Oops... missed that.. my bad. If that's the case, I would suspect you have a shorted winding... what did you do with the 5-volt winding? I even did a ARC ST-70C3 upgrade on one of the potted versions and it did not run as hot as yours... plus it did not have mechanical coupling as you described. A shorted winding (or even a partial shorted winding) would account for the high temps and the the mechanical issue.
Regards, KM
Regards, KM
Re: Hot Transformer
Normally you should do that in reverse, now you'll want to walk away for at least 8 hrs, but measure the ambient temperature now and again when you measure the cold resistance - if they are not the same you can correct the Rcold for the temperature difference, or if the difference is a degree C or less you can just ignore it.
You could have a shorted winding, but I think that is unlikely - I would expect much higher temperatures and magic smoke a lot sooner than 8 hrs.
I bet it rises gradually over the course of several hours before achieving equilibrium.
How large are the filter caps? Increased ripple current due to the low forward resistance of your dampers combined with several hundred uF of input capacitance might very well account for the increased heating you are observing.
Was the cage on during operation - if so you really don't have an issue because the internal ambient in that amp is not many degrees lower after some hours of operation.
FWIW I recommend small quiet fans with caged vintage dyna ST-70 and MKIII in particular.
opc said:
Normally you should do that in reverse, now you'll want to walk away for at least 8 hrs, but measure the ambient temperature now and again when you measure the cold resistance - if they are not the same you can correct the Rcold for the temperature difference, or if the difference is a degree C or less you can just ignore it.
You could have a shorted winding, but I think that is unlikely - I would expect much higher temperatures and magic smoke a lot sooner than 8 hrs.
I bet it rises gradually over the course of several hours before achieving equilibrium.
How large are the filter caps? Increased ripple current due to the low forward resistance of your dampers combined with several hundred uF of input capacitance might very well account for the increased heating you are observing.
Was the cage on during operation - if so you really don't have an issue because the internal ambient in that amp is not many degrees lower after some hours of operation.
FWIW I recommend small quiet fans with caged vintage dyna ST-70 and MKIII in particular.
Hot or Not Transformer Competition
Well, I've had a fan blasting the thing for the past few hours, and it's back down to room temp. So after taking a second resistance measurement and adding the ambient temp I'm getting 80.1 degrees C for the copper.
I guess that seems reasonably accurate since the outside lamination temp is 74-75C, and the potting probably helps couple heat reasonably well.
As for the remaining 5V leads, I put heat-shrink tubing over the entire wire, then another bit at the end and pinched it together. I just double checked them now, and they're definitely not shorted.
The rest of the cloth lead wires are really showing their age though, and although I covered each one with heat-shrink, I can't really be sure that they're not shorted somewhere inside the casing. the bottom of the transformer won't come off because of the potting compound, so I can't get in there to take a better look.
Well the fact that it's running hotter than other people's, and the mechanical vibration problem both point me in the new transformer direction.
Anyone have any suggestions? I wanted to use an Antek toroidal AN-4T360 but it doesn't have the 55V bias tap, and I don't have room for yet another 50V transformer under the chassis. Is there a clever way around the bias tap that I haven't thought of?
Any suggestions would be appreciated!
Well, I've had a fan blasting the thing for the past few hours, and it's back down to room temp. So after taking a second resistance measurement and adding the ambient temp I'm getting 80.1 degrees C for the copper.
I guess that seems reasonably accurate since the outside lamination temp is 74-75C, and the potting probably helps couple heat reasonably well.
As for the remaining 5V leads, I put heat-shrink tubing over the entire wire, then another bit at the end and pinched it together. I just double checked them now, and they're definitely not shorted.
The rest of the cloth lead wires are really showing their age though, and although I covered each one with heat-shrink, I can't really be sure that they're not shorted somewhere inside the casing. the bottom of the transformer won't come off because of the potting compound, so I can't get in there to take a better look.
Well the fact that it's running hotter than other people's, and the mechanical vibration problem both point me in the new transformer direction.
Anyone have any suggestions? I wanted to use an Antek toroidal AN-4T360 but it doesn't have the 55V bias tap, and I don't have room for yet another 50V transformer under the chassis. Is there a clever way around the bias tap that I haven't thought of?
Any suggestions would be appreciated!
Hi Kevin,
Sorry about that, I was writing my last post when yours went up. You're absolutely right, I was a little hasty with the resistance measurements. I just checked again and it's about 0.8 ohms lower than it was an hour ago, so it's still cooling. I'll leave it off all night and check it when it's cold in the morning.
The first cap is 68uF, followed by the choke and three more 120uF caps so:
6DE4 - 68uF - choke -120uF - 6K8 - 120uF - 22K - 120uF.
I don't think that's a significantly high enough capacitance to really cause a problem, but I guess it could be with the lower Rforward of the dampers.
I have to original cage, but I never use it. It would break my heart to hide away those tubes, and I don't have any kids or animals, so it's sufficiently safe.
Sorry about that, I was writing my last post when yours went up. You're absolutely right, I was a little hasty with the resistance measurements. I just checked again and it's about 0.8 ohms lower than it was an hour ago, so it's still cooling. I'll leave it off all night and check it when it's cold in the morning.
The first cap is 68uF, followed by the choke and three more 120uF caps so:
6DE4 - 68uF - choke -120uF - 6K8 - 120uF - 22K - 120uF.
I don't think that's a significantly high enough capacitance to really cause a problem, but I guess it could be with the lower Rforward of the dampers.
I have to original cage, but I never use it. It would break my heart to hide away those tubes, and I don't have any kids or animals, so it's sufficiently safe.
Hi Owen,
Actually it is pretty likely that the 68uF is contributing in a big way to the winding heating as the pk current could be as much as 3X larger than with the stock cap alone, add in the probable lower forward resistance of the dampers and that compounds the situation. The lower Rcold is going to result in an overall higher winding temperature once you get there.. I bet it is somewhere in the vicinity of 90 degrees C once it has cooled and you have the new reading.
Frankly I am not sure that most people take the trouble to measure the transformer as carefully as you did. My recollection with the small power transformer is that it ran quite hot in the stock amp - IIRC I measured in excess of 70 degrees C using your technique after about 8hrs of operation.
Dynakits, Triode Electronics, and Tubezone all had suitable replacement transformers last time I checked.
Actually it is pretty likely that the 68uF is contributing in a big way to the winding heating as the pk current could be as much as 3X larger than with the stock cap alone, add in the probable lower forward resistance of the dampers and that compounds the situation. The lower Rcold is going to result in an overall higher winding temperature once you get there.. I bet it is somewhere in the vicinity of 90 degrees C once it has cooled and you have the new reading.
Frankly I am not sure that most people take the trouble to measure the transformer as carefully as you did. My recollection with the small power transformer is that it ran quite hot in the stock amp - IIRC I measured in excess of 70 degrees C using your technique after about 8hrs of operation.
Dynakits, Triode Electronics, and Tubezone all had suitable replacement transformers last time I checked.
@Kevinkr
Well, your prediction was pretty much spot on. I took a measurement last night after the amp had been on for 9 hours, and then I took another measurement this afternoon when it had been off for 9 hours. The new value I got for the copper temp was 87.5 C and I think that's about as accurate of a reading as I'm going to get.
As for the capacitor, I'm honestly surprised. I've seen people out there running silicon diodes and 400uF caps without any reported problems, so I thought I was being pretty conservative with my 68uF value and the tube rectifiers.
If it is the size of the first cap, would choke coupling the damper diodes to the first 68uF cap ease the transformers burden by reducing the peak current value? If so, I could get my hands on a choke and give it a shot.
Otherwise, I think I'm just going to replace the thing with a toroidal transformer with a much higher VA rating. I'd really appreciate any suggestions on what to do about the bias tap. Could I use a resistor divider between one of the 360V lines and the CT to get close to 55V? How much current do those bias supplies eat up?
Any suggestions would be appreciated, and a big thanks to kevinkr for the transformer temp measurement suggestion!
Cheers,
Owen
Well, your prediction was pretty much spot on. I took a measurement last night after the amp had been on for 9 hours, and then I took another measurement this afternoon when it had been off for 9 hours. The new value I got for the copper temp was 87.5 C and I think that's about as accurate of a reading as I'm going to get.
As for the capacitor, I'm honestly surprised. I've seen people out there running silicon diodes and 400uF caps without any reported problems, so I thought I was being pretty conservative with my 68uF value and the tube rectifiers.
If it is the size of the first cap, would choke coupling the damper diodes to the first 68uF cap ease the transformers burden by reducing the peak current value? If so, I could get my hands on a choke and give it a shot.
Otherwise, I think I'm just going to replace the thing with a toroidal transformer with a much higher VA rating. I'd really appreciate any suggestions on what to do about the bias tap. Could I use a resistor divider between one of the 360V lines and the CT to get close to 55V? How much current do those bias supplies eat up?
Any suggestions would be appreciated, and a big thanks to kevinkr for the transformer temp measurement suggestion!
Cheers,
Owen
Attachments
hey-Hey!!!,
There are a few quantities all tied together in this power supply. I suggest employing PSUDii to examine their relationship. For a cap-input filter, and a given DC output, the output ripple, input RMS current, and DC output voltage are all tied to the size of the input cap.
To experiment, hold all but one constant and examine the change in the other outputs.
cheers,
Douglas
There are a few quantities all tied together in this power supply. I suggest employing PSUDii to examine their relationship. For a cap-input filter, and a given DC output, the output ripple, input RMS current, and DC output voltage are all tied to the size of the input cap.
To experiment, hold all but one constant and examine the change in the other outputs.
cheers,
Douglas
Bandersnatch said:
I'm with Doug "forensically" 🙂 on this one, and I don't really recommend toroid power transformers for tube amplifiers. The leakage inductance in a standard EI transformer is actually your friend here, reducing the effects of line borne noise and transients on the performance of your amplifier. Big toroids have big inrush currents which can create other issues with fuse blowing, etc. Any of the transformers I mentioned previously will drop in without difficulty. Plus it is quite possible that the current transformer will operate forever as is, or with just minor changes to the input circuitry.
hey-Hey!!!,
For a given set of conditions, you'll find RMS current delivered to the filter increasing with the size of the first filter cap( even if the delivered power does not increase ). The filter can be tweaked a bit, drop the first cap, increase the second, increase the inductance of the choke a bit...
cheers,
Douglas
( it is a C-L-C pi filter isn't it? )
For a given set of conditions, you'll find RMS current delivered to the filter increasing with the size of the first filter cap( even if the delivered power does not increase ). The filter can be tweaked a bit, drop the first cap, increase the second, increase the inductance of the choke a bit...
cheers,
Douglas
( it is a C-L-C pi filter isn't it? )
Hey Guys,
@Bandersnatch
I remember using that program some time ago, but didn't even think of it until you suggested it. I ran a few simulations, and things don't look too good for the poor transformer.
I found that varying the value of the capacitor doesn't seem to help very much with regards to the current peaks in the transformer. I think the dominant values are the low Rforward of the damper diodes, and the low ESR of the excellent Panasonic capacitor. Using a smaller cap reduces the duration (width) of the spike slightly, but they're still quite high at nearly 900mA peak.
The biggest improvement came from choke coupling the diodes to the first cap in the power supply, so I think I might give that a try. It reduced the current spikes to 200mA peak, which I think will really help.
I'll include a screen-shot so you can see what I did. Any comments are welcome.
@kevinkr
I had never heard that toroidal transformers weren't as good as EI types for tube circuits. I have a pair of 1500VA's in my Aleph 2's, and inrush current is a problem with those, but I've never had troubles with anything below 500VA. I guess I'll try dropping in a choke, and reducing that first cap to 42uF, and maybe the transformer in there now will settle down.
Another thing I noticed while I was poking around was the unusually high AC voltage on the unloaded transformer. It runs 400-0-400 under no load, and I'm getting nearly 450VDC after the choke. I've adjusted the bias on the output tubes down to 45 mA each to compensate for this, but that voltage seems a little high to me. The line voltage here is high (around 125VAC at the outlet) so that might be contributing, considering the transformer was designed for 110-115VAC. The damper diodes also have much lower voltage drop than the usual FW crap, so that makes the DC voltage even higher.
@scottw
The board in the picture is one I just finished designing and building. I'm working on a second revision that only uses one standard FW rectifier in place of the two damper diodes, and makes a few fixes to some of the little things I didn't like with the first revision like the placement of the bias adjustment and a few other things.
I didn't like the huge boards that sit above the chassis, and I didn't like the small boards that sit below, so I measured out all the holes and tube locations and made a board that covers the whole bottom of the chassis and has PCB mount tube sockets.
I think it looks much cleaner than the usual spaghetti mess you see under ST-70's, and it allows better parts placement and layout since you're not confined to the tiny space of the original board.
If you're interested drop me a note and I can send you some details on it.
Cheers,
Owen
@Bandersnatch
I remember using that program some time ago, but didn't even think of it until you suggested it. I ran a few simulations, and things don't look too good for the poor transformer.
I found that varying the value of the capacitor doesn't seem to help very much with regards to the current peaks in the transformer. I think the dominant values are the low Rforward of the damper diodes, and the low ESR of the excellent Panasonic capacitor. Using a smaller cap reduces the duration (width) of the spike slightly, but they're still quite high at nearly 900mA peak.
The biggest improvement came from choke coupling the diodes to the first cap in the power supply, so I think I might give that a try. It reduced the current spikes to 200mA peak, which I think will really help.
I'll include a screen-shot so you can see what I did. Any comments are welcome.
@kevinkr
I had never heard that toroidal transformers weren't as good as EI types for tube circuits. I have a pair of 1500VA's in my Aleph 2's, and inrush current is a problem with those, but I've never had troubles with anything below 500VA. I guess I'll try dropping in a choke, and reducing that first cap to 42uF, and maybe the transformer in there now will settle down.
Another thing I noticed while I was poking around was the unusually high AC voltage on the unloaded transformer. It runs 400-0-400 under no load, and I'm getting nearly 450VDC after the choke. I've adjusted the bias on the output tubes down to 45 mA each to compensate for this, but that voltage seems a little high to me. The line voltage here is high (around 125VAC at the outlet) so that might be contributing, considering the transformer was designed for 110-115VAC. The damper diodes also have much lower voltage drop than the usual FW crap, so that makes the DC voltage even higher.
@scottw
The board in the picture is one I just finished designing and building. I'm working on a second revision that only uses one standard FW rectifier in place of the two damper diodes, and makes a few fixes to some of the little things I didn't like with the first revision like the placement of the bias adjustment and a few other things.
I didn't like the huge boards that sit above the chassis, and I didn't like the small boards that sit below, so I measured out all the holes and tube locations and made a board that covers the whole bottom of the chassis and has PCB mount tube sockets.
I think it looks much cleaner than the usual spaghetti mess you see under ST-70's, and it allows better parts placement and layout since you're not confined to the tiny space of the original board.
If you're interested drop me a note and I can send you some details on it.
Cheers,
Owen
Attachments
Yes, nice looking board. I hadn't recalled ever seeing an after-market board that covered that much chassis real estate and had the octals mounted to it.
Maybe some of those high voltage SIC schottky's?
If I had a "real" ST70.....I ended up with a Series II and I don't think the mounting holes match up. Dang it. I'm stuck with the series II front end, 6gh8/6u8/7687, a while longer.
Thanks,
Scott
Maybe some of those high voltage SIC schottky's?
If I had a "real" ST70.....I ended up with a Series II and I don't think the mounting holes match up. Dang it. I'm stuck with the series II front end, 6gh8/6u8/7687, a while longer.
Thanks,
Scott
I finally got around to putting the Mapletree board into my Dynaco. It sounds much better than other boards I have used in the past.
Hope no one minds me jumping in here a little off topic, but I came across this thread and had a great time reading about the different experiences and opinions.....really makes me regret selling my two old st-70's I had a few years back. They were basically stock, but they were my first "real" tubes, minus the McIntosh Ma230 that I sold to get the Dynacos. (ugh......yeah....probably not the best financial decision I ever made....but, cest la vie).....I just had to use my CJ PV-1....which i DO still have.....and love, maybe I'm just senti"mental".
I had always wanted to do the Audio Research driver board mod(my memory is a bit faded....was it called the ARC-90 maybe????) to the 70's, and was curious if anyone here had any experience with this driver board mod.
Anyone doing active crossovers with the 70's or any of their "derivatives"? Ive always liked the tuning capabilities, and the lack of power robbing passive components.....
FWIW, I really liked the midrange of the the "stock" 70's and had originally planned on using them in an active three way setup, one for the mids and the other for highs, maybe *gasp* solid state for the lows. At the time I had some Infinity Q5's....again my first "real" speakers....which I still "kinda have". Those are a different story now though....
I am again being drawn back to tubes with a different el34 based project(see my "Knight" thread if interested), and have considered getting another 70 or three....lol.
any and all opinions welcome....
thanks again,
Nate
I had always wanted to do the Audio Research driver board mod(my memory is a bit faded....was it called the ARC-90 maybe????) to the 70's, and was curious if anyone here had any experience with this driver board mod.
Anyone doing active crossovers with the 70's or any of their "derivatives"? Ive always liked the tuning capabilities, and the lack of power robbing passive components.....
FWIW, I really liked the midrange of the the "stock" 70's and had originally planned on using them in an active three way setup, one for the mids and the other for highs, maybe *gasp* solid state for the lows. At the time I had some Infinity Q5's....again my first "real" speakers....which I still "kinda have". Those are a different story now though....
I am again being drawn back to tubes with a different el34 based project(see my "Knight" thread if interested), and have considered getting another 70 or three....lol.
any and all opinions welcome....
thanks again,
Nate
ST-70 - not new, but excited anyway
OK, it's my first post and I've just been having a blast reading everyone's comments. I've been piecing together a Dynaco ST-70 for a few weeks now...buying a piece here and another there. So far, I thought I've done a pretty good job.
I still lack the A470 trannies though. What are your thoughts on the A470-S versions. I've been building the amp slowly due to my limited resources. I'm doing it slowly to keep off of my wife's check-book radar. I'm sure most of you can relate - as for the rest of you single guys...enough said there.
I almost forgot, I'm also missing the bottom plat to the chasse - any clues where I could find one? No hurry - just getting excited to hear (or see) this thing work without making it into a smoke generator. 🙂
Brian
OK, it's my first post and I've just been having a blast reading everyone's comments. I've been piecing together a Dynaco ST-70 for a few weeks now...buying a piece here and another there. So far, I thought I've done a pretty good job.
I still lack the A470 trannies though. What are your thoughts on the A470-S versions. I've been building the amp slowly due to my limited resources. I'm doing it slowly to keep off of my wife's check-book radar. I'm sure most of you can relate - as for the rest of you single guys...enough said there.
I almost forgot, I'm also missing the bottom plat to the chasse - any clues where I could find one? No hurry - just getting excited to hear (or see) this thing work without making it into a smoke generator. 🙂
Brian
Re: ST-70 - not new, but excited anyway
See Heyboer TX in Grand Haven MI for OPT's. They have the best prices on A470 and you can also get its 6k6 brother the A420( my preference ) too at the same price. If you want some other exotica new vintage models they've got quite a list, and with a few on my shelf as engineering samples the list can easily expand.
Best Regards,
Douglas
brosenau said:
See Heyboer TX in Grand Haven MI for OPT's. They have the best prices on A470 and you can also get its 6k6 brother the A420( my preference ) too at the same price. If you want some other exotica new vintage models they've got quite a list, and with a few on my shelf as engineering samples the list can easily expand.
Best Regards,
Douglas
A470 vs A420
OK, you have my attention - what's the primary difference between the two models of trannies (no punn intended)? I like this forum because I can tap into other peoples' knowledge base without doing a huge amount of research. You guys definately know your stuff.
On some occasions, I've seen the text go back and forth on some opinion issues - but it's good to keep the blood flowing - just not high enough to take away the fun.
As I plunk around in here for a while, I'll let you all know more about my background and the fun I've had. Until then, happy toobing!
B
OK, you have my attention - what's the primary difference between the two models of trannies (no punn intended)? I like this forum because I can tap into other peoples' knowledge base without doing a huge amount of research. You guys definately know your stuff.
On some occasions, I've seen the text go back and forth on some opinion issues - but it's good to keep the blood flowing - just not high enough to take away the fun.
As I plunk around in here for a while, I'll let you all know more about my background and the fun I've had. Until then, happy toobing!
B
hey-Hey!!!,
The A470 is 4300 Ohms plate to plate, and the A420 is 6600 Ohms plate-to-plate. Same power ratings, and same size core.
cheers,
Douglas
The A470 is 4300 Ohms plate to plate, and the A420 is 6600 Ohms plate-to-plate. Same power ratings, and same size core.
cheers,
Douglas
- Status
- Not open for further replies.