Another finished Cornet, break-in time?

Braden · « **on:** 30 Sep 2003, 07:48 pm »

I finished the Cornet last week, and have been playing it for a couple hours each day. For the first few hours it didn't sound bad at all, but now I can't be in the same room! There is an upper mid-range glare that's just unbearable. I get a headache after about ten minutes. Last night I put on a cd and was actually relieved. Also, there's very little bass anymore. I'm hoping this is just part of the break-in process, but I haven't heard of similar occurances. The only changes I made in parts was R21 I used a 2R5 instead of a 2R4, and of course using the 0.47 uf capacitor instead of the 1 uf in the output stage. The input impedance of my preamp is 50k ohms. Anyone have any suggestions on what might be happening?

Wayne1 · « **Reply #1 on:** 30 Sep 2003, 09:17 pm »

The parts, mostly the capacitors and tubes, need LOTS of time to break-in.

I would suggest you pick up the Hagerman inverse RIAA and send a signal through it to your Cornet.

I would suggest a minimum of 50 hours constant signal. 150 would be better.

You could also pick up Jim's FryKleaner for the signal source. Its signal tends to work out the power supply caps a bit quicker than just music.

When I received the Trumpet from Jim, it was alredy partially broken in. It still needed 24-50 hours of constant FryKleaner to finish the job.

Braden · « **Reply #2 on:** 30 Sep 2003, 11:20 pm »

Now, I'm not an expert on this, but I have an idea. Why couldn't Mr. Hagerman, or someone else, use both the frykleaner and the inverse RIAA device and record the result onto a cd-r or something and sell it for a small fee? For someone like me, who it only building this one unit, it doesn't make sense for me to spend another $50 CDN to break it in without my ears bleeding.

I've been spending most of the afternoon messing around on my cooledit program. I used Jim's chart for the inverse RIAA curve, and applied it to full frequency sweeps, and various music. I'll make a cd-r of this and let it play into the Cornet for the next few days.

Two questions, should I leave the Cornet on for 3 days straight, or in, say, 15 or 20 hour increments? Also, do I need to connect anything to the outputs?(ie. do I have to listen to it?)

Thanks again!

Wayne1 · « **Reply #3 on:** 1 Oct 2003, 12:43 am »

The FryKleaner and its signal is Mr. Hagerman's property.

Perhaps he should make such a CD available. I will mention it to him at VSAC.

I had no trouble running the Trumpet, Cornet or Bugle 24/7.

Just make sure there is lots of room around the tubes for air flow.

There is no need to listen to it. Just plug the Cornet into your preamp, select the input on your preamp and leave the power amp off.

mgalusha · « **Reply #4 on:** 1 Oct 2003, 12:51 am »

Braden,

One of the functions of the Inverse RIAA circuit is to drop the levels by either 40 or 60dB. If you connect the CDP outputs to the Cornet you will very likely overload the input stage.

You could of course use a voltage divider to acomplish this or record the track for your break in CD at -40 or -60dB.

In looking at your pictures I noticed you used Holco resistors in many places. I have read (in the product catalog) that the Holco's are only rated to 200V and the HV rail in the Cornet runs from 300 to 375 volts. You may want to keep tabs on R1. In your picture you have a 1/4 watt Holco in this location and it sees 330V. I'm not trying to nitpick, just trying to help prevent a possible future problem.

Mike G.

Braden · « **Reply #5 on:** 1 Oct 2003, 02:02 am »

mgalusha,

The cd's levels will be low, I'll see how it works.

As far as the resistors go, I only ordered 1 watt Holco's, but then again, I wouldn't know the visual difference between 1/4 and 1 watt. I thought the Holco's would work better than the Xicon's. I don't mind ordering a few resistors to replace mine if it'll work better and last longer. I used some H2 and some H4P Holco's, depending on what was available. The Mills were 12 watts.

Thanks.

mgalusha · « **Reply #6 on:** 1 Oct 2003, 04:03 am »

Braden,

I suspect the Holco's are quieter than the Xicon's and if so, would probably sound better.

I didn't notice the "P" on the resistor.

Normally the H4's are 1/2 Watt (sorry I wrote 1/4 Watt previously) but the H4P's are 1 Watt. My mistake, the H4P's are rated 500V. They should work very nicely.

I left my Cornet powered up and on the FryKleaner/Inverse RIAA setup for 3 days before I was reasonably happy with the sound. Even so, it has steadily improved since then. What tubes are you using? IME the Cornet can really change it's personality with different tubes. So far I've been happiest with some EI Gold Elite, JJ's and Sovtek 12AX7LPS for the first two stages and a NOS Mullard CV-4003 for the 12AU7.

I am also currently using a NOS RCA 5Y3GT. I've tried a Sovtek and NOS Sylvania and preferred the RCA. To be fair the Sylvania wasn't broken in yet, so it may be fine once it has some time on it.

Hope it comes around for you. It probably will. When we tried some broken in tubes in Wayne's Cornet it was transformed, so break in will change it.

Mike G.

hagtech · « **Reply #7 on:** 1 Oct 2003, 06:18 am »

Braden, you certainly have put some very high quality part into this!

"Glare" is not something you will get from a Cornet. Even raw, it is far from unlistenable. This isn't a break-in issue. My guess is that something is awry. Probably an oscillation. Stock parts fit nicely onto the circuit board and all signal currents are well controlled and routed. With exotic parts, they don't always fit. That might lead to problems.

Maybe it's the 0.047uF caps. They seem to route over everything there in the first two gain sections. Loops of wire, especially near the inputs can couple positive feedback and cause an ultrasonic oscillation.

Any chance you have an oscilloscope handy? Try to move the caps around so that leads don't appear to be coupling. Maybe you can temporarily run with smaller caps? Are both channels the same? Or is one worse?

As far as the FryKleaner and iRIAA, they are by far the cheapest solutions on the market. Hey, I even give you the circuit if you decide to wire it up yourself. The CD doesn't work too well. Signal levels are so low you run out of bits. Consider 20Hz, it'll need to be at -60dBFS. That leaves only 6 bits to produce music.

jh

Braden · « **Reply #8 on:** 1 Oct 2003, 04:41 pm »

mgalusha,

I'm using EH 12AX7's, NOS Mullard 12AU7, and I think NOS ITT 5Y3GT.
I picked my tubes based on what you and the other reviewers suggested, and what parts connexion had.

Jim,

An oscilliscope would be really handy right now. I'll see what I can do as far as shortening the wires, maybe tying the caps down to the board. If that doesn't help, I guess I order new caps. The glare seems to be uniform between channels. It's not overly apperant initially, but upper-mid frequencies, like vocals etc. become fatiguing quickly. After awhile I was getting a headache. It's more bearable with headphones, the speakers I'm using really seem to tell it how it is.

Maybe I should record an album from both the Cornet and my old phono stage to my computer and look at the difference?

Braden

Braden · « **Reply #9 on:** 2 Oct 2003, 06:11 pm »

Okay, I tidied up the wires. I tied back the transformer wires as much as I could. Then, I shortened one end each of the 0.047 hovland caps and stood them on end, then snugged up the remaining wire with nylon tie.

What I came to realize later was that the Cornet is really sensitive to other equipment. Just having it in different spots near my cd player could greatly increase or decrease the amount of audible hum. When I was experiencing great listening fatigue was when the unit was right below my turntable, and about 16" from a power bar. Also, I ground the unit to my CJ PF-R grounding nut and that reduced low level RFI. I haven't had much time to listen to the unit with these new conditions, but I'll give it a go tonight and post any other comments.

My bass is still on the lean side, but I'm thinking that's a combination of it still not being totally broken in yet and, more likely, needing to roll some tubes.

Braden

mgalusha · « **Reply #10 on:** 2 Oct 2003, 10:24 pm »

Braden,

Glad to hear it's coming around. When I first had mine up and running it was on the shelf under the table and I was getting a lot of hum. It turned out that the cartridge was picking up the magnetic field from the transformer.

The Cornet is now well away from the table and it's much quieter. My Cornet is quite sensitive to tubes. While I like the sound of the JJ's on the front end, the Sovtek's provide deeper and cleaner bass. The EI Gold Elite's provide the best overall sound but they are also the noisiest.

Mike G.

pretzel_logic · « **Reply #11 on:** 3 Oct 2003, 12:12 pm »

I an using GE5814A/12AU7 and GE5751's at the moment, still looking for some different rec tubes to try. My Cornet sounds pretty good though I noticed yesterday when I touch the far right 5751 I get a major hum. Flip-flopped the tubes with no change. I'm wondering if I have a loose connection or something though I looked it over and didn't see anything. I don't hear anything while playing but it obviously has me wondering. My unit is underneath the TT shelf and I'm going to move it a little forward today.

Brian

penance · « **Reply #12 on:** 3 Oct 2003, 01:32 pm »

I have exactly the same problem mate
but not noticeable during play

Braden · « **Reply #13 on:** 3 Oct 2003, 02:40 pm »

Well, I've got some more work to do. Yesterday when I returned home from work, my Cornet had blown a resistor and smelled burnt. Great.

It was resistor R20, that's the 6K8 nearest to the transformer. Hmm, I don't think C9 was hurt, but it has some burn marks on it. Oh well, pull out the soldering iron I guess.

Braden

pretzel_logic · « **Reply #14 on:** 3 Oct 2003, 06:11 pm »

I may have some more work to do myself. I mistakenly put a 5814A in place of the 5751. I wasn't listening at the time, was doing other things but the unit had been on for some time when all of a sudden I heard a little crackle and the Cornet was powered down. Only thing I can see is the fuse burned out but am having trouble locating another fuse in town, will have to travel 15 miles to the nearest Radio Shack. I love living out in the country except during times like this. And hopefully I didn't burn up something else.

Brian

Braden · « **Reply #15 on:** 3 Oct 2003, 10:10 pm »

I replaced the R20 resistor, and, it still doesn't work. C9 must have blown as well. I've got the new parts on order, and until they come in I'm left wondering what happened. The only reason I can come up with is when I tried breaking the unit in using a cd, for the first minute or two I used a cd at a -5 to 0db sound level, which, well, probably overloaded something. It sure sounded like hell. But I did run the unit for a good thirty hours after that without any problems, but I guess that could have weakened something. Any opinions? I had myself fooled there for a minute, thinking my first DIY project would go without a hitch. HA!

Braden · « **Reply #16 on:** 8 Oct 2003, 05:03 pm »

Before I flip the switch again I just wanted to make sure I've got my wires in order. Of the two black transformer wires, one is going to the hot ac input, the other to one lead on the switch. The ac neutral is connected to the other switch terminal, then the ground to the board. This seems correct to me, but I'm also unclear about where the fuse comes into play. Could someone please explain to me how the two fuse leads are connected to the ac leads.

Thanks,
Braden

penance · « **Reply #17 on:** 8 Oct 2003, 05:16 pm »

not sure where you comeing from there mate

should go --

AC input live > Fuse>Sitch>transformer
Neutral> transformer
Earth >chasis

penance · « **Reply #18 on:** 8 Oct 2003, 05:30 pm »

here ya go mate

HTH

Braden · « **Reply #19 on:** 8 Oct 2003, 09:25 pm »

Penance,

Thanks! That makes it much more clear. I had it right, I was just by-passing the fuse.

Well, I replaced the R20 resistor, and the C9 capacitor, but I've still got nothing! The tubes don't heat up at all. I measure 120 volts comings into, then every point on the board reads 22-23 volts when it's on. All the resistors test okay, the capacitors that are connected to resistors measure okay, but all the other ones get maxed out by the capacitors that are rated too high for my meter. I guess I would have to remove them from the board to test them.

Any suggestions for what to test next?

Thanks,
Braden