Altec can capacitor upgrade?
- Thread starter grey
- Start date
Russellc
Addicted Member
This is my line of thinking for working on just about anything. I'm also reading that technology limitations and cost sometimes made the decision on values here. Stock values will definitely keep me out of trouble with this.
Good advice! I have three .1uf bypass caps.(One in each section?) for ONE amp. in the order list.
Changed the bias cap on the list to (100UF 100V ELECT AUDIO).
Sure, I will doc. the build and post when done. Hopefully, there will not be anything major along the way. If there is there is no thread, you will know that that one of them turned into flames.
Looks like two votes for original values.:scratch2:
Oh we like flames too! If something horrible happens, (heaven forbid the loss of Altec) document the event with pics. I'm sure It'll go fine, and I'm looking forward to seeing the complete project.
russellc
Jim-
I replaced C11 A and B with 50uF (100 combined) C6 A and B also 50uF.
Replaced all the other caps (these were the parts I ordered from you)
The amps now work, but I'm getting quite a bit of hum from both and not sure why. I do notice that if I adjust the bias from the recommended -37v to -50v or so the hum goes away completely ( I think this mean the output tubes are not being driven as hard...) You think going to 100/100 for C6 would help here?
Also the bias trim pots are shot...do you have any replacements?
Thanks
Kirk
Jim McShane
Well-Known Member
Hi Kirk!
What exactly do you mean by "quite a bit" (I'm not trying to nit-pick, just narrow things down)? Can you hear the hum from your normal listening position at normal volume? Or do you have to put your ear up to the speaker to hear it?
You probably either have an error in the wiring or you have a ground problem. Did the amps hum like this before you worked on them? Do both hum identically? It could be a ground loop.
Are you sure the amp is the source of the hum. If you run the amps with the inputs shorted does the hum go away?
When you adjust the bias more negative it does reduce the current draw which could reduce hum, yes.
On the pots - if you can post/send a picture of them it will be a big help.
If you can answer the questions I'll do my best to help you out. Hum is TOUGH - it takes a lot of effort sometimes to get rid of it.
Thanks for the input and support from everyone. Ordered the 50/50 F&T caps. along with everything else they should need. I will probably step up the C6 values later as recommended by Jim and others.
One change I want to do is bias the tubes individually. I have somewhat of an idea as to how I would do it. I'm taking a guess that I will need 4) 1K+ pots, 1 or 10ohm 1% resistors to check against and individual caps? Can anyone assist with this?
One change I want to do is bias the tubes individually. I have somewhat of an idea as to how I would do it. I'm taking a guess that I will need 4) 1K+ pots, 1 or 10ohm 1% resistors to check against and individual caps? Can anyone assist with this?
similost
Rockin and Rollin....
There is a member here that rebuilt one with a way to bias each tube by pressing a button for that tube, then reading the bias on a meter he installed too..
You might search around here to see if you can find the couple pictures and posts about it and find who it was. He could help you out with it... I'll look around some myself..
You might search around here to see if you can find the couple pictures and posts about it and find who it was. He could help you out with it... I'll look around some myself..
similost
Rockin and Rollin....
similost
Rockin and Rollin....
These two posts in particular..
http://www.audiokarma.org/forums/showthread.php?t=232717&page=2
http://www.audiokarma.org/forums/showpost.php?p=2977610&postcount=80
http://www.audiokarma.org/forums/showthread.php?t=232717&page=2
http://www.audiokarma.org/forums/showpost.php?p=2977610&postcount=80
bharper
tube junkie...
Altec 1569 - part one
ok boys, here it comes... altec purest may want to turn their head!
I bought a couple of 1569a's off ebay probably 5 years ago. A modder had started on one, and gave up. Was changed to dc filaments for the drivers, guts hanging out, extremely rustry transformers and chassis, and the tube sockets and wiring were gone for the 5u4b's, converted to solid state. The first hv cap in the supply was changed to two 260mfd 350v caps in series, mounted where both of the original caps were mounted. I decided to tackle this one. The dc filaments were added because of hum problems he was having during the mods due to ground loops.
I had previously mod'd a pair of webster theater amps, and decided to maybe go a step or two further on this one. It was junk as it sat.
First thing I noticed about the 1569a was that there were no cathode resistors in the output stage, and just a single bias pot. So, one could not even measure plate current on the output tubes, and a quad of output tubes would have to be carefully matched, to maintain reasonable current offset from side to side on the output transformer. General rule of thumb I seem to have heard is that you would like the side to side current on the output tranny to be within 1% of each other. Hmmmm, must use el34's, and be closely matched to get good results. Not my idea of a good time, I wanted to be able to measure each tube, did not want to be stuck on el34's and wanted to be able to bias up anything I wanted, and not in matched quads.
So I decided that I wanted independent bias for each tube, onboard metering and switching to make it convenient. And while I was at it, I also decided being able to run triode or pentode would be another worthy addition.
First, I pulled both of the transformers, wire brushed the endbells and the cores. Repainted the endbells in satin black, and sealed the cores with multiple coats of hair spray. Painted the mounting brackets of the power transformer silver. Repainted the top and bottom sides of the chassis that were quite rusted with an old can of gold paint, that had partly seperated. But ended up with a nice patena, with essentially a clear coating.
While the transformers were out, I milled the chassis to accept an iec socket that had rfi filtering built in so could replace the trashed power cord. The area around the previous rectifier sockets was fairly trashed, so milled out that area rectangular in shape, and made a cover out of a piece of gold anodized aluminum from a scrapped piece of electronic equipment. This is where I put the bias pots, switch, and meter for the bias supply.
I purchased some vintage minature meters, a few minature rotary switches, and a large lot of 10k pots off ebay. Ordered in some solen caps and some small electrolytic caps, and started in on the beast.
Next post are a few shots of the amp when completed, and then will walk you through what was done, and how to do the independent bias and triode modes.
Brian
ok boys, here it comes... altec purest may want to turn their head!
I bought a couple of 1569a's off ebay probably 5 years ago. A modder had started on one, and gave up. Was changed to dc filaments for the drivers, guts hanging out, extremely rustry transformers and chassis, and the tube sockets and wiring were gone for the 5u4b's, converted to solid state. The first hv cap in the supply was changed to two 260mfd 350v caps in series, mounted where both of the original caps were mounted. I decided to tackle this one. The dc filaments were added because of hum problems he was having during the mods due to ground loops.
I had previously mod'd a pair of webster theater amps, and decided to maybe go a step or two further on this one. It was junk as it sat.
First thing I noticed about the 1569a was that there were no cathode resistors in the output stage, and just a single bias pot. So, one could not even measure plate current on the output tubes, and a quad of output tubes would have to be carefully matched, to maintain reasonable current offset from side to side on the output transformer. General rule of thumb I seem to have heard is that you would like the side to side current on the output tranny to be within 1% of each other. Hmmmm, must use el34's, and be closely matched to get good results. Not my idea of a good time, I wanted to be able to measure each tube, did not want to be stuck on el34's and wanted to be able to bias up anything I wanted, and not in matched quads.
So I decided that I wanted independent bias for each tube, onboard metering and switching to make it convenient. And while I was at it, I also decided being able to run triode or pentode would be another worthy addition.
First, I pulled both of the transformers, wire brushed the endbells and the cores. Repainted the endbells in satin black, and sealed the cores with multiple coats of hair spray. Painted the mounting brackets of the power transformer silver. Repainted the top and bottom sides of the chassis that were quite rusted with an old can of gold paint, that had partly seperated. But ended up with a nice patena, with essentially a clear coating.
While the transformers were out, I milled the chassis to accept an iec socket that had rfi filtering built in so could replace the trashed power cord. The area around the previous rectifier sockets was fairly trashed, so milled out that area rectangular in shape, and made a cover out of a piece of gold anodized aluminum from a scrapped piece of electronic equipment. This is where I put the bias pots, switch, and meter for the bias supply.
I purchased some vintage minature meters, a few minature rotary switches, and a large lot of 10k pots off ebay. Ordered in some solen caps and some small electrolytic caps, and started in on the beast.
Next post are a few shots of the amp when completed, and then will walk you through what was done, and how to do the independent bias and triode modes.
Brian
Glassweights
Compulsive about tubes
Wow!
That was junk when you found it?
Very nice work without going crazy on it!
Dan
That was junk when you found it?
Very nice work without going crazy on it!
Dan
bharper
tube junkie...
Altec 1569 - part 3
Since I am not much of an artist, I am going to talk you through how to approach these mods referencing the schematic shown on the first page of this thread, I believe I can do that accurately. So grab a copy, rotate it so you do not get a stiff neck, and lets start.
Many parts were selected depending what was on hand, I have for example a lot of dale resistors, but as it would go, frequently do not have what I need at the moment. So, you will see in the pictures I may series or parallel resistors to get exactly what is needed.
To read the bias on each tube, we need 10 ohm resistors in the cathode to ground path. In my case, I left pin one of the el34's grounded, and then removed the jumper to pin 8 and inserted a 1/2 watt 10 ohm dale resistor. Your mileage may vary here, some tubes may not like pin one at ground. All I have run so far in this amp are kt90's, and a pair of genelex kt88's, and neither of these minded.
With a 10 ohm resistor in place, you can now measure plate current between pin 8 and pin 1, or pin 8 and ground. Many may want to stop here, you can now read plate current, if you are competent in probing around high voltage with your meter, and don't mind having your amp on its side to read it.
It is a lot easier however, to be able to switch what tube is being read, and have a meter onboard to display that.
With the resistors in place, now just run a wire from each tubes cathode (pin 8), to a rotary switch that has at least 4 positions or more available. Mine I think has like eight or ten. I left the first position full counter clockwise blank, so that would be meter turned off, then rotating clockwise pickup tubes 1,2,3, & 4. With these hooked up now to the contacts at four positions of rotation, we can just bring out the common contact on the rotary switch to a meter movement(through a resistor).
Most common meter movements are a 1ma full scale movement, and most are what is generally a 1000 ohm per volt meter, meaning if you hook a 1k resistor in series with a 1ma meter, most of the time you have just built a 1v meter, full scale.
Back to our 10 ohm cathode resistors. If that resistor had a current of 100ma flowing through it - E = I*R - the voltage seen would be 1 volt.
1volt = .1a*10ohm. So our above 1v full scale meter would be full scale at 100ma, 1/2 scale at 50ma, ect. Depending on tube, I generally bias between 40 and 50ma, which would be at the .4 or .5 position on a 0-1 meter.
Meter would have the series R of about 1k from the + terminal to the common on the rotary switch, with the - lead of the meter connecting to ground.
On the vintage meter I used, I ended up with a series R of 498 ohms, so it must have been a 500ohm/volt movement. I started with I think like a 10k resistor, not knowing the sensitivity of the meter, and started reducing until I saw deflection ok, then connected my dvm to monitor the voltage across the 10 ohm resistor, and fine tuned the resistor so the dvm and the onboard meter were both reading the same.
If I have not totally lost you, next will go on to part 4, on how to individually bias the tubes, now that we can read the current of them individually.
Since I am not much of an artist, I am going to talk you through how to approach these mods referencing the schematic shown on the first page of this thread, I believe I can do that accurately. So grab a copy, rotate it so you do not get a stiff neck, and lets start.
Many parts were selected depending what was on hand, I have for example a lot of dale resistors, but as it would go, frequently do not have what I need at the moment. So, you will see in the pictures I may series or parallel resistors to get exactly what is needed.
To read the bias on each tube, we need 10 ohm resistors in the cathode to ground path. In my case, I left pin one of the el34's grounded, and then removed the jumper to pin 8 and inserted a 1/2 watt 10 ohm dale resistor. Your mileage may vary here, some tubes may not like pin one at ground. All I have run so far in this amp are kt90's, and a pair of genelex kt88's, and neither of these minded.
With a 10 ohm resistor in place, you can now measure plate current between pin 8 and pin 1, or pin 8 and ground. Many may want to stop here, you can now read plate current, if you are competent in probing around high voltage with your meter, and don't mind having your amp on its side to read it.
It is a lot easier however, to be able to switch what tube is being read, and have a meter onboard to display that.
With the resistors in place, now just run a wire from each tubes cathode (pin 8), to a rotary switch that has at least 4 positions or more available. Mine I think has like eight or ten. I left the first position full counter clockwise blank, so that would be meter turned off, then rotating clockwise pickup tubes 1,2,3, & 4. With these hooked up now to the contacts at four positions of rotation, we can just bring out the common contact on the rotary switch to a meter movement(through a resistor).
Most common meter movements are a 1ma full scale movement, and most are what is generally a 1000 ohm per volt meter, meaning if you hook a 1k resistor in series with a 1ma meter, most of the time you have just built a 1v meter, full scale.
Back to our 10 ohm cathode resistors. If that resistor had a current of 100ma flowing through it - E = I*R - the voltage seen would be 1 volt.
1volt = .1a*10ohm. So our above 1v full scale meter would be full scale at 100ma, 1/2 scale at 50ma, ect. Depending on tube, I generally bias between 40 and 50ma, which would be at the .4 or .5 position on a 0-1 meter.
Meter would have the series R of about 1k from the + terminal to the common on the rotary switch, with the - lead of the meter connecting to ground.
On the vintage meter I used, I ended up with a series R of 498 ohms, so it must have been a 500ohm/volt movement. I started with I think like a 10k resistor, not knowing the sensitivity of the meter, and started reducing until I saw deflection ok, then connected my dvm to monitor the voltage across the 10 ohm resistor, and fine tuned the resistor so the dvm and the onboard meter were both reading the same.
If I have not totally lost you, next will go on to part 4, on how to individually bias the tubes, now that we can read the current of them individually.
Last edited:
bharper
tube junkie...
Altec 1569 - part 4
Last thoughts on metering... One could also use a digital panel meter, cheap chinese ones are typically like $10 or less out there. If you do go away from rectifier tubes, the 5vac winding could be used to power the dpm.
Instead of removing the rectifiers, one could also just make a new panel, say right of the speaker connectors, and run wiring through the matching transformer socket hole. If you don't mind hacking up the chassis, there is room to mount the pots, switch, and meter on one side or the other of the amp. I mounted mine where I did since the rectifier sockets were gone, chassis was mangled when they had been previously removed, and I was already committed to going ss rectification, because I wanted more B+ to run a wider range of tubes, and wanted to gain the filament power that went to the 5u4's back to making more power available for bigger output tubes.
Now lets talk about independent bias. Looking at the schematic, we have a tap off of the tranny secondary, slightly off center, that is rectified half wave and connects to the current bias pot. That rectifer was replaced with a 1 amp 1n4007 type rectifier, and the existing bias pot was disconnected. The anode of that diode is now filtered with a 200mfd/150v cap ( - side to anode, + side to ground), bypassed with a .01 solen fastcap (polypropylene).
Now we need to add 4 pots to individually control each tube, and split up the common circuitry that used to do that. I used four 10k pots, and connected the top and bottom connections of the pots in parallel across all 4. So now have two wires, one paralleling all the top side connections, and one wire paralleling all the bottom side connections. One side connects to a 1k ohm 5 watt resistor, that then goes to ground. The other side hooks to the above 200mfd cap to its negative lead, same lead the anode of the rectifier is hooked to. The 4 pots in parallel are 2.5k, the 1k in series to ground makes the total now 3.5k, which is closer to our original 5k pot (did not want to double the current to ground, plus with the 1k on the bottom, we never need to take the bias to 0 ground level anyway). Now we have four variable voltage bias pots to hook up. Obviously, the center wiper contact is the source of this variable voltage we will be hooking to each tube individually. If when done, you want the direction of the shaft movement to bias with to reverse, just flip the parallel wires on the ends of the pots with each other, that connect to the 1k resistor, or the 200mfd cap.
Ok, lets rewire the output tubes to use the new pots. First, look at my photo of the turret board. Instead of 2 couplng caps feeding the pairs of output tubes, now there are 4. I used .47mfd 630v solens here, instead of the stock .22, in hoping to increase bottom end just a little. After removing the old caps, I added turret pins to this board to accomidate te 4 coupling caps, series resistors to the bias pots, and electrolytic bypass caps for each of the 4 circuits. The two dale resistors to the left of each coupling caps are 49.9k each hooked in series, because I was out of 100k dales at that time.
Back to the schematic to make sense of this madness.
Take a look at V3, the upper left el34. On pin 5, you have a 100 ohm grid stop resistor, r16, hooked to c9, cap back to the plate of the driver. At the intersection of r16 and c9, we also have r15, which is a 100k resistor that is hooked back to c7, the 50mfd cap, and the wiper of the old bias pot. All we need to do is do that seperately 4 times, with each 100k hooking to its own of the 4 bias pots wiper. R17 will no longer connect to r16, but to its own coupling cap back to the same plate side connection of c9 to the driver, and to its own 100k resistor back to another bias pot wiper, bypassed on the wiper end with a 47mfd/100v electrolytic (those are seen in the turret board photo, left end of the turret board). R18 will no longer connect to r19, and r15 will no longer connect to r14. You must keep the tube pairs as they are, connected to the same side of the drivers, or the results will only be pretty for a few seconds as all goes up in sparkes and smoke.
You probably do not absolutely have to have the new 47mfd caps on each circuit, but I like having those there to bypass any noise picked up on the way to the 100k resistors, somewhat removing noise from the bias pots, and making interaction between the bias pots and tubes a little less. The original bias filter - c7 - is not retained.
after a short break, will be back regarding pentode - triode mode, and closing thoughts in part 5.
Last thoughts on metering... One could also use a digital panel meter, cheap chinese ones are typically like $10 or less out there. If you do go away from rectifier tubes, the 5vac winding could be used to power the dpm.
Instead of removing the rectifiers, one could also just make a new panel, say right of the speaker connectors, and run wiring through the matching transformer socket hole. If you don't mind hacking up the chassis, there is room to mount the pots, switch, and meter on one side or the other of the amp. I mounted mine where I did since the rectifier sockets were gone, chassis was mangled when they had been previously removed, and I was already committed to going ss rectification, because I wanted more B+ to run a wider range of tubes, and wanted to gain the filament power that went to the 5u4's back to making more power available for bigger output tubes.
Now lets talk about independent bias. Looking at the schematic, we have a tap off of the tranny secondary, slightly off center, that is rectified half wave and connects to the current bias pot. That rectifer was replaced with a 1 amp 1n4007 type rectifier, and the existing bias pot was disconnected. The anode of that diode is now filtered with a 200mfd/150v cap ( - side to anode, + side to ground), bypassed with a .01 solen fastcap (polypropylene).
Now we need to add 4 pots to individually control each tube, and split up the common circuitry that used to do that. I used four 10k pots, and connected the top and bottom connections of the pots in parallel across all 4. So now have two wires, one paralleling all the top side connections, and one wire paralleling all the bottom side connections. One side connects to a 1k ohm 5 watt resistor, that then goes to ground. The other side hooks to the above 200mfd cap to its negative lead, same lead the anode of the rectifier is hooked to. The 4 pots in parallel are 2.5k, the 1k in series to ground makes the total now 3.5k, which is closer to our original 5k pot (did not want to double the current to ground, plus with the 1k on the bottom, we never need to take the bias to 0 ground level anyway). Now we have four variable voltage bias pots to hook up. Obviously, the center wiper contact is the source of this variable voltage we will be hooking to each tube individually. If when done, you want the direction of the shaft movement to bias with to reverse, just flip the parallel wires on the ends of the pots with each other, that connect to the 1k resistor, or the 200mfd cap.
Ok, lets rewire the output tubes to use the new pots. First, look at my photo of the turret board. Instead of 2 couplng caps feeding the pairs of output tubes, now there are 4. I used .47mfd 630v solens here, instead of the stock .22, in hoping to increase bottom end just a little. After removing the old caps, I added turret pins to this board to accomidate te 4 coupling caps, series resistors to the bias pots, and electrolytic bypass caps for each of the 4 circuits. The two dale resistors to the left of each coupling caps are 49.9k each hooked in series, because I was out of 100k dales at that time.
Back to the schematic to make sense of this madness.
Take a look at V3, the upper left el34. On pin 5, you have a 100 ohm grid stop resistor, r16, hooked to c9, cap back to the plate of the driver. At the intersection of r16 and c9, we also have r15, which is a 100k resistor that is hooked back to c7, the 50mfd cap, and the wiper of the old bias pot. All we need to do is do that seperately 4 times, with each 100k hooking to its own of the 4 bias pots wiper. R17 will no longer connect to r16, but to its own coupling cap back to the same plate side connection of c9 to the driver, and to its own 100k resistor back to another bias pot wiper, bypassed on the wiper end with a 47mfd/100v electrolytic (those are seen in the turret board photo, left end of the turret board). R18 will no longer connect to r19, and r15 will no longer connect to r14. You must keep the tube pairs as they are, connected to the same side of the drivers, or the results will only be pretty for a few seconds as all goes up in sparkes and smoke.
You probably do not absolutely have to have the new 47mfd caps on each circuit, but I like having those there to bypass any noise picked up on the way to the 100k resistors, somewhat removing noise from the bias pots, and making interaction between the bias pots and tubes a little less. The original bias filter - c7 - is not retained.
after a short break, will be back regarding pentode - triode mode, and closing thoughts in part 5.
Last edited:
bharper.....that thing rocks!:ntwrthy: I really appreciate all of the details. This entire thread solves plenty of questions that I had rolling around in my head.
The stock bias circuit is the only thing I don't like so far. My amps should have something similar to your circuit VERY shortly after I have them working. Every amp should have this ability in my eyes. The 128b had it out of the box...at least the ability to check. Why not all of them? There is a matched quad of tubes that will serve to get them working for the short term.
I'm aiming to keep the face/cover and rectifier tubes. I May add some some tall feet to roll it over on it's face. What I'm thinking is four banana jacks to check the bias. I do admire what you have done with the bias items externally.:scratch2:
The stock bias circuit is the only thing I don't like so far. My amps should have something similar to your circuit VERY shortly after I have them working. Every amp should have this ability in my eyes. The 128b had it out of the box...at least the ability to check. Why not all of them? There is a matched quad of tubes that will serve to get them working for the short term.
I'm aiming to keep the face/cover and rectifier tubes. I May add some some tall feet to roll it over on it's face. What I'm thinking is four banana jacks to check the bias. I do admire what you have done with the bias items externally.:scratch2:
Attachments
Last edited:
bharper
tube junkie...
Altec 1569 - part 5
Pentode - Triode switching is pretty simple. You need a dpdt switch, and a couple of 47 ohm - 100 ohm 2 watt resistors. Plus, I added in a 22ufd 630v polypropylene solen cap.
Back on the schematic, look at v5 and v6 pin 4, the screen connection. Also note that on each side, pin 4 also connects to its partner parallel tube.
What I see here stock, I do not like, but maybe it is just me. R20 that is feeding the screens, the 250 ohm 5 watt resistor drops down the plate vcc, but is not bypassed. So, on the tube side of r20, I added a 22mfd solen cap to reduce the impedence seen by the screens, and hopefully reduce them moving around as much in voltage.
With the dpdt switch, the two center terminals will hook to pin 4 of each of the sides of the output stage, in place of the connections to r20. So one connection from one center to v5 pin 4, and one connection from the other center pin to v6 pin 4, with no remaining connection to r20. Pick an end of the toggle switch, and hook the side of r20 that had been connected to the pin 4's to both sides of the connections on that end. And again, that end of r20 or the switch connected to the 22mfd solen that the other side of connects to ground (of the solen). Now we have pentode mode again, when the switch is in the correct position to reconnect the pin 4's to r20/solen.
For triode mode, now hook one connection each of the other end of the switch through a 47 ohm 2 watt resistor back to its corresponding output tube pairs plate (pin 3). Be very certain that the side of the switch that goes to pin 4 for example of v5, is the same side on the other end that has the resistor going to pin 3 of v5. You end up with two resistors, each going to one side of the same end of the dpdt switch, both matching the output pairs pin 4 side of the switch, with one resistor to pin 3 of v5, the other to pin 3 of v6. Get the tubes sides/switch interchanged, results will not be pretty.
Keep in mind that pentode mode gives the 80 watts out rating, and triode will probably drop that to 1/2 or more of the above amount. Even though I have not measured it, I would expect the quad of kt90's could push at least a buck and a quarter, maybe more, if you are using this as a high side amp in a biamp setup where the bass is not going to saturate the output transformer. A pair of the kt90's will probably beat the stock quad of el34's in output.
That's pretty much it for bias metering, independent bias, and pentode/triode mode.
On with a few other changes that were made to this amp, and closing thoughts.
The two large black objects in the upper right of one of the photos are angela 47mfd 630v polypropylenes, they replaced c6a and c6b on the schematic, filtering for the 6ccg7's plates. Those are bypassed on the turret board with .01 solens. The HV caps are bypassed with a .01 solen right after the rectifiers.
The third, smaller large black object is the solen 22mfd for the screen bypass, and you can barely see the pentode/triode switch under the left lower corner.
The coax from the phono input was directly connected to pin 7 of v1, with none of the other associated volume control or circuitry hooked to that pin.
There is however a new grid leak resistor of 100k from pin 7 to ground.
My main ground point is the - of the lower hv cap, and pretty much everything stars back to that point. Even a stock altec has some ground loop problems, this one does not - no hum, dead quiet.
This amp is also very quick, detailed, without really getting at all harsh. That said, some people do not like solens, for me in this amp, and my speakers, they are great. For some, with say nose bleed horn speakers, you might want to look elsewhere, but I have not yet heard anything that would make me want to go to some boutique caps in this amp. russian pio's or teflon caps might be fun to try however.
Only mods considered having just reviewed my work from 5 years ago, would be for less than $5 cost to upgrade the two hv rectifiers and the bias rectifier to vishay hexfreds. I also have some 600mfd/500v polypropylene caps, that one may end up replacing the two electrolytic caps currently used for the high voltage filtering.
Since the output tubes are now all split up regarding bias and metering, I am tempted to do away with the 100 ohm grid stop resistors, and attach the coupling caps and the 100k bias resistors directly to the grid of the output tubes. I really do not feel they are necessary any longer, and could be very slightly degrading the signal path. Could you hear the difference? Probably not.....
I stuck with 1/2 watt 10 ohm cathode resistors for a reason, with a tube fart, they will normally blow pretty quickly, I have already replaced a couple. Mills 3 watt 10 ohm would probably be the resistor of choice here, but I would rather not stress the output trannies when you loose a tube, or have a flashover, and let the cathode resistor act as a fuse.
Plus should probably bring the sister of this amp up to the same rev/resto mod level. Right now, this is probably going to be my test bench mule for matching an such, in that this amp has been superceeded by 5 antique sound lab monoblocks in the main system, two of which have only arrived in the last couple of weeks, a pair of asl hurricane dt's. Next project might be a source/speaker switcher so I can switch up to 7.1 home theater from glass to sand with a single switch. Probaby going after a sunfire or parasound multichannel amp, so can still live in the summer heat wise, and cut back on the electric when not doing serious listening. The 5 asl's are pulling down around 2kw of power, and that is not even up on the pipe! I turned off the heat yesterday, and pulled the living room temp up about 2 degrees in an hour!
Are these mods for everyone? Probably not. If I had known how much these altecs would have shot up in the past 5 years, I probably would have rebuilt them more toward stock. However, since I really do not plan on selling the pair I have, that is probably really moot point. The pair I bought were very rough anyway, and in bad need of help. Bottom line, I would rather have the improved performance over the possible future sales price, and if someone that knew altecs heard this one, I believe it might be worth a lot more than a more attractive stock unit.
I did just aquire a scott 272 that is hitting the bench soon, and it will retain most of its original looks, except changing all the old paper caps to illinois poypropylenes.
On the tube rectifiers vrs ss rectifiers, we could argue that point until the cows come home, and it has been argued enough already. However in this configuration, I believe ss is the better way to go. I have probably 20vdc-30vdc more plate voltage available, the load of the 5u4's filaments are gone, and with the metering, individual biasing, and pentode/triode features I believe the higher hv allows more tube options to be used a little better. I'm not sure this amp will ever go back to el34's, I like 6550's, kt88's, kt90's, and this thing will run just fine on even one pair of those. You should hear this thing on a pair of original genelex kt88's! Would love to try a quad of the reissues.
Sorry for a gross hijack of this thread, hopefully some of the madness spewed here will help someone out. Do not attempt any of these mods if you are not qualified to work on HV electronic gear, or at least keep one hand in your pocket!
Good luck, and Good listening,
Brian
Pentode - Triode switching is pretty simple. You need a dpdt switch, and a couple of 47 ohm - 100 ohm 2 watt resistors. Plus, I added in a 22ufd 630v polypropylene solen cap.
Back on the schematic, look at v5 and v6 pin 4, the screen connection. Also note that on each side, pin 4 also connects to its partner parallel tube.
What I see here stock, I do not like, but maybe it is just me. R20 that is feeding the screens, the 250 ohm 5 watt resistor drops down the plate vcc, but is not bypassed. So, on the tube side of r20, I added a 22mfd solen cap to reduce the impedence seen by the screens, and hopefully reduce them moving around as much in voltage.
With the dpdt switch, the two center terminals will hook to pin 4 of each of the sides of the output stage, in place of the connections to r20. So one connection from one center to v5 pin 4, and one connection from the other center pin to v6 pin 4, with no remaining connection to r20. Pick an end of the toggle switch, and hook the side of r20 that had been connected to the pin 4's to both sides of the connections on that end. And again, that end of r20 or the switch connected to the 22mfd solen that the other side of connects to ground (of the solen). Now we have pentode mode again, when the switch is in the correct position to reconnect the pin 4's to r20/solen.
For triode mode, now hook one connection each of the other end of the switch through a 47 ohm 2 watt resistor back to its corresponding output tube pairs plate (pin 3). Be very certain that the side of the switch that goes to pin 4 for example of v5, is the same side on the other end that has the resistor going to pin 3 of v5. You end up with two resistors, each going to one side of the same end of the dpdt switch, both matching the output pairs pin 4 side of the switch, with one resistor to pin 3 of v5, the other to pin 3 of v6. Get the tubes sides/switch interchanged, results will not be pretty.
Keep in mind that pentode mode gives the 80 watts out rating, and triode will probably drop that to 1/2 or more of the above amount. Even though I have not measured it, I would expect the quad of kt90's could push at least a buck and a quarter, maybe more, if you are using this as a high side amp in a biamp setup where the bass is not going to saturate the output transformer. A pair of the kt90's will probably beat the stock quad of el34's in output.
That's pretty much it for bias metering, independent bias, and pentode/triode mode.
On with a few other changes that were made to this amp, and closing thoughts.
The two large black objects in the upper right of one of the photos are angela 47mfd 630v polypropylenes, they replaced c6a and c6b on the schematic, filtering for the 6ccg7's plates. Those are bypassed on the turret board with .01 solens. The HV caps are bypassed with a .01 solen right after the rectifiers.
The third, smaller large black object is the solen 22mfd for the screen bypass, and you can barely see the pentode/triode switch under the left lower corner.
The coax from the phono input was directly connected to pin 7 of v1, with none of the other associated volume control or circuitry hooked to that pin.
There is however a new grid leak resistor of 100k from pin 7 to ground.
My main ground point is the - of the lower hv cap, and pretty much everything stars back to that point. Even a stock altec has some ground loop problems, this one does not - no hum, dead quiet.
This amp is also very quick, detailed, without really getting at all harsh. That said, some people do not like solens, for me in this amp, and my speakers, they are great. For some, with say nose bleed horn speakers, you might want to look elsewhere, but I have not yet heard anything that would make me want to go to some boutique caps in this amp. russian pio's or teflon caps might be fun to try however.
Only mods considered having just reviewed my work from 5 years ago, would be for less than $5 cost to upgrade the two hv rectifiers and the bias rectifier to vishay hexfreds. I also have some 600mfd/500v polypropylene caps, that one may end up replacing the two electrolytic caps currently used for the high voltage filtering.
Since the output tubes are now all split up regarding bias and metering, I am tempted to do away with the 100 ohm grid stop resistors, and attach the coupling caps and the 100k bias resistors directly to the grid of the output tubes. I really do not feel they are necessary any longer, and could be very slightly degrading the signal path. Could you hear the difference? Probably not.....
I stuck with 1/2 watt 10 ohm cathode resistors for a reason, with a tube fart, they will normally blow pretty quickly, I have already replaced a couple. Mills 3 watt 10 ohm would probably be the resistor of choice here, but I would rather not stress the output trannies when you loose a tube, or have a flashover, and let the cathode resistor act as a fuse.
Plus should probably bring the sister of this amp up to the same rev/resto mod level. Right now, this is probably going to be my test bench mule for matching an such, in that this amp has been superceeded by 5 antique sound lab monoblocks in the main system, two of which have only arrived in the last couple of weeks, a pair of asl hurricane dt's. Next project might be a source/speaker switcher so I can switch up to 7.1 home theater from glass to sand with a single switch. Probaby going after a sunfire or parasound multichannel amp, so can still live in the summer heat wise, and cut back on the electric when not doing serious listening. The 5 asl's are pulling down around 2kw of power, and that is not even up on the pipe! I turned off the heat yesterday, and pulled the living room temp up about 2 degrees in an hour!
Are these mods for everyone? Probably not. If I had known how much these altecs would have shot up in the past 5 years, I probably would have rebuilt them more toward stock. However, since I really do not plan on selling the pair I have, that is probably really moot point. The pair I bought were very rough anyway, and in bad need of help. Bottom line, I would rather have the improved performance over the possible future sales price, and if someone that knew altecs heard this one, I believe it might be worth a lot more than a more attractive stock unit.
I did just aquire a scott 272 that is hitting the bench soon, and it will retain most of its original looks, except changing all the old paper caps to illinois poypropylenes.
On the tube rectifiers vrs ss rectifiers, we could argue that point until the cows come home, and it has been argued enough already. However in this configuration, I believe ss is the better way to go. I have probably 20vdc-30vdc more plate voltage available, the load of the 5u4's filaments are gone, and with the metering, individual biasing, and pentode/triode features I believe the higher hv allows more tube options to be used a little better. I'm not sure this amp will ever go back to el34's, I like 6550's, kt88's, kt90's, and this thing will run just fine on even one pair of those. You should hear this thing on a pair of original genelex kt88's! Would love to try a quad of the reissues.
Sorry for a gross hijack of this thread, hopefully some of the madness spewed here will help someone out. Do not attempt any of these mods if you are not qualified to work on HV electronic gear, or at least keep one hand in your pocket!
Good luck, and Good listening,
Brian
Last edited:
bharper
tube junkie...
Altec 1569A manual
I put a pdf version of the 1569a manual up for download, where the schematic on the first page of the thread showed 390 volts for the plate supply, this manual shows 440vdc... hmmm, quite a difference!
altec 1569a manual
here is just the schematic
altec 1569a schematic
I put a pdf version of the 1569a manual up for download, where the schematic on the first page of the thread showed 390 volts for the plate supply, this manual shows 440vdc... hmmm, quite a difference!
altec 1569a manual
here is just the schematic
altec 1569a schematic
Last edited:
I didn't view it as a hijack. All of the knowledge here is greatly appreciated.:thmbsp: