutions?
Edit: In starting a preemptive search for replacement TIP31c/32c transistors, I don't see any on Digikey that are faster than 3mhz fT... ?!?!? Where does one find a 20mHz one?
Thank you very much!
Last edited:
I think through experiment, that TIP31c/32c with Ft 3 mhz are unsuited for q3 q4 drivers. I told you I didn't like my TIP41c/42c which were same Ft bigger Pd. I do like MJE15028/29 which come with 30 mhz Ft. My hearing goes to 14 hz, and 3 mhz drivers don't sound right on top octave Steinway grand or tinkly bells. My original dynaco ST120 was panned by reviewers in 1966 for having lifeless sound, and they came with 400 khz output transistors. Somebody that ran a simulation said changing Q5 Q6 from 40636 (400 hz) to TIP3055 (3 mhz) really pepped up the high frequency HD & ID readings. That was on original ST120 which had RCA 2n5320/22 as drivers which were 50 mhz Ft (no longer in production).
Lest channel Q3 is a TIP3055 which IMHO is a 3 mhz part. suitable for output transistors (Q5 Q6) but in my repairs to my blown ST120, not suited for good sounding driver. The other part numbers I don't recognize except to say ECG292 is part of the generic transistor line of either NTE, GE or RCA which were usiversally **** transistors with all the specs left off. I doubt if there is any Ft spec at all on ECG292. You might check on datasheetcatalog.com , my go to website for datasheets.
I think you would be happy with Q3 Q4 MJE15028/29 or 30/31. 2n3772 should be fine for Q5Q6 as long as the 2-3 mhz Ft was maintained. Nearly all epitaxial transistors are that fast or faster.
Indianajo handle is a joke. I live in Indiana. I was looking for my first 8 letter Internet handle, and came up with indianaj because I am a bit adventurous, get around to odd places, and like old archaic stuff. I don't need a hat like IndianaJones with my full head of oily hair, which comes from the 3/8th of my genetic sources (great grandparents) that were native American. The rest of the joke. If people misspell the handle "injunjo" I don't mind. Many of my characteristics come from the native side of the family, whereas my brother got mostly German characteristics. I laugh to think some might think jo is female, others male. I'm mostly a copy of my Mother, with different gonads. She was adventurous, wandered the mountains alone as a preteen and worked at a coal mine after high school. I am patient, play piano & organ, and had fun teaching kids at one point in my life.
Lest channel Q3 is a TIP3055 which IMHO is a 3 mhz part. suitable for output transistors (Q5 Q6) but in my repairs to my blown ST120, not suited for good sounding driver. The other part numbers I don't recognize except to say ECG292 is part of the generic transistor line of either NTE, GE or RCA which were usiversally **** transistors with all the specs left off. I doubt if there is any Ft spec at all on ECG292. You might check on datasheetcatalog.com , my go to website for datasheets.
I think you would be happy with Q3 Q4 MJE15028/29 or 30/31. 2n3772 should be fine for Q5Q6 as long as the 2-3 mhz Ft was maintained. Nearly all epitaxial transistors are that fast or faster.
Indianajo handle is a joke. I live in Indiana. I was looking for my first 8 letter Internet handle, and came up with indianaj because I am a bit adventurous, get around to odd places, and like old archaic stuff. I don't need a hat like IndianaJones with my full head of oily hair, which comes from the 3/8th of my genetic sources (great grandparents) that were native American. The rest of the joke. If people misspell the handle "injunjo" I don't mind. Many of my characteristics come from the native side of the family, whereas my brother got mostly German characteristics. I laugh to think some might think jo is female, others male. I'm mostly a copy of my Mother, with different gonads. She was adventurous, wandered the mountains alone as a preteen and worked at a coal mine after high school. I am patient, play piano & organ, and had fun teaching kids at one point in my life.
Last edited:
Well... it's been a while. An update... I've replaced every transistor, every resistor, every capacitor, and every diode on both PC-18s, PC-19, and the Q5/6 output transistors.
Here's what I used for the transistors..
Q1 BC108B
Q2 2N5320
Q3 TIP31C
Q4 TIP32C
Q5/6 MJ15003
On PC19, I used Ohmite 400ohm 10w resistors and the following diode, an STTH3R04RL, for D4-7...
STTH3R04RL
I had ordered those, not knowing a whole lot about diode specs, prior to someone elsewhere (very respected) recommending 1N5404 diodes for PC19.
Once I buttoned everything back up, R21, a 68 ohm 1/2watt resistor on PC19, blows instantly upon power up. That's happened twice. Could the diodes I used cause that? Here's the circuit....ignore the penciled figures... it's from someone else's pdf.
If anyone wants to read the whole saga/tale of woe, here's the thread from another forum... feel free to zoom to the last page...
Dynaco SCA80Q Attempted rebuild
Thanks for any ideas...
Here's what I used for the transistors..
Q1 BC108B
Q2 2N5320
Q3 TIP31C
Q4 TIP32C
Q5/6 MJ15003
On PC19, I used Ohmite 400ohm 10w resistors and the following diode, an STTH3R04RL, for D4-7...
STTH3R04RL
I had ordered those, not knowing a whole lot about diode specs, prior to someone elsewhere (very respected) recommending 1N5404 diodes for PC19.
Once I buttoned everything back up, R21, a 68 ohm 1/2watt resistor on PC19, blows instantly upon power up. That's happened twice. Could the diodes I used cause that? Here's the circuit....ignore the penciled figures... it's from someone else's pdf.
If anyone wants to read the whole saga/tale of woe, here's the thread from another forum... feel free to zoom to the last page...
Dynaco SCA80Q Attempted rebuild
Thanks for any ideas...
STTH3R04RL is a 3 amp 400 PIV diode. Those specs are suitable for bridge rectifier for a 80 watt amp. It is an ultrafast diode, which tends to accentuate the 120 hz hash you get out of a diode rectifier. C37 may be big enough to filter these sharp edges off, I don't know. The fourrier transform of a more vertical edge has more high frequency components than a more gradual edge of a bland ordinary silicon diode. High frequencies transmit.
I would expect some load problem from terminal 1 to cause the burn up of R21. I'd disconnect c11 triangle then c11 square to see if either is soaking up all that current. Meanwhile you can increase size of R21 to 1 or 2 watts to allow you to feel around for which item is getting hot.
I warned you how unsuitable TIP31c 32c were for drivers. Fine if you only listen to voice & strumma strumma nylon string guitar. No high frequencies to miss. I test my amps with a top octave (solo) Steinway piano recording to listen for a really difficult high frrequency source. I have a Steinway console piano, my recordings are supposed to sound like that. Tinkly bells are hard to reproduce, also. My hearing stops at 14 khz, this is not an acid test. But difficult enough that the first 3 versions of the ST-120 repair failed it.
I'm very suspicious of any modern "2n5320" also. If they are ST, the datasheet I have does show 50 mhz FT. The factory that made RCA 50 mhz Ft 2n5320 in 1965 is now a county park in New Jersey. The number is not copywrited. People can sell anything they want. You have to download the datasheet the day you buy something from the distributor to see what specs they left off. I got caught by On semi selling TIP142 in TO220 instead of TO3p because they changed the datasheet the day I bought them. Such houses as central semi, new jersey semi, NTE, sell commodity parts with highly sought out part numbers, with all the specs left off. Frankly, you won't find any transistor with EBC layout suitable for TO5 drivers or VAS, IMHO. Newark sometimes has a multicomp part in TO39 which can be heatsinked up to 10 W, and they have (sometimes) Ft specs. I bought some, salted them away for my future upgrades, and the burglar hauled them off to the copper scrapyard for me.
I have NTE60 whitebox transistors, probably MJ15003, on my ST120. Same Ft spec, same SOA spec. Pass the top octave piano test when driven by MJE15028/29 or 2n5320/5322 either one ( I have one surviving pair 5320/5322)
I would expect some load problem from terminal 1 to cause the burn up of R21. I'd disconnect c11 triangle then c11 square to see if either is soaking up all that current. Meanwhile you can increase size of R21 to 1 or 2 watts to allow you to feel around for which item is getting hot.
I warned you how unsuitable TIP31c 32c were for drivers. Fine if you only listen to voice & strumma strumma nylon string guitar. No high frequencies to miss. I test my amps with a top octave (solo) Steinway piano recording to listen for a really difficult high frrequency source. I have a Steinway console piano, my recordings are supposed to sound like that. Tinkly bells are hard to reproduce, also. My hearing stops at 14 khz, this is not an acid test. But difficult enough that the first 3 versions of the ST-120 repair failed it.
I'm very suspicious of any modern "2n5320" also. If they are ST, the datasheet I have does show 50 mhz FT. The factory that made RCA 50 mhz Ft 2n5320 in 1965 is now a county park in New Jersey. The number is not copywrited. People can sell anything they want. You have to download the datasheet the day you buy something from the distributor to see what specs they left off. I got caught by On semi selling TIP142 in TO220 instead of TO3p because they changed the datasheet the day I bought them. Such houses as central semi, new jersey semi, NTE, sell commodity parts with highly sought out part numbers, with all the specs left off. Frankly, you won't find any transistor with EBC layout suitable for TO5 drivers or VAS, IMHO. Newark sometimes has a multicomp part in TO39 which can be heatsinked up to 10 W, and they have (sometimes) Ft specs. I bought some, salted them away for my future upgrades, and the burglar hauled them off to the copper scrapyard for me.
I have NTE60 whitebox transistors, probably MJ15003, on my ST120. Same Ft spec, same SOA spec. Pass the top octave piano test when driven by MJE15028/29 or 2n5320/5322 either one ( I have one surviving pair 5320/5322)
Last edited:
Joe, thanks SO much for the detailed and helpful answer. I do recall your warning against the TIP31/32C for drivers, but since they were originally what was used in this version of the amp and I could get them, and based on a bunch of warning elsewhere against faster transistors causing oscillation issues in this amp, I played it "safe" and went ahead and used the TIP31/32C for Q3/4.
I'm such a beginner I didn't think to look PAST PC19 for any reason that R21 might be blowing... duh. Here's where it leads on PC18 in on "A" at eyelet 10... supplying power through R6/R7 and R3 to the base and collector of Q2 and collector of Q1, yes? I'll pull the PC18 boards again and check those resistors, and also make sure I have Q1 and Q2 pinned correctly. I checked a zillion times against the datasheets, and did the typical diode test measurements on the transistors to confirm pinouts.. but screwing up is CERTAINLY not beyond me.
"E" and "F" from PC19 supply power to the preamp boards here, via eyelets 4 and 12. I had completely repopulated the preamp boards about a year ago and they were both performing very well before I started this current PC18 and PC19 renewal... nice clean scope traces comparing input to output on PC17 eyelet 19.
Thank you again, very much, for your input. I'll follow up as soon as I check those resistors and transistors on PC18.
I'm such a beginner I didn't think to look PAST PC19 for any reason that R21 might be blowing... duh. Here's where it leads on PC18 in on "A" at eyelet 10... supplying power through R6/R7 and R3 to the base and collector of Q2 and collector of Q1, yes? I'll pull the PC18 boards again and check those resistors, and also make sure I have Q1 and Q2 pinned correctly. I checked a zillion times against the datasheets, and did the typical diode test measurements on the transistors to confirm pinouts.. but screwing up is CERTAINLY not beyond me.
"E" and "F" from PC19 supply power to the preamp boards here, via eyelets 4 and 12. I had completely repopulated the preamp boards about a year ago and they were both performing very well before I started this current PC18 and PC19 renewal... nice clean scope traces comparing input to output on PC17 eyelet 19.
Thank you again, very much, for your input. I'll follow up as soon as I check those resistors and transistors on PC18.
To speed debug up of R21 problem, all loads are wires. I would disconnect all load wires on R21, marking them with tape & a sharpie. Then I would add them back one at a time with a DVM 10 amp scale in series. Use alligator clip leads, you don't want to use 2 hands on a powered up amplifier. I use a power strip to make turning on & off power convenient to where I sit. The lead that uses a lot of current, that is where your problem lies. Could be a solder bridge or part insertion error, don't assume it is a bad part.
The sound of the dynaco transistor amps prior to ST400 was universally panned in reviews as bad. One problem on the 1966 design ST-120 I have was slow output transistors - 400 khz Ft 40636. MJ15003 corrects that. The other was inadequate cold idle current in the output transistors. The obsolete bias control network D1 C4 C6 C7 is the cause of that problem. D2 D3 need to be mounted out over the Q5 Q6 to pick up the heat of the output transistors & collapse the voltage when hot. They also need a series pot, 50 or 100 ohms, to adjust the cold idle bias out to 20 ma on R19 (40 mv). Apex AX6 has the same 6 transistors but <.1% HD even cold. Since my driver board had lifted broken traces under the drivers, I replaced it with a point to point AX6 board. If your chassis is taller than 3.5" you might be able to fit in a AX6 artwork PCB and solve the whole problem. The difference between ST-120 and your SC80 appears to be that yours has a less capable transformer & filter caps. You'll still have flying wires out to Q5 Q6 on the heat sink. I do and it works. I did put a 8 turn coil in wire 7 back from R19, to trap ultrasonic oscillations.
The sound of the dynaco transistor amps prior to ST400 was universally panned in reviews as bad. One problem on the 1966 design ST-120 I have was slow output transistors - 400 khz Ft 40636. MJ15003 corrects that. The other was inadequate cold idle current in the output transistors. The obsolete bias control network D1 C4 C6 C7 is the cause of that problem. D2 D3 need to be mounted out over the Q5 Q6 to pick up the heat of the output transistors & collapse the voltage when hot. They also need a series pot, 50 or 100 ohms, to adjust the cold idle bias out to 20 ma on R19 (40 mv). Apex AX6 has the same 6 transistors but <.1% HD even cold. Since my driver board had lifted broken traces under the drivers, I replaced it with a point to point AX6 board. If your chassis is taller than 3.5" you might be able to fit in a AX6 artwork PCB and solve the whole problem. The difference between ST-120 and your SC80 appears to be that yours has a less capable transformer & filter caps. You'll still have flying wires out to Q5 Q6 on the heat sink. I do and it works. I did put a 8 turn coil in wire 7 back from R19, to trap ultrasonic oscillations.
Last edited:
Look at it. R21 feeds point A which feeds a 30k and a 1k, either of which will blow-up before a 68r. Oh, and it feed C11. And those wires. Inspect wires for shorts, then replace C11. Maybe it is bad, maybe you put it backward, whatever. If it's not the wires it is that cap.
Thanks very much, Joe and PRR, for the continued suggestions. Joe, I attempted to follow your directions, but I fear I'm still too much a neophyte to fully digest them. The good news is I didn't get electrocuted.. the safety stuff I understood and appreciated! 🙂
Here's some more info as to what I did, the status of C11, and a possible step-by-step troubleshooting plan. Is this basically what you were suggesting?
C11 is no longer the original cap; it's a substitute circuit furnished from UpdateMyDynaco.com ...
From Dan's website...
SCA80 and/or SCA80-Q Replacement Cap for C11 (SCA80C11)-$29.00
![[IMG]](https://updatemydynaco.com/pictures/SCA80C11.JPG "[IMG]")
C11 was the silver-can 3-section filter capacitor that supplied power to:
---end website quote----
I installed that C11 kit, along with his replacements for both C7s and C9, along with his jacks, binding posts, front panel resistors/capacitors, and full preamp board repops, most about a year ago. That C11 sub has always worked fine, but I just took your advice and checked for shorts to ground where you suggested.... no issues that I could find.
However, you've definitely got it narrowed down for me somewhat. When I disconnected the wire from PC19 eyelet 1 to the A terminal of C11, then reconnected all of the other leads to PC19 2 through 6 (I left 7, the ground, attached), R21 stayed fine. The test voltages on PC19 and PC18 per the manual all measure within tolerance, with the expected lack of voltage at eyelet 10 on PC18 (since I don't have anything attached to eyelet 1 on PC19 right now), and that eyelets 1 and 2 on PC19 are identical voltages due to the unattached load of C11 instead of 3v less on eyelet 1 as specified.
Sooo... something either on the C11 board, or something drawing FROM that, must be drawing too much. I had left the wires to eyelets 10 on both PC18s disconnected while connecting the C11 A terminal to PC19 eyelet 1, but R21 still blew like that so I don't think it's anything on the amp boards; the "B" source from PC19 eyelet 2 via C9 to PC18 eyelet 4, as are C and D to eyelets 8 and 6, are connected without any problem.
So, and confirm this for me if you would.. the issue is either something onboard C11 that just coincidentally went south, or a problem with the phono or preamp circuits on one or both PC17 boards drawing from E and/or F via pins 4 or 12.
Would the best way to continue to troubleshoot be ...
1. Remove the connections to 4 and 12 on PC17s, and leave eyelet 10 disconnected on the amp boards.
2. Reconnect C11 A to PC19 eyelet 1 and see if R21 stays intact. If not, the issue on C11.
3. If so, measure voltages on C11 and verify 71V at A/Triangle, 24V at E/Square, and 17.5V at F/Halfmoon.
4. If one of those is bad, the issue is on C11 somewhere. If all good, then....
5. Connect A to eyelet 10, PC18, one board at a time, see if R21 blows, check voltage. If good then,
6. Connect F to PC17 eyelet 4, one board at a time, check R21, check voltage. If good then,
7. Connect # to PC18 eyelet 12, one board at a time, check R21, voltage.
I did order a whole bunch of 1/2w 68 ohms 5% resistors, thankfully.... here's hoping I have enough on hand. Who likes butter with their popped resistors?
Thank you all so much. Apologies for being so dense and needing so much hand-holding. The upside is I'm slightly less dense than I was six months ago.
Here's some more info as to what I did, the status of C11, and a possible step-by-step troubleshooting plan. Is this basically what you were suggesting?
C11 is no longer the original cap; it's a substitute circuit furnished from UpdateMyDynaco.com ...
From Dan's website...
SCA80 and/or SCA80-Q Replacement Cap for C11 (SCA80C11)-$29.00
C11 was the silver-can 3-section filter capacitor that supplied power to:
- amplifier driver sections
- the pre-amps high level/tone control stage
- the phono preamp
---end website quote----
I installed that C11 kit, along with his replacements for both C7s and C9, along with his jacks, binding posts, front panel resistors/capacitors, and full preamp board repops, most about a year ago. That C11 sub has always worked fine, but I just took your advice and checked for shorts to ground where you suggested.... no issues that I could find.
However, you've definitely got it narrowed down for me somewhat. When I disconnected the wire from PC19 eyelet 1 to the A terminal of C11, then reconnected all of the other leads to PC19 2 through 6 (I left 7, the ground, attached), R21 stayed fine. The test voltages on PC19 and PC18 per the manual all measure within tolerance, with the expected lack of voltage at eyelet 10 on PC18 (since I don't have anything attached to eyelet 1 on PC19 right now), and that eyelets 1 and 2 on PC19 are identical voltages due to the unattached load of C11 instead of 3v less on eyelet 1 as specified.
Sooo... something either on the C11 board, or something drawing FROM that, must be drawing too much. I had left the wires to eyelets 10 on both PC18s disconnected while connecting the C11 A terminal to PC19 eyelet 1, but R21 still blew like that so I don't think it's anything on the amp boards; the "B" source from PC19 eyelet 2 via C9 to PC18 eyelet 4, as are C and D to eyelets 8 and 6, are connected without any problem.
So, and confirm this for me if you would.. the issue is either something onboard C11 that just coincidentally went south, or a problem with the phono or preamp circuits on one or both PC17 boards drawing from E and/or F via pins 4 or 12.
Would the best way to continue to troubleshoot be ...
1. Remove the connections to 4 and 12 on PC17s, and leave eyelet 10 disconnected on the amp boards.
2. Reconnect C11 A to PC19 eyelet 1 and see if R21 stays intact. If not, the issue on C11.
3. If so, measure voltages on C11 and verify 71V at A/Triangle, 24V at E/Square, and 17.5V at F/Halfmoon.
4. If one of those is bad, the issue is on C11 somewhere. If all good, then....
5. Connect A to eyelet 10, PC18, one board at a time, see if R21 blows, check voltage. If good then,
6. Connect F to PC17 eyelet 4, one board at a time, check R21, check voltage. If good then,
7. Connect # to PC18 eyelet 12, one board at a time, check R21, voltage.
I did order a whole bunch of 1/2w 68 ohms 5% resistors, thankfully.... here's hoping I have enough on hand. Who likes butter with their popped resistors?
Thank you all so much. Apologies for being so dense and needing so much hand-holding. The upside is I'm slightly less dense than I was six months ago.
Quick follow up. Disconnected all loads from the C11 board; no connections via E, F, or A to PC17s or PC18s. Connected A to PC19 pin 1. R21 still blows.
I don't have the build diagram to show where the wires go between the boards. I can't help you on troubleshooting between the boards. A & B are obvious but I don't have a clue about what F & E are for. You have to eliminate all loads then reconnect one at a times to figure out which one is blowing the R21. Reconnecting the loads with a alligator clip lead through the amps scale of a DVM can show which part is hogging the current. You may have to go to milliamp scale instead of 10 amp scale of DVM to get a decent reading, but start on 10 amps. Blowing those tiny 400 ma fuses inside the DVM is so annoying. Be sure to switch DVM off amps everytime you put it down, because as soon as you do a voltage check on amps, you'll blow another meter fuse.
Djoffe sells a quality product, but preserving the dynaco channel schematic is slavish. Nobody liked that sound when soft. Full power HD was okay, but not soft. Dynaco continued to sell tube ST70 Mark III and IV their entire life because it sounded better. Not that other transistor amps could not sound as good with slightly different wiring. AX6 is about as good as 6 transistors at 70 v can be. Dynaco looks good the controls are industrial grade, the transformers are first rate. The heat sinks may be adequate on SCA-80, the ones on my ST-120 were totally inadequate except for passing the FTC prescribed 1 hour watts test.
Note in general, reducing part burn during debug can be done by putting a 60 w incandescent bulb in series with the AC plug. I have my edison socket inside of a steel chassis that is grounded with a circuit breaker. Those wires can pop out from under the screws of a lamp socket and hit the case. Especially if stranded wire. Electricians use solid wire on lamp fixture screws, that they bend a certain way with electrician's pliers. Higher wattage amps you use a 100 w bulb.
Djoffe sells a quality product, but preserving the dynaco channel schematic is slavish. Nobody liked that sound when soft. Full power HD was okay, but not soft. Dynaco continued to sell tube ST70 Mark III and IV their entire life because it sounded better. Not that other transistor amps could not sound as good with slightly different wiring. AX6 is about as good as 6 transistors at 70 v can be. Dynaco looks good the controls are industrial grade, the transformers are first rate. The heat sinks may be adequate on SCA-80, the ones on my ST-120 were totally inadequate except for passing the FTC prescribed 1 hour watts test.
Note in general, reducing part burn during debug can be done by putting a 60 w incandescent bulb in series with the AC plug. I have my edison socket inside of a steel chassis that is grounded with a circuit breaker. Those wires can pop out from under the screws of a lamp socket and hit the case. Especially if stranded wire. Electricians use solid wire on lamp fixture screws, that they bend a certain way with electrician's pliers. Higher wattage amps you use a 100 w bulb.
Last edited:
Thanks, Joe. F on the C11 board supplies 17.5V to the phono section eyelet 4 on PC17, E supplies 24V to the line level preamp section eyelet 12 on PC17, and A supplies 71V to eyelet 10 on PC18. With NONE of those loads connected, but with C11 connected to its voltage source at eyelet 1 on PC19 coming off R21, R21 still blows. I've pulled the C11 board and inspected it.. no bridges, traces issues, or component failures that I can see. Sooo strange. Dan Joffe has suggested a test to take a raw 72V signal via one of the 400ohm 10w resistors straight to C11 and see what gets hot. Scares me a little bit, but I guess that's the next step.
I haven't rigged a dim bulb tester, although I'm sure that would be a smart thing to do.
I also understand and accept your thoughts on just how good, or not-so-good, sounding this amp is. I built it myself as an 18-year-old kid way back when, and it has a lot of sentimental value, so between that and using it as a tool to learn a lot more about this whole process of troubleshooting and repairing, I guess it's still worth it to me. It served me VERY well through college, marriage, raising three boys, seven dogs, and decades of great music listening. Have moved on to Hafler, Van Alstine, and B&K amps, Akitika, Adcom, and Hafler preamps (still building at least kits whenever I can), and a whole bunch of other stuff that does, indeed, confirm your assertions re/ the entry-levelness of the Dynaco SCA80Q. Still... I love it. I hate it now... but I love it. 🙂
I haven't rigged a dim bulb tester, although I'm sure that would be a smart thing to do.
I also understand and accept your thoughts on just how good, or not-so-good, sounding this amp is. I built it myself as an 18-year-old kid way back when, and it has a lot of sentimental value, so between that and using it as a tool to learn a lot more about this whole process of troubleshooting and repairing, I guess it's still worth it to me. It served me VERY well through college, marriage, raising three boys, seven dogs, and decades of great music listening. Have moved on to Hafler, Van Alstine, and B&K amps, Akitika, Adcom, and Hafler preamps (still building at least kits whenever I can), and a whole bunch of other stuff that does, indeed, confirm your assertions re/ the entry-levelness of the Dynaco SCA80Q. Still... I love it. I hate it now... but I love it. 🙂
Be sure to wear safety glasses when doing this test. Shorted electrolytic capacitors can explode. Earplugs too, perhaps.
Thanks, Joe... I'll be careful. Just to reiterate, "C11" in my amp isn't the original multi-tap cap w/ resistors. It's a circuit board consisting of 2 470uf, 1 2200uf, and 1 22uf electrolytic caps, a few resistors, a 1N4004 diode, a TL783 regulator (for the 24V) and a TL431 regulator (for the 17.5V).
Will report back in a bit .. hopefully with most of my limbs intact.
Will report back in a bit .. hopefully with most of my limbs intact.
Well.... I guess it's time for the "thrilling conclusion." After, with Dan's help, calculating the expected current draw of his C11 board (which was a measly 11.9mA, which would mean 9.6mW across the 1/2 watt 68 ohm resistor using P=R(I^2).. which you wouldn't THINK would be a problem. After confirming that C11 was not drawing more than it should (by feeding it raw 72V from one of the 400ohm 10W R16 resistors with an ammeter in series... read 11mA as expected), reinspecting and confirming every trace and ground for continuity where there should be, and everything else I could think of, I wired everything back up.. and blew another 68ohm R21. Sigh..
Thinking, maybe, that the batch of 68 ohm resistors I had were not rated 1/2 watt, or defective.. or just because I was totally frustrated... I opened up an Arduino starter kit I have and pawed through the resistors that came with it. Found a 100ohm resister that looked MUCH more substantial than the 68ohm ones I was using. Installed one for R21. Powered up... slight little spark/puff, but the resistor held and still measured OK with power off. Checked to make sure dummy loads were still attached well, fired up the scope, attached function generator (OK, cell phone) with 1kHz sine, power on... and... scope looks good!!! 100hZ square wave... some tilt, but no ringing (other than my phone's lame output)... same w/ 50hZ. Soooo... function generator off, dummy loads off, music source and speakers attached and...
IT WORKS!!!! Sounds much better than it ever has.. ever. Both channels are beautifully matched, mono is centered and stable.
Soooo... what was the problem?
There's almost 1000uF of capacitance on C11 that charges upon power up and, I guess, for a very tiny amount of time, there's (72V^2)/68 watts (76.2W!) going through that resistor until the caps are charged. The original resistors in this amp were all carbon comp, and I'm told they excel at withstanding surges much in excess of their rated power. Not knowing any better (you can put that phrase in front of pretty much ANYTHING I do inside an amp, I think), I had ordered and was using these...
https://www.digikey.com/en/products...oric-bc-components/SFR16S0006809JA500/5065274
... metal film resistors... which turns out HATE surges. Will pick up some more appropriate resistors today and install the correct 68ohm carbon comp.
Anyway... I THINK we finally have a happy ending.
Thank you, Joe and everyone!
Thinking, maybe, that the batch of 68 ohm resistors I had were not rated 1/2 watt, or defective.. or just because I was totally frustrated... I opened up an Arduino starter kit I have and pawed through the resistors that came with it. Found a 100ohm resister that looked MUCH more substantial than the 68ohm ones I was using. Installed one for R21. Powered up... slight little spark/puff, but the resistor held and still measured OK with power off. Checked to make sure dummy loads were still attached well, fired up the scope, attached function generator (OK, cell phone) with 1kHz sine, power on... and... scope looks good!!! 100hZ square wave... some tilt, but no ringing (other than my phone's lame output)... same w/ 50hZ. Soooo... function generator off, dummy loads off, music source and speakers attached and...
IT WORKS!!!! Sounds much better than it ever has.. ever. Both channels are beautifully matched, mono is centered and stable.
Soooo... what was the problem?
There's almost 1000uF of capacitance on C11 that charges upon power up and, I guess, for a very tiny amount of time, there's (72V^2)/68 watts (76.2W!) going through that resistor until the caps are charged. The original resistors in this amp were all carbon comp, and I'm told they excel at withstanding surges much in excess of their rated power. Not knowing any better (you can put that phrase in front of pretty much ANYTHING I do inside an amp, I think), I had ordered and was using these...
https://www.digikey.com/en/products...oric-bc-components/SFR16S0006809JA500/5065274
... metal film resistors... which turns out HATE surges. Will pick up some more appropriate resistors today and install the correct 68ohm carbon comp.
Anyway... I THINK we finally have a happy ending.
Thank you, Joe and everyone!
I use a lot of metal film resistors since I discovered they hiss a lot less than carbon comp: In the 220 kohm size that are the plate resistors in my dynakit ST70. BUT 600 mw metal film resistors are so tiny, they are not rated for 350 v. I still do some tube work on ST70 and Hammond organs, so I stock 2 and 3 watt metal film resistors. Haven't burned one yet. I didn't know about the surge current limit.
Another thought, the Allen Bradley and Sprague mil-spec carbon comp resistor factories are now parking lots or shopping malls or something. Dynakit used the good ones produced on same lines as RCR07g### parts. Carbon comps are now made by boutique specialty shops for radio restorer purists. The reliability might be there, it might not? A lot of cheapo carbon comps used in TV's etc in the sixties did not have the mil-spec paint that kept Navy equipment from absorbing water out of the air after 1961. The new carbon comps have what paint? US hammond organs (post 1961) are fine on islands. Netherlands built Hammond organs, the resistors go all low value on the channel Islands.
There is a time to use a steam locomotive (1950) and there is a time to change to diesels (1960) as even N&W that hauled a billion tons of coal a year found out. Supply of feedwater pumps etc dried up.
Another thought, the Allen Bradley and Sprague mil-spec carbon comp resistor factories are now parking lots or shopping malls or something. Dynakit used the good ones produced on same lines as RCR07g### parts. Carbon comps are now made by boutique specialty shops for radio restorer purists. The reliability might be there, it might not? A lot of cheapo carbon comps used in TV's etc in the sixties did not have the mil-spec paint that kept Navy equipment from absorbing water out of the air after 1961. The new carbon comps have what paint? US hammond organs (post 1961) are fine on islands. Netherlands built Hammond organs, the resistors go all low value on the channel Islands.
There is a time to use a steam locomotive (1950) and there is a time to change to diesels (1960) as even N&W that hauled a billion tons of coal a year found out. Supply of feedwater pumps etc dried up.