Leslie Speaker Conversion, Part 1

This post and the ones that will follow address a project we have undertaken with my associate V. The intended goal for this project is a functioning Leslie speaker.

Background

Recently, V and I acquired a Leslie speaker. It doesn't quite have all it takes to function properly as a Leslie speaker, so some restoring, re-engineering and construction is necessary. What we actually have are the following parts:
- Leslie foam rotor (this is the most important component of the whole set-up)
- speaker (Utah)
- pulleys (foam rotor pulley and engine pulley)
- belt
- AC motor (fast and slow)
- mounts (for AC motor and foam rotor)
- a piece of wood, on which the above are mounted

The Plan

The plan is to do the following:
- swap the AC motor for a DC motor for safety reasons
- plan and make the necessary electronics for DC motor control
- construct a cabinet for the speaker

List of parts needed for the Fender Princeton Chorus overhaul

Thinking about purchasing the zeners from ELFA. The jacks and pots would be from UralTone, except 50K15A and 1M5C would be from Music Parts Guru or some other store. Total without postage would be about 58€.

Description	Schematic pos.	Schematic/part info	Fender pno	Store part no	number	price á
9-pin stereo 1/4” jack, Fender	J1, J2, J6-J8		0031570000 0025931000	428-H507-A	5	3.55€
4-pin mono 1/4” jack, Fender	J3-J5		0037036000 0025929000	428-H506-A	3	3.30€
pot	R10 (pos. 1 VOLUME)	50K B / B50K 8839	0027941000	636-R-VSN-50KL-DS	3	2.50€
pot	R40 (pos. 2 TREBLE)	50K 15A / 15A50K 8840	0027940000	FPHD-50KA-DS, FEN-002-7950-000	2	$4
pot	R42 (pos. 3 MID)	2K 15A / 15A2K 8846	0027937000	FPHD-2KA-DS / 636-R-VSN-2KA-DS	1	2.50€
pot	R41 (pos. 4 BASS)	50K 15A / 15A50K 8840	0027940000	FPHD-50KA-DS	see above
pot	R55 (pos. 5 REVERB)	50K B / B50K 8839	0027941000	636-R-VSN-50KL-DS	see above
pot	R20 (pos. 6 GAIN)	100K 30C / 30C100K 8830	0027947000	636-R-VSN-100KC-DS	1	2.50€
pot	R23 (pos. 7 LIMITER)	1M 5C / O5C1M 8835	0027950000	FEN-002-7950-000	1	$3
pot	R31 (pos. 8 PRES.)	50K 2B / 2B50? ????	0027942000	636-R-VSN-50KLD-DS	1	2.50€
pot	R34 (pos. 9 VOLUME)	50K B / B50K 8839	0027941000	636-R-VSN-50KL-DS	see above
pot	R131 (pos. 10 RATE)	100K B / B100K 882?	0027945000	636-R-VSN-100KL-DS	2	2.50€
pot	R139 (pos. 11 DEPTH)	100K B / B100K 8840	0027945000	636-R-VSN-100KL-DS	see above
zener diode	CR19, CR20	1N5353	?	1N5353BG	2 (+spare)	0.76€

Fender Princeton Chorus disassembly and inspection Part 1

I disassembled my Fender Princeton Chorus. This meant undoing four screws on top of the amp and then one on each side. After I had slid out the panel which holds the circuit board, I took photos for reference, detached all leads, removed the nuts around the pots and jacks and finally undid six screws holding the circuit board to the panel.

Immediately as I glanced upon the backside of the circuit board, I noticed a dark spot surrounding two components placed next to each other. The two components had the schematic numbers CR19 and CR20, which I quickly found in my paper schematic. Next to the markings CR19 and CR20 read the text '1N5353', which I assumed to be some sort of a part number. Using google, I found out the two components were 16V zener diodes (http://www.datasheetcatalog.com/datasheets_pdf/1/N/5/3/1N5353.shtml).

After a second round of googling, I had found a store selling such diodes in my home town. The price for three was 2.79€. Not too bad. Will get them on Monday.

Coming up: Fender Princeton Chorus maintenance?

The circuit board of a 1988 Fender Princeton Chorus solid-state amp

Let's see what happens...

A Smiths tachometer adventure

Left to right: Smiths RVC 1410/00AF, Smiths RVC 1414-01F

Bought a Smiths RVC 1410/00AF tachometer with the purpose of installing it to my '66 Volvo Amazon. The car I own didn't originally have one, but many people installed Smiths tachs to Amazons in the 60s and 70s, so it sort of is OK to put one in. Also, I gather the sportier models apparently had 80mm Smiths tachs installed as standard.

The tach that didn't work (RVC 1410/00AF, 4-cylinder negative earth,
no calibration pot, 80mm fitting size, possibly inteded for an MG MGB?)

The tachs I have don't read 'Volvo', as they are from British cars of the era. Why are there two? After buying the older one (RVC 1410),  I noticed that it doesn't work. One wire was loose, but the tach didn't work properly even after re-soldering it - the indication was something like 5500rpm at idle, although the needle did respond to changes in rpm. I noticed that disconnecting the +12V lead had no effect on the read-out. I suspect some component on the (very simple) circuit board has gone bad, possibly a resistor. Notably, Rick Astley posits in MGB Electrical Systems that the 1410/00AF "has been shown to be quite unreliable" and that it is unserviceable as the integrated circuit is unavailable nowadays (2006: 152).

The tach that worked (RVC 1414-01F, 4-cylinder negative earth tach,
no calibration pot, 80mm fitting size, probably intended for an MG Midget)

So I bought another one off eBay, an RVC 1414-01F that was supposed to work. And it did work. Opened both up and found them very similar. The 1414 had a better casing, because it had a green bulb cover relevant to my interests, and, of course, it had the better mechanism/circuit board. However, the 1410/00AF was prettier: it had a nice chrome bezel, a prettier needle and a prettier, more "old-skool" face.

Identical circuit boards, nearly identical construction otherwise, RVC1410 on the left and 1414 on the right

So I combined the two into a single unit, pretty and functional. From the RVC1410 I took the face, protective glass, bezel, needle and light bulb, and from the RVC1414 I took all the rest (the mechanism, internal structural components, circuit board, wiring and the case). The face is attached to the internal structure with two screws, and the bezel is even easier to separate from the case, so no problems there. Needle removal was the potentially tricky part, but I used two tea spoons and a piece of thick paper to pry it off. No problem. End result? A prettier RVC1414-01F, basically. Shame it still doesn't have a calibration pot, though...
 

Tools for needle removal

Bought a T61

Haven't even seen it yet, but it's supposed to be a Lenovo Thinkpad T61 with:

1.8GHz Core 2 Duo T7100 (Merom, I think, they call it)
3GB of RAM
80GB HDD (probably 5400rpm)
Intel Graphics X3100 (I assume)
1280x800 widescreen (I'm guessing 14.1" rather than 15.4", although both are possible)
DVD-RW/CD-RW recorder (Ultrabay Slim Super Multi-Burner Drive, I'm assuming)
two hours of battery life when gaming
a dock of some sort (has to be Ultrabase X6, I think)
mass of approx. 2.5kg not including the dock

The 14.1" widescreen would obviously be preferable in terms of portability, whereas the 15.4" widescreen version would be better in terms of potential battery life with a very significant margin, because it would seem that the 14.1" wide cannot be equipped with a nine-cell battery. BUT there might be hope, as a different page on the Lenovo site would suggest there is a 9-cell for the 14.1" wide. If this proves to be true, I really hope my T61 is a 14.1".

All in all, sounds like a promising 'puter and a great improvement over my customized X40. This one'll probably attract some botch jobs as well. The acronym SSD springs to mind, especially as I hear there's a hacked BIOS which increases SATA transfer rates on T61s. Also might consider a MB+CPU+GFX swap if a promising (and cheap) donor T61 crops up, as my future laptop in its current state (digest that) has the lowest spec components possible in those departments. Battery upgrade to a new 9-cell is inevitable, but there's not that much room for botching in replacing a battery. Or is there?

Certainly the most promising upgrades performance-wise would be the SSD (maybe a 40 or 60Gb OCZ Vertex 2) and the battery upgrade. The MB swap would be a huge project and would not really offer that much more bang for the extra buck. Would be a fun one, though. Stay tuned.

Further info: http://www.thinkwiki.org/wiki/Category:T61

Replacing the IBM X40 Hitachi 1.8" HDD with a CF card

 

Why this botch job?

Thinkwiki: http://www.thinkwiki.org/wiki/Compact_Flash_boot_drive
Russell's blog: http://vort.org/2008/02/21/converting-an-ibm-x40-to-flash/

Simply put, the Hitachi 1.8" hard-drive used in the x40 is a botch job to begin with. I guess it is OK for its age and the period when it was made, but still. Using Windows XP was incredibly painful, so I started using Ubuntu last autumn. Then I thought: why can't I have it better still? I would toss the old 1.8" clunker of a hard-drive and do as suggested at thinkwiki.org: replace it with a CompactFlash "SSD". VERY COOL (and proved to be fast, see the benchmark results below).

WHAT I SHOULD HAVE DONE IN THE FIRST PLACE:
Step 1: Get a CF card and an adapter card. Combine.
I bought a Taiwanese Silicon Power 16GB 400x (60MB/s) I found on eBay for about $75/€55 including international shipping. I found one of these adapters http://www.sintech.cn/en/cf%20ide/CF-IDE%20STCI4007.htm for $7.89 on eBay. It has a red activity led and two jumpers (master/slave and 2.5"/1.8"). Some VERY basic looking hardware here.
Step 2: Set the jumper to 1.8"

Step 3: Remove old HD and replace it with the fresh DIY-SSD
Step 4: Boot from USB, partition the CF and install an OS (do not bother with a swap)
Step 5: Start using your newly installed OS
Step 6: To improve performance, add the stuff below to your hdparm.conf (from Russell's)

/dev/sda {
write_cache = on
io32_support = 3
dma = on
lookahead = on
interrupt_unmask = on
}

Step 7: Make sure swap is not enabled and enjoy!

Benchmark results (DIY-SSD is faster):

Hitachi 4200/40GB copy 699.5MB 1:22, 1:16, 1:16
hdparm -tT /dev/sda 850MB in 2.00 = 424.94 MB/s  // 482.63 MB/s  // 477.59 MB/s
58MB in 3.07 = 18.87 MB/s  // 18.90 MB/s  // 18.89 MB/s

CF SSD defaults copy 701.0MB 1:08, 0:52, 0:54
hdparm -tT /dev/sda 982MB in 2.00 = 491.14 MB/s // 521.72 MB/s // 520.05 MB/s
513MB in 3.05 = 44.00 MB/s // 44.08 MB/s // 37.09 MB/s

CF SSD Russell's  copy 701.0MB 0:37, 0:38, 0:35
hdparm -tT /dev/sda 912MB in 2.00 = 455.95 MB/s // 514.12 MB/s // 503.06 MB/s
134MB in 3.04 = 44.05 MB/s // 44.01 MB/s // 44.37 MB/s

------------------------------------------------------
 

WHAT I ACTUALLY DID SO YOU WOULDN'T HAVE TO:

Step 1: Bought a CompactFlash memory card

Step 2: Bought a PA-CF18H CF-to-IDE adapter

Step 3: Used a USB card reader to partition the CF and built a minimal Debian install

Suffice to say the instructions at Russell's blog were somewhat dated.

Step 4: Removed the Hitachi drive
Removed the screw that holds in place the cover. Tried to pull out the cover. Couldn't get it out, so you bet I pulled harder. What came out was a combination of the plastic cover and the hard-drive that was attached to it with two further screws. I opened the screws and separated the plastic cover from the hard-drive. Easy, but requires quite a lot of force. Do not drop any the tiny screws on a thick carpet. I did.

Step 5: Prepared the CF card
Slid the CF card in its place on the adapter card. Left the jumpers as they were: MASTER/SLAVE and 2.5"/1.8". Although I didn't discover what the problem was until much later, these jumper settings proved to be a mistake, because when I...

Step 6: Tried to boot
It didn't work. To be more accurate, the X40 didn't find anything to boot from.


 
Step 7: Set the correct jumper settings and booted Ubuntu from a USB stick
I don't know if it was setting the jumper to 1.8" or booting Ubuntu from USB and re-partitioning and installing the OS, but I managed to install Ubuntu and boot. WIN!

Photos from Flickr:
Kai Hendry, turboalieno, Rickydavid, fireflythegreat