Good idea. I never heard it before pulling it. It all does look a little big so I should have figured it would be heavy on the mids. Do you know what level of efficiency I should be looking for? Maybe ~95db/watt + or so?
More Fun With Magnavox: The 9300 Series
- Thread starter dcgillespie
- Start date
eickmewg
Well-Known Member
A few years back, AKer kidmoe put his Gillespie-inspired Magnavox 9304 (as I recall?) up on Bartertown for a very reasonable price. I was interested in a 6BQ5/EL84 amp to complement my EL34/6CA7 Yaqin and my Latino VTA ST 120 KT88 amp. It came to me with some right-channel hum. I ended up putting a fair amount of time and effort into this problem: relocating the power cord back to its original position, twisted the AC lines to the power switch, and playing with wiring dress and grounding options. It is now virtually silent and is running Russian 6P12P-EV power tubes, 6EU7 preamp tubes, and with a JJ GZ34 rectifier. My understanding is that this may be the very first 9300-series amp that Dave worked with kidmoe on and got the EFB treatment. So this is likely the amp that got this whole thread rolling way back when!
The purpose for this post is simply to say what a wonderful sounding amp this is with my Polk SRS SDA 2.3 speakers! There is something about the liquid and detailed sound this amp puts out. Excellent sound staging too. In some respects, it even bests the VTA ST 120, which is no slouch. Those 13 watts can be magic! Thanks again to both Dave and kidmoe for their interest in these amps.
The purpose for this post is simply to say what a wonderful sounding amp this is with my Polk SRS SDA 2.3 speakers! There is something about the liquid and detailed sound this amp puts out. Excellent sound staging too. In some respects, it even bests the VTA ST 120, which is no slouch. Those 13 watts can be magic! Thanks again to both Dave and kidmoe for their interest in these amps.
Does anybody (other than DG) have experience going from the original Magnavox OPT to the Z565 OPT? I bought the Dynaco clone OPTs a couple of years ago , but liked the amp as is enough that I couldn't justify taking it out of service to install new iron.
My only minor quibbles with the amp as it is would be
1) upper-frequency extension is a tad rolled off. A Spectral it ain't
2) bass is a little 'bloomy'
Neither of these are very objectionable and actually somewhat pleasing if I'm not being too critical
Now the amp has developed some minor problem that forced me to take it out of service and open it up. Also, its winter and gets dark at 4 pm, so it's indoor project time. Now would be the time to do the swap.
I know that there are sound technical reasons for the Dyna iron being better. I'm looking for a more subjective, listening-based opinion.
Anybody?
My only minor quibbles with the amp as it is would be
1) upper-frequency extension is a tad rolled off. A Spectral it ain't
2) bass is a little 'bloomy'
Neither of these are very objectionable and actually somewhat pleasing if I'm not being too critical
Now the amp has developed some minor problem that forced me to take it out of service and open it up. Also, its winter and gets dark at 4 pm, so it's indoor project time. Now would be the time to do the swap.
I know that there are sound technical reasons for the Dyna iron being better. I'm looking for a more subjective, listening-based opinion.
Anybody?
rjpjnk
Active Member
I switched to the Z565 clones (dynakitparts) but I can't say if the sound improved subjectively because it's too hard to remember what it sounded like with the old OPTs. They both sounded great to me. Yes, the bass is pleasantly lose, but I don't think the OPTs had any impact on that. I can say only that it measures better on the bench with the Z565s. Btw, it's not just the transformers that get swapped for this mod. There are a total of 20 resistors and capacitors that need to be changed to go from the original DG design to the version with the updated OPTs. Just get the schematic for each and compare. Feedback loop, step filter, cathode bypass, and grid resistors are different.
I've rebuilt one with stock iron, and I have mine with the Edcor iron. If I'm being honest about it, I don't know that the Edcor is significantly better to listen to. Its got a little more bass but its nothing you couldn't fill in with a small amount of subwoofer.
If I had a second one of my own to do, I'd probably just leave the stock iron. It honestly sounds better than it seems like it should.
Might be another story if I already owned the iron, but more likely I'd be tempted to just build something around those transformers on a fresh chassis.
Might be another story if I already owned the iron, but more likely I'd be tempted to just build something around those transformers on a fresh chassis.
Hmmm...
Morrism
Active Member
What about the 2.7K and 100uF? They were changed to 1.5K and 470uF in the new version. Do they need to be changed? I have the parts for the other 20 items than need changing.
rjpjnk
Active Member
Yes, definitely. That 1.5K is an essential part of the feedback circuit designed to support the new transformers. I thought I included those when I counted 20? In any event, change exactly what is in the schematic. I think the latest was this one:
http://audiokarma.org/forums/index....-the-9300-series.687735/page-20#post-10220075
Kkbubar
Member
Got mine finished up and did the power up this morning Had a few craftsmanship related issues. Now I've got a oscillation problems even with inputs grounded it takes off on it's own. I'm not sure where to start to sort this out. Adding some pictures to aid in the troubleshooting This is a scratch build w/ Edcor OTs w/ 8 ohm secondaries and a scrounged Motorola PT from a three channel console amp. Tubes are 12AX7 & EL84, and NFB is 2.7K w/ 680pf in parallel and 18pf to the top EL84 anode. On the phase inverter it's 3.3K w/ 18pf. Everything power wise is good.
Attachments
Last edited:
Kkbubar
Member
I went back and reread Dave’s dissertation about the HF instability in this amp and was able to get the oscillating under control. To get there first I had to pull the buffer tube and jumper the input around it. With the buffer in place with >0.1 volt from the frequency generator the goes into a grossly overdriven state. I’m using 12AX7 tubes for my buffer and inverter stages. I’m going back to read the notes about the buffer, if someone has input about this it would be greatly appreciated.
Kkbubar
Member
I’m following your drawing that I’ve attached to this post. But I seem to remember Dave stating it should be 47K.
Kkbubar
Member
This is it I’ll give it a try and report back.
I think Dave's comment was for the buffer added to the amp circuit with stock transformers. Increasing that resistor won't hurt anything though. It will knock down some of the HF response, which may be a help for stability.
Ideally you need a scope and a signal generator for this since your setup is slightly different from what has already been done.
Ideally you need a scope and a signal generator for this since your setup is slightly different from what has already been done.
Kkbubar
Member
Gadget you are correct changing that resister from 10 to 47K had no real effect. What I found was if I removed the 680pf cap from the NFB network it goes away and everything is stable. My current configuration is 3.9K ohm and 680pf in parallel connected to the output of the OT and going to pin 3 of the phase inverter tube. I have also tried a 2.7K in this network with no change.And from that point a 20pf cap going to the brown transformer input lead. I do have a scope and a signal generator. I’m feeding a 1Khz signal in at .3 vac.
I think I’m going to put the 10K back in,I’m a fan of if it doesn’t work put it back the way it was. Maybe stick a pot in the NFB network and see if I can find a value for the resistor. I’m very much on the front end of learning about this it’s nice to have people who are willing to help.
I think I’m going to put the 10K back in,I’m a fan of if it doesn’t work put it back the way it was. Maybe stick a pot in the NFB network and see if I can find a value for the resistor. I’m very much on the front end of learning about this it’s nice to have people who are willing to help.
What does a 1khz square wave look like? Output should be somewhere around 2-3v p-p.
For the designed feedback level, 2.7k from the 8 ohm should be the right value. Its supposed to be 12db. There are ways to calculate it, I prefer to just measure output with the feedback disconnected, then connect it and measure the difference. A bit of math will turn the voltage difference into db.
The phase cap value may be wrong for your particular transformers. Thats just what I happened to end up with for a decent compromise of stability and frequency response. Feeback is something of a balancing act, the wider the frequency response, the more likely it is to become unstable.
For the designed feedback level, 2.7k from the 8 ohm should be the right value. Its supposed to be 12db. There are ways to calculate it, I prefer to just measure output with the feedback disconnected, then connect it and measure the difference. A bit of math will turn the voltage difference into db.
The phase cap value may be wrong for your particular transformers. Thats just what I happened to end up with for a decent compromise of stability and frequency response. Feeback is something of a balancing act, the wider the frequency response, the more likely it is to become unstable.
Similar threads
