Hey All,
I recently modded my Mark VI's to use Kt 88's. I replaced the four power supply caps with some large (380uf 350V) Vishay computer grade caps. And I replaced the dual power supply cap with a 100uf 100uf 350VDC one. I was told that to get more bias voltage to make the KT's work I needed to change the 18k resistor before the main bias pot to a 1k. And then the 10k between the pot and ground to an 18k. This done I installed a set of JJ KT 88's and NOS 7199's and was disappointed to find serious distortion in the high frequencies. Its an edgy scratchy sound that seems worse as high frequency content increases. It may be a problem that was there in the amps originally, or I may have installed it. The tubes biased great, and the bass is much better. One of the original caps was open, and two of the original 8417's had lost vacuum so they weren't in great shape to begin with. Without signal the amps are dead quiet. So I don't think think its a power supply problem. Since the distortion increases with volume and is there in the precense of signal does it seem a safe bet its in the signal path? I've ordered all new caps for the signal path but haven't recieved them yet. Any ideas?
I recently modded my Mark VI's to use Kt 88's. I replaced the four power supply caps with some large (380uf 350V) Vishay computer grade caps. And I replaced the dual power supply cap with a 100uf 100uf 350VDC one. I was told that to get more bias voltage to make the KT's work I needed to change the 18k resistor before the main bias pot to a 1k. And then the 10k between the pot and ground to an 18k. This done I installed a set of JJ KT 88's and NOS 7199's and was disappointed to find serious distortion in the high frequencies. Its an edgy scratchy sound that seems worse as high frequency content increases. It may be a problem that was there in the amps originally, or I may have installed it. The tubes biased great, and the bass is much better. One of the original caps was open, and two of the original 8417's had lost vacuum so they weren't in great shape to begin with. Without signal the amps are dead quiet. So I don't think think its a power supply problem. Since the distortion increases with volume and is there in the precense of signal does it seem a safe bet its in the signal path? I've ordered all new caps for the signal path but haven't recieved them yet. Any ideas?
Rule 1: When modifying an amp, make sure that it's working in the original condition!
Is the distortion in both amps?
Do check all of the DC voltages (carefully!) to see if there's anything obviously broken. And double check the wiring of the output tube sockets- there are differences between KT88 and 8417! You could also have an oscillation. That could take some work with an oscilloscope to track down.
Is the distortion in both amps?
Do check all of the DC voltages (carefully!) to see if there's anything obviously broken. And double check the wiring of the output tube sockets- there are differences between KT88 and 8417! You could also have an oscillation. That could take some work with an oscilloscope to track down.
When I first got the amps they seemed to play alright. I did hear that they had an odd distortion. It showed up in the mid to higher frequencies in dense spots. Like double tracked backround voices. Individual instruments or lead voice didn't seem to have this distortion. It seemed minor, and people I talked to said it was probably in the recordings. The amps had been given individual bias pots. And had a mix of different manufacturer8417's. For that reason I wanted to replace the tubes. Rather than use 8417's which are no longer in production, I wanted to use soemthing that was. So I went to KT 88's. I checked a tube website for pinouts. The 8417, KT 88 and 6550 all had the same pinouts.
The distortion I'm now hearing is the same as before. But it is much more pronounced. It is in both amps. But it seems about twice as bad in one amp.
I replaced all the electrolytics because they don't age well. As was recommended. I believe with the stronger power supply the distorion is that much more evident. I don't think the modification is the problem so much as it amplifys a problem that was already there. The amps have much more gain than before
I've ordered new coupling caps as well as all the caps in the driver section in the signal path. They all cost less than $20 for good quality orange drop 716's from Tube Depot and shouldn't be difficult to install. I know Dynaco used good quality caps but these are thirty years old now. When I get the signal paths caps replaced I'll check the voltages. I don't have a scope so at that point I would have to take them to a shop.
The distortion I'm now hearing is the same as before. But it is much more pronounced. It is in both amps. But it seems about twice as bad in one amp.
I replaced all the electrolytics because they don't age well. As was recommended. I believe with the stronger power supply the distorion is that much more evident. I don't think the modification is the problem so much as it amplifys a problem that was already there. The amps have much more gain than before
I've ordered new coupling caps as well as all the caps in the driver section in the signal path. They all cost less than $20 for good quality orange drop 716's from Tube Depot and shouldn't be difficult to install. I know Dynaco used good quality caps but these are thirty years old now. When I get the signal paths caps replaced I'll check the voltages. I don't have a scope so at that point I would have to take them to a shop.
hey-Hey!!!,
I'd suspect the 7199 input/phase splitter circuit. Of all the Dyna amps, the Mk.VI put the most load on it, and it was marginal for use in the SCA35. Individual bias pots also double the load on it( with each tube now having its own grid circuit, which are now in parallel ).
For a simple experiment, try reducing the resistors in the split-load inverter to 33k from 47kOhm. If this helps, do consider one of the Mk.III input stage boards to deliver a more capable input/phase splitter stage.
cheers,
Douglas
I'd suspect the 7199 input/phase splitter circuit. Of all the Dyna amps, the Mk.VI put the most load on it, and it was marginal for use in the SCA35. Individual bias pots also double the load on it( with each tube now having its own grid circuit, which are now in parallel ).
For a simple experiment, try reducing the resistors in the split-load inverter to 33k from 47kOhm. If this helps, do consider one of the Mk.III input stage boards to deliver a more capable input/phase splitter stage.
cheers,
Douglas
The JJ KT88 (I've never actually seen one), is unlike any of the previous type KT88. For the ones with which I'm familiar (MO, Genelex), there was an internal connection on pin 6, which shouldn't be used as a tie point. If memory serves, Dynaco did just that, and right at the output tube grid, as a convenient point to mount the grid stopper resistor. You're probably OK if the JJ version has pin 6 cut off. However, make sure that 1 and 8 are tied together- this is a difference between versions of 8417, 6550, and KT88. The JJ datasheet is contradictory about whether the suppressor is internally connected to the cathode, so best to do it at the socket.
Had those bastards some 10 years ago. Due to the fact that the 6550´s needed more drive-voltage than the 8417´s, the 7199 driver probably wasn´t capable. Rebuilt them with ECC88 differential cascode´s followed by ECC182 CF´s. Worked great driving my(at that time) ML CLSII´s. Also rebuilt the OPT´s for the ML´s by parallelling the 0-4ohm with the 4-16ohm windings to get lower secondary RDC.
The problem is that the MarkVIs are unstable as designed and built.
I think it is largely due to layout, not circuit design per se.
If you put a scope on the thing I'd bet you see HF parasitics on the signal.
Consider cleaning up the wiring under the board (routing etc) and finding the parasitics.
That has been my experience with the amp.
I think it is largely due to layout, not circuit design per se.
If you put a scope on the thing I'd bet you see HF parasitics on the signal.
Consider cleaning up the wiring under the board (routing etc) and finding the parasitics.
That has been my experience with the amp.
Hey!
I had the circuit all along! Its a circuit that uses two 12AU7's and point to point wiring instead of the board. I got it from Bill Hardy at Vintage Amplifier Restoration in Marietta Georgia. Four nine pin sockets, four 12AU7's and a handful of passive components and it'll be done. This circuit gets rid of the 7199 and the time delay realy tube. The parts are ordered I should have it done by next weekend.
Kevin
I had the circuit all along! Its a circuit that uses two 12AU7's and point to point wiring instead of the board. I got it from Bill Hardy at Vintage Amplifier Restoration in Marietta Georgia. Four nine pin sockets, four 12AU7's and a handful of passive components and it'll be done. This circuit gets rid of the 7199 and the time delay realy tube. The parts are ordered I should have it done by next weekend.
Kevin
well how did it go?
Im interested since I have 4 mark 6's. two built and two unbuilt kits. I have enough 8417's to go that route or enough kt-88's to go that way.
Sorry I haven't gotten back to you. It went really well. The amps sound great. I'm driving them with an Audio Research SP3A into a pair of original Snell Type A's. I had to modify the bias circuit to make it more stable. I did this by giving each tube its own filter to ground. When all four tubes were going trhough one resistor/cap I couldn't get the bias adjustment on one tube to stop interfering with the others. Also the I had to lower the resistors on the 12AU7's to get them to draw enough Amperage. I have a copy of the preamp/phase splitter diagram and most ot the parts list if you need copies.
The Mark VI is known to be very unstable with parasitics.
Changing the layout is somewhat mandatory if you want the thing to be stable...
I'd take a strong look at a mid freq square wave and look for ringing, especially with some capacitance added to the load... also put the thing up toward clipping with a sinewave, and look for any "blips" at the peak...
_-_-bear
PS. wait! I already said this pretty much...
Changing the layout is somewhat mandatory if you want the thing to be stable...
I'd take a strong look at a mid freq square wave and look for ringing, especially with some capacitance added to the load... also put the thing up toward clipping with a sinewave, and look for any "blips" at the peak...
_-_-bear
PS. wait! I already said this pretty much...
Actually the only original Dynaco parts are the chassis, transformers and meters. The new amp uses 12AU7's as preamp tubes and phase splitters. And KT88's as outputs. The preamp section is now a long-tail phase splitter. I knew the previous owner that had these amps with the original configuration and they never sounded as good a s they do now, not even close.
Are you saying the JJ has an internal connection on pin six, or are you saying that it should but it doesn't?
The datasheet currently published at their website clearly shows G2 tied to K. They show nothing for pin 6, not even a warning not to use it as a tie point.
http://www.jj-electronic.com/pdf/KT88.pdf
I think I've got a JJ KT88 somewhere in my box of junk. I might have to go take a look at it.
- Status
- Not open for further replies.